HEDIS Table of Contents

Transcription

1 HEDIS Table of Contents HEDIS Overview... 1 The HEDIS Measure Development Process... 1 Submitting Comments... 1 NCQA Review of Public Comments... 2 Value Set Directory... 2 Items for Public Comment... 2 Questions... 3 Proposed New Measures... Utilization of the PHQ-9 to Monitor Depression Symptoms for Adolescents and Adults... 6 Depression Remission, Response or Treatment Adjustment for Adolescents and Adults Depression Screening and Follow-up for Adolescents and Adults Inpatient Hospital Utilization Emergency Department Utilization Statin Therapy for Patients with Cardiovascular Disease Statin Therapy for Patients with Diabetes Hospitalization for Potentially Preventable Complications Proposed Changes to Existing Measures... Asthma Medication Ratio Medication Management for People with Asthma Medication Reconciliation Post-Discharge Proposed Measures to Retire... Use of Appropriate Medications for People with Asthma Guideline Updates... General Guideline 28: Members who Switch Products Guidelines for Relative Resource Use Measures Guideline Announcement... General Guidelines 12 16: Retirement of the Measure Rotation Strategy HEDIS is a registered trademark of the National Committee for Quality Assurance (NCQA).

2 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, HEDIS Public Comment Overview HEDIS Overview HEDIS is a set of standardized performance measures designed to ensure that purchasers and consumers can reliably compare the performance of health plans. It also serves as a model for emerging systems of performance measurement in other areas of health care delivery. HEDIS is maintained by NCQA, a not-for-profit organization committed to evaluating and publicly reporting on the quality of physicians, HMOs, PPOs, ACOs and other organizations. The HEDIS measurement set consists of 83 measures across five domains of care. Items available for Public Comment are proposed changes to HEDIS 2016, which will be published in July 2015 and reported in June 2016, based on activity that occurred during the 2015 measurement year. HEDIS Measure Development Process The NCQA Committee on Performance Measurement (CPM) oversees the evolution of the measurement set. Several Measurement Advisory Panels (MAP) provide clinical and technical knowledge required to develop the measures. Additional HEDIS Expert Panels and the Technical Measurement Advisory Panel (TMAP) provide invaluable assistance by identifying methodological issues and providing feedback on new and existing measures. Synopsis NCQA seeks feedback on proposed new measures and changes to specifications and guidelines for HEDIS Reviewers are asked to submit their comments to NCQA in writing via the Public Comment Web site by 5:00 PM (EST) Wednesday, March 18. Submitting Comments Submit all comments via NCQA s Public Comment Web site, using the following link: Note: NCQA does not accept comments via mail, or fax. How to Submit a Comment 1. Enter the following information: Your address. Your contact information. 2. Choose from the following options: Select HEDIS 2016 Public Comment. Select the measure on which you would like to comment. Select your support option (e.g., Support, Do not support, Support with modifications). Note: If you choose Do not support, include your rationale in the text box. If you choose Support with modifications, enter the suggested modification in the text box.

3 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Enter your comment into the text box. There is a 2,500 character limit. If you exceed the limit, your comment will be cut off at 2,500 characters. Note: We suggest that you develop your comment in Word, in order to check your character limit, and save a copy for reference. Use the cut and paste function to copy your comment into the text box. 4. Submit additional comments using the same process. NCQA must receive your comments by 5:00 PM (EST) Wednesday, March 18. Note: For more detailed information on posting a comment, please see Posting a Comment Instructions document. NCQA Review of Public Comments NCQA appreciates the time and effort required to submit comments, and reviews all feedback submitted within the Public Comment period. Due to the high volume of comments received, NCQA cannot respond to individual comments; however, NCQA MAPs and Expert Panels will consider all comments and advise NCQA staff. Based on the review, NCQA staff will bring recommendations to the CPM, a committee of representatives from purchasers, consumers, health plans, health care providers and policy makers that oversees the evolution of the measurement set. The CPM also reviews all comments before making final decisions for HEDIS Value Set Directory Effective with HEDIS 2014, code tables are not included in measure specifications; they are included in a separate Excel workbook called the Value Set Directory. Measure specifications reference value sets that must be used for HEDIS reporting. A value set contains the complete set of codes used to identify the service or condition included in the measure. All codes for all measures listed below are in the HEDIS 2016 Public Comment Value Set Directory, which is included with the measure materials on the NCQA Public Comment Web page. Items for Public Comment Refer to the NCQA Public Comment page for detailed documentation (e.g., memos, specifications, workups, performance data, guidelines, value sets) on the items described below. Proposed New Measures NCQA proposes the following new measures: 1. Utilization of the PHQ-9 to Monitor Depression Symptoms for Adolescents and Adults. 2. Depression Remission, Response or Treatment Adjustment for Adolescents and Adults. 3. Depression Screening and Follow-up for Adolescents and Adults. 4. Inpatient Hospital Utilization. 5. Emergency Department Utilization. 6. Statin Therapy for Patients with Cardiovascular Disease. 7. Statin Therapy for Patients with Diabetes. 8. Hospitalization for Potentially Preventable Complications.

4 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Proposed Changes to Existing Measures NCQA proposes changes to the following measures: 1. Asthma Medication Ratio. 2. Medication Management for People with Asthma. 3. Medication Reconciliation Post-Discharge. Proposed Measures to Retire NCQA proposes retiring the following measures: 1. Use of Appropriate Medications for People with Asthma. Guideline Updates NCQA would like feedback on the following guideline updates: 1. General Guideline 28: Members who Switch Products. 2. Guidelines for Relative Resource Use Measures Guideline Announcement NCQA would like to announce the following guideline retirement: 1. General Guidelines 12 16: Retirement of the Measure Rotation Strategy. Questions? Contact NCQA Customer Support at , Monday Friday, 8:30 a.m. 5:00p.m. (EST). 1 HEDIS is a registered trademark of the National Committee for Quality Assurance (NCQA).

5 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Proposed New Measures for HEDIS 1 Learning Collaborative: Depression Care Measures Set NCQA s theme for 2015 our 25th anniversary is Looking Forward. As part of planning for NCQA s future, we propose new HEDIS measures that address patient-reported outcomes and utilize electronic clinical data. The measures are proposed for phased implementation through a learning collaborative, with voluntary reporting using a new data collection method. We have been encouraged by stakeholders to explore measures that assess depression care; specifically, do health plan members get screened for depression appropriately and do they receive care that leads to improved outcomes? Throughout the development process, NCQA focused on specifications assessing the quality of patient management and ones that encourage collaborative care by utilizing technology in an efficient and effective manner. To achieve these goals, detailed electronic clinical data, such as electronic health records (EHRs) and clinical registries, were required in order to provide valid and reliable health plan HEDIS measures. NCQA seeks comments on the proposed measure set, Depression Care for Adolescents and Adults, which will require a new data collection method using electronic clinical data sources: 1. Utilization of the PHQ-9 to Monitor Depression Symptoms for Adolescents and Adults. The percentage of members 12 years of age and older with a diagnosis of major depression or dysthymia who have a PHQ-9 or PHQ-A tool administered at least once during a four-month period. 2. Depression Remission, Response or Treatment Adjustment for Adolescents and Adults. The percentage of members 12 years of age and older with a diagnosis of depression and an elevated PHQ-9 or PHQ-A score, who had evidence of response or remission within 5 7 months of the elevated PHQ-9 score or an indication of treatment adjustment within 30 days of the PHQ-9 score that showed no evidence of response or remission. 3. Depression Screening and Follow-up for Adolescents and Adults. The percentage of members 12 years of age and older who were screened for clinical depression using a standardized tool and, if screened positive, who received appropriate follow-up care. The new measures are adapted from existing NQF-endorsed provider-level measures, developed by Minnesota Community Measurement and the Centers for Medicare & Medicaid, and used in various federal quality reporting programs. They assess care for depression along the continuum of care and will be included in a new HEDIS domain. In recognition of the unique challenges and opportunities new data sources present, we propose to phase-in voluntary reporting through a learning collaborative, beginning with Utilization of the PHQ-9 for HEDIS NCQA will work with health plans to refine data collection and reporting guidelines. We expect to identify several innovative solutions and better understand the role of health plans in aggregating electronic data for quality improvement. Although we expect limited ability to report these measures initially, this will be a large step forward in adapting HEDIS to efficient use of health information technology. The new data collection method introduces possibilities for future HEDIS measures that let plans track meaningful outcomes and earn recognition for investing in innovative quality improvement strategies. NCQA field-tested the measures using different data sources (e.g., EHRs, a statewide database for quality reporting, administrative claims, medical records). Performance results were generated for Medicare, Medicaid and commercial plans using 2012 and 2013 data. They demonstrate low performance, suggesting that gaps in care exist for all three measures, independent of data source. Testing also demonstrated the feasibility of calculating and reporting the measures at the health plan level, using EHR and other electronic 1 HEDIS is a registered trademark of the National Committee for Quality Assurance (NCQA).

6 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, clinical data. Supporting documents for the proposed measures include the draft measure specifications and associated measure rationale work-up. NCQA would like to hear from health plans who are interested in participating in a learning collaborative for further testing of these measures in Please send inquiries to [email protected]. NCQA acknowledges the contributions of the Geriatric Measurement Advisory Panel, the Behavioral Health Measurement Advisory Panel and the Technical Measurement Advisory Panel. This project was supported by grant number U18HS (PI: Scholle) from the Agency for Healthcare Research and Quality (AHRQ) and contract number HHSM C from the Centers for Medicare and Medicaid Services (CMS). The content is solely the responsibility of the authors and does not necessarily represent the official views of AHRQ or CMS.

7 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Utilization of the PHQ-9 to Monitor Depression Symptoms for Adolescents and Adults SUMMARY OF CHANGES TO HEDIS 2016 This is a first year measure for HEDIS 2016 using a new reporting methodology for electronic clinical data. Description The percentage of members 12 years of age and older with a diagnosis of major depression or dysthymia who have a PHQ-9 or PHQ-A tool administered at least once during a four-month period. Two rates are reported. 1. ECDS Data Coverage Rate. The percentage of members 12 and older with a diagnosis of major depression or dysthymia who are covered by an electronic clinical data system (ECDS). 2. PHQ Utilization Rate. The percentage of PHQ utilization. Members with a diagnosis of major depression or dysthymia who are covered by an ECDS and, if they had an outpatient encounter, have either a PHQ-9 or PHQ-A Score present in their record. Definitions ECDS Electronic clinical data system. An electronic version of a patient s comprehensive medical experiences, maintained over time. May include some or all of the key administrative clinical data relevant to the patient s care (e.g., demographics, progress notes, problems, medications, vital signs, past medical history, social history, immunizations, laboratory data, radiology reports). The ECDS provides automated access to the patient s comprehensive information and has the ability to create data files to be used for quality reporting (e.g., QRDA 1, C-CDA, CCD) and may also support other care-related activities through various interfaces, including evidence-based decision support, quality management and outcomes reporting. To qualify for this measure, ECDS data must be automated and accessible to the care team at the point of care (e.g., EHRs, registries and case management or disease management systems that the provider has access to during a patient interaction). Measurement period The measurement year is segmented to establish regular utilization of the PHQ assessment tool in the management of depression. The first qualifying encounter in each period determines the denominator events for the performance measure. The measurement year is divided into three periods: Assessment Period One. Dates of service January 1 through April 30 of the measurement year. Assessment Period Two. Dates of service May 1 through August 31 of the measurement year. Assessment Period Three. Dates of service September 1 through December 31 of the measurement year.

8 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Eligible Population Product line Ages Commercial, Medicare, Medicaid (report each product line separately). 12 years and older as of January 1 of the measurement year. Report four age stratifications and a total rate years. 65+ years years. Total years. Continuous enrollment Allowable gap Anchor date Benefit Event/ diagnosis Step 1 Step 2 Step 3 The measurement year. No more than one gap in enrollment up to 45 days during the measurement year. To determine continuous enrollment for a Medicaid beneficiary for whom enrollment is verified monthly, the member may not have more than a 1-month gap in coverage (i.e. a member whose coverage lapses for 2 months [60 days] is not considered continuously enrolled.) December 31 of the measurement year. Medical Follow the steps below to identify the eligible population. Identify all members with an active diagnosis of major depression or dysthymia (Major Depression and Dysthymia Value Set) that starts before the beginning of the measurement year or during the measurement year. Identify all members in step 1 with an active diagnosis of depression that starts before or during an outpatient encounter (Depression Encounter Value Set) that occurs in the measurement year. Determine continuous enrollment. For all members identified in step 2, identify members continuously enrolled in the health plan for the measurement year, with no more than a 45-day gap in enrolment. Electronic Specification ECDS data coverage denominator ECDS data coverage numerator The eligible population. Identify all members for whom the plan has access to the electronic clinical data. This includes any provider or provider groups that can submit clinical data to the health plan via electronic file transfer (e.g., QRDA 1, C-CDA, CCD) or through third-party aggregator applications for purposes of quality reporting. Include members only if the electronic clinical interface is accessible by health care providers at the point of care.

9 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Utilization of PHQ-9 denominator Step 1 Step 2 Step 3 Step 4 Utilization of PHQ-9 numerator Step 1 Step 2 Step 3 Step 4 Step 5 Total performance Rate Count the first outpatient encounter in any or all of the three assessment periods. Members need only have one event in any measurement period to be counted. Follow the steps below to determine denominator events. For all members from the ECDS data coverage numerator identify all outpatient encounters (Depression Encounter Value Set) during the measurement year where an active diagnosis of depression starts before or during the encounter. For each outpatient encounter in step 1, identify the date of service and classify each encounter in one of the three assessment periods. For each assessment period, count only the first qualifying encounter for each member. Each member may have up to three qualifying encounters (one from each assessment period) for the measurement year. Count the number of qualifying encounters for each member. The denominator is the sum of the member s first qualifying encounter for each assessment period. A PHQ-9 or PHQ-A total score in the patient s record during the assessment period in which a qualifying encounter occurred. Follow the steps below to determine numerator events. For each member, identify PHQ questionnaires completed during the measurement year. The presence of a PHQ total score indicates completion of a PHQ assessment tool. Assign a date to each PHQ score. Use the date when the PHQ total score was recorded, not the date when the assessment was performed. Classify each PHQ score in an assessment period. For each assessment period, count only the first qualifying PHQ total score for each member. Each member may have up to three qualifying PHQ total scores (one from each assessment period) for the measurement year. Sum the qualifying PHQ events across the three periods to report the total PHQ utilization numerator. To calculate the performance rate, divide the sum of the qualifying PHQ events across the three assessment periods (step 5) by the sum of the qualifying encounters across the three assessment periods. Exclusions (Required) Any member with an active diagnosis of any of the following at any time during the measurement year: Bipolar disorder (Bipolar Disorder Value Set; Bipolar Disorder ECDS Value Set). Personality disorder (Personality Disorder Value Set; Personality Disorder ECDS Value Set). Psychotic disorder (Psychotic Disorders Value Set). Autism spectrum disorder (Pervasive Developmental Disorder Value Set). Also exclude: Members admitted to hospice during the measurement year or with permanent residence in a nursing home. Note: Value Sets will contain all applicable SNOMED codes.

10 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Data Elements for Reporting Organizations that submit HEDIS data to NCQA must provide the following data elements. Table XXX-1/2/3: Data Elements for Utilization of the PHQ-9 to Monitor Depression Symptoms Measurement year Data collection methodology (Electronic Clinical Data) Eligible population ECDS denominator ECDS numerator Required exclusions Numerator total rate Lower 95% confidence interval Upper 95% confidence interval Electronic Clinical Data For each age stratification and total For each age stratification and total Each rate, for each age stratification and total Each rate, for each age stratification and total Each rate, for each age stratification and total Each rate, for each age stratification and total Each rate, for each age stratification and total

11 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Measure Flow Diagrams Figure 1: Electronic Data Capture Rate for Depression Care Eligible Population Is the member aged 12 or older? YES Did the member NO Not included in eligible population Have an active diagnosis of MDD or dysthymia starting before the start of the measurement year or during the measurement year? YES Have an outpatient encounter preceded by an active diagnosis of depression during the measurement year? YES YES Have continuous enrollment in the health plan for the measurement year, with medical benefits? NO Not included in eligible population Exclusions Did the member have an active diagnosis of bipolar disorder, personality disorder, schizophrenia, psychotic disorder or pervasive developmental disorder anytime during the measurement year? OR Was the member in hospice or a permanent resident in a nursing home during the measurement year? YES Not included in denominator NO Does the health plan have direct access to the electronic clinical data as required by the measure specification? This includes any submission of clinical data directly to the health plan via electronic file transfer (e.g. QRDA 1, C-CDA, or CCD) for purposes of quality reporting. NO Not numerator compliant Numerator YES Are these data also directly accessible at the point of care by the reporting provider? NO Not numerator compliant YES Numerator compliant

12 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Figure 2: Measure Performance Eligible Population Is the member aged 12 or older? YES Did the member NO Not included in eligible population Have an active diagnosis of MDD or dysthymia starting before the start of the measurement year or during the measurement year? YES Have an outpatient encounter preceded by an active diagnosis of depression during the measurement year? YES YES Have continuous enrollment in the health plan for the measurement year, with medical benefits? NO Not included in eligible population Exclusions Did the member have an active diagnosis of bipolar disorder, personality disorder, schizophrenia, psychotic disorder or pervasive developmental disorder anytime during the measurement year? OR Was the member in hospice or have permanent residence in a nursing facility during the measurement year? YES Not included in denominator NO Did the member have a qualifying encounter in Assessment Period 1? (January 1 through April 30) YES Count in denominator Denominator Did the member have a qualifying encounter in Assessment Period 2? (May 1 through August 31) YES Count in denominator Did the member have a qualifying encounter in Assessment Period 3? (September 1 through December 31) YES Count in denominator Numerator Does the member have a PHQ total score in Assessment Period 3? (September 1 through December 31) Does the member have a PHQ total score in Assessment Period 2? (May 1 through August 31) Does the member have a PHQ total score in Assessment Period 1? (January 1 through April 30) YES YES YES Numerator compliant

13 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Depression Remission, Response or Treatment Adjustment for Adolescents and Adults SUMMARY OF CHANGES TO HEDIS This is a new measure for HEDIS, using a new reporting methodology for electronic clinical data. Description The percentage of members 12 years of age and older with a diagnosis of depression and an elevated PHQ-9 or PHQ-A score, who had evidence of response or remission within 5 7 months of the elevated PHQ-9 score or an indication of treatment adjustment within 30 days of the PHQ-9 score, that showed no evidence of response or remission. Five rates are reported: 1. ECDS Data Coverage Rate. The percentage of members 12 years of age and older whose health information is accessible in an electronic clinical data system (ECDS). 2. Depression Remission Rate. The percentage of members who achieved remission (PHQ score <5) within five seven months after the initial elevated PHQ-9 score. 3. Depression Response Rate. The percentage of members who were not in remission and showed response within five seven months after the initial elevated PHQ-9 score. 4. Depression Treatment Adjustment Rate. The percentage of members who have an indication of treatment adjustment within 30 days of the PHQ-9 score that showed no response or remission. 5. Remission, Response and Treatment Adjustment Total Rate. The sum of the Remission, Response and Treatment Adjustment performance rates. Definitions ECDS Electronic clinical data system. An electronic version of a patient s comprehensive medical experiences, maintained over time. May include some or all of the key administrative clinical data relevant to the patient s care (e.g., demographics, progress notes, problems, medications, vital signs, past medical history, social history, immunizations, laboratory data, radiology reports). The ECDS provides automated access to the patient s comprehensive information and has the ability to create data files to be used for quality reporting (e.g., QRDA 1, C-CDA, CCD) and may also support other care-related activities through various interfaces, including evidence-based decision support, quality management and outcomes reporting. To qualify for this measure, ECDS data must be automated and accessible to the care team at the point of care (e.g., EHRs, registries and case management or disease management systems that the provider has access to during a patient interaction). Intake period IESD May 1 of the year prior to the measurement year through April 30 of the measurement year. Index Episode Start Date. The earliest date during the intake period where either a PHQ-9 or a PHQ-A score >9 is documented in the ECDS within 30 days of an interactive encounter between the member and a provider. An interactive encounter can be face-to-face, phone-based or via secure messaging (e.g., or patient portal).

14 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Depression reevaluation period Treatment remission Treatment response Treatment adjustment Depression response score The day (inclusive) period following the IESD, during which depression symptoms are reevaluated using the PHQ-9 or PHQ-A tool. Members who achieve remission from depression, as demonstrated by a PHQ-9 or PHQ-A score of <5 recorded in the ECDS during the Depression Reevaluation Period. Members with indication of response to treatment for depression, as demonstrated by a PHQ-9 total score or PHQ-A total score reduction of at least 50 percent during the Depression Reevaluation Period. Members with no indication of remission or response on a follow-up PHQ-9 or PHQ-A score during the Depression Reevaluation Period, but who have evidence that treatment was adjusted (i.e., dispensed a new antidepressant or non-antidepressant psychotropic medication, had increased dosage of an antidepressant, switched antidepressants, were delivered a new therapy or changed therapy type) within 30 days of the follow-up PHQ-9 or PHQ-A score that showed no remission or response. PHQ-9 or PHQ-A score indicating a member s depression symptom level. The PHQ score must be the last noted in the member s record during the Depression Reevaluation Period (i.e., if multiple PHQ scores are present in the member record, choose the score closest to the end of the Depression Reevaluation Period). Eligible Population Product lines Ages Commercial, Medicaid, Medicare (report each product line separately). 12 years and older as of May 1 of the year prior to the measurement year. Report four age stratifications and a total rate years years years. 65+ years. Total. The total is the sum of the age stratifications. Continuous enrollment Allowable gap Anchor date Benefit May 1 of the year prior to the measurement year through December 31 of the measurement year. No more than one gap in continuous enrollment of up to 45 days during each year of continuous enrollment. To determine continuous enrollment for a Medicaid beneficiary for whom enrollment is verified monthly, the member may not have more than a 1-month gap in coverage (i.e., a member whose coverage lapses for 2 months [60 days] is not considered continuously enrolled). IESD. Medical.

15 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Event/diagnosis Step 1 Step 2 Step 3 Follow the steps below to identify the eligible population. Identify members in the specified age range with at least one interactive outpatient encounter (Depression Encounter Value Set) during the Intake Period, with an active diagnosis of major depressive disorder or dysthymia (Major Depression and Dysthymia Value Set). Identify all eligible episode dates. For each member in step 1, identify the dates for each interactive outpatient encounter (Depression Encounter Value Set) occurring during the Intake Period. Identify the IESD. For each member in step 2, identify the first date during the Intake Period where an elevated PHQ-9 or PHQ-A score (>9) was recorded in the ECDS 15 days prior to the eligible episode through 15 days after the eligible episode. Electronic Specification ECDS data coverage denominator ECDS data coverage numerator Denominator: Depression rate Numerator: Depression remission Numerator: Depression response The eligible population. Identify all members for whom the plan has access to the electronic clinical data. This includes any provider or provider groups that can submit clinical data to the health plan via electronic file transfer (e.g., QRDA 1, C-CDA, CCD) or through thirdparty aggregator applications for quality reporting. Include members only if the electronic clinical interface is accessible by health care providers at the point of care. The eligible population: all members who meet the eligibility criteria for the ECDS data coverage numerator. Identify all members who achieve remission of depression symptoms as demonstrated by a PHQ-9 or PHQ-A Depression Response Score of <5 recorded in the ECDS during the Depression Reevaluation Period. From the group of members who do not meet depression remission numerator criteria, identify members who indicate a response to treatment for depression as demonstrated by a PHQ-9 or PHQ-A Depression Response Score at least 50 percent lower than the PHQ score associated with the IESD, recorded in the ECDS during the Depression Reevaluation Period. The condition of PHQ-9 or PHQ-A score response is met when the following is true: Numerator: Depression Treatment adjustment PPPPPP DDDDDDDDDDDDDDDDDDDD RRRRRRRRRRRRRRRR SSSSSSSSSS PPPPPP tttttttttt ssssssssss ffffffff IIIIIIII 22 From the group of members who do not meet the Depression Response or Remission Numerator criteria, identify members who have evidence of treatment adjustment within 30 days of the Depression Response Score indicating no remission or response during the Depression Reevaluation Period. Documentation of any of the following indicate treatment adjustment: New antidepressant. Dispensed an antidepressant medication (Table AMM-C), where the member had no pharmacy claims for the same medication during the period between the IESD and the Depression Response Score. This includes starting an antidepressant medication, adding a new antidepressant medication to existing medications or switching antidepressants.

16 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Note: Switching between a generic and a brand name medication does not count as treatment adjustment. Adjusting the dose of the same medication counts as treatment adjustment. New Non-Antidepressant. Dispensed a non-antidepressant psychotropic medication (Table XXX-X), where the member had no pharmacy claims for the same medication during the period between the IESD and the Depression Response Score. New Therapy. Any service for individual therapy (Individual Therapy Value Set), family therapy (Family Therapy Value Set), group therapy (Group Therapy Value Set) or other therapy (Other Therapy Value Set) that was not used (i.e., no services documented) between the IESD and the Depression Response Score. This includes starting any type of therapy, adding a new type of therapy to an existing therapy (e.g., adding family therapy to individual therapy) or switching between types of therapies (e.g., from family therapy to individual therapy). Note: Switching between different lengths of therapy does not count as treatment adjustment. Total performance rate To calculate the performance rate, sum the numerators for remission, response and treatment adjustment and divide by the Depression Rate denominator. Table AMM-C: Antidepressant Medications Description Prescription Miscellaneous Bupropion Vilazodone antidepressants Monoamine oxidase inhibitors Phenylpiperazine antidepressants Psychotherapeutic combinations SSNRI antidepressants SSRI antidepressants Tetracyclic antidepressants Tricyclic antidepressants Isocarboxazid Phenelzine Nefazodone Selegiline Tranylcypromine Trazodone Amitriptyline-chlordiazepoxide Amitriptyline-perphenazine Desvenlafaxine Venlafaxine Duloxetine Citalopram Escitalopram Maprotiline Amitriptyline Amoxapine Clomipramine Fluoxetine Fluvoxamine Mirtazapine Desipramine Doxepin (>6 mg) Imipramine Fluoxetineolanzapine Paroxetine Sertraline Nortriptyline Protriptyline Trimipramine

17 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Table XXX-X: Non-Antidepressant Psychotropic Medications Antipsychotics Description Antidepressant and antipsychotic combination Anticonvulsants Anti-anxiety Benzodiazepines Other hypnotics Stimulants Prescription Note: NCQA will post a comprehensive list of medications and NDC codes to by November 1, Exclusions (Required) Exclude members with any of the following at any time during the measurement period: Bipolar disorder (Bipolar Value Set; Bipolar Disorder ECDS Value Set). Personality disorder (Personality Disorder Value Set; Personality Disorder ECDS Value Set). Psychotic disorder (Psychotic Disorders Value Set). Autism spectrum disorder (Pervasive Developmental Disorder Value Set). Members admitted to hospice during the measurement year or with permanent residence in a nursing home. Note: Value Sets will contain all applicable SNOMED codes. Data Elements for Reporting Organizations that submit HEDIS data to NCQA must provide the following data elements. Table DRRT-1/2/3: Data Elements for Remission, Response and Treatment Adjustment Measurement year Data collection methodology (electronic clinical data) Eligible population ECDS Denominator ECDS Numerator Required exclusions Remission Numerator Response Numerator Treatment Adjustment Numerator Reported rate Lower 95% confidence interval Upper 95% confidence interval Electronic Clinical Data For each age stratification and total For each age stratification and total Each rate, for each age stratification and total Each rate, for each age stratification and total Each rate, for each age stratification and total Each rate, for each age stratification and total Each rate, for each age stratification and total Each rate, for each age stratification and total Each rate, for each age stratification and total Each rate, for each age stratification and total

18 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Measure Flow Diagrams Figure 1: Electronic Data Capture Rate for Depression Care Eligible Population Was the member aged 12 or older at the beginning of the intake period? YES Did the member NO Not included in eligible population Have at least one encounter during the intake period where there was an active diagnosis of depression? YES Have a PHQ-9 score >9 within 30 days of a depression encounter? YES Have continuous enrollment in the health plan for the specified time frame? NO Not included in eligible population YES Exclusions Did the member have a diagnosis of bipolar disorder, personality disorder, psychotic disorder or pervasive developmental disorder during the measurement year? OR Was the member in hospice or have permanent residence in a nursing facility during the measurement year? YES Not included in denominator NO Does the health plan have direct access to the electronic clinical data as required by the measure specification? This includes any submission of clinical data directly to the health plan via electronic file transfer (e.g. QRDA 1, C-CDA, or CCD) for purposes of quality reporting. NO Not numerator compliant Numerator YES Are these data also directly accessible at the point of care by the provider? NO Not numerator compliant YES Numerator compliant

19 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Figure 2: Measure Performance Eligible Population Was the member aged 12 or older at the beginning of the intake period? YES Did the member NO Not included in eligible population Have at least one encounter during the intake period where there was an active diagnosis of depression? YES Have a PHQ-9 score > 9 within 30 days of a depression encounter? YES Have continuous enrollment in the health plan for the specified time frame? NO Not included in eligible population YES Exclusions Did the member have a diagnosis of bipolar disorder, personality disorder, psychotic disorder or pervasive developmental disorder during the measurement year? OR Was the member in hospice or have permanent residence in a nursing facility during the measurement year? YES Not included in denominator NO Did the member have a Depression Response Score of <5 during the Depression Reevaluation Period? YES Numerator compliant NO Numerator Did the member have a Depression Response Score reduced by at least 50% during the Depression Reevaluation Period? NO YES Numerator compliant Did the member have Treatment Adjustment within 30 days of the Depression Response Score that showed no response during the Depression Reevaluation Period? YES Numerator compliant

20 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Depression Screening and Follow-Up for Adolescents and Adults SUMMARY OF CHANGES TO HEDIS This is a new measure for HEDIS, using a new reporting methodology for electronic clinical data. Description The percentage of members 12 years of age and older who were screened for clinical depression using a standardized tool and, if screened positive, who received appropriate follow-up care. Two rates are reported. 1. ECDS Data Coverage Rate. The percentage of members 12 years of age who are covered by an electronic clinical data system (ECDS). 2. Depression Screening Rate. The percentage of members who were screened for clinical depression using a standardized tool and, if screened positive, received appropriate follow-up care. Definitions ECDS Electronic clinical data system. An electronic version of a patient s comprehensive medical experiences, maintained over time. May include some or all of the key administrative clinical data relevant to the patient s care (e.g., demographics, progress notes, problems, medications, vital signs, past medical history, social history, immunizations, laboratory data, radiology reports). The ECDS provides automated access to the patient s comprehensive information and has the ability to create data files to be used for quality reporting (e.g., QRDA 1, C-CDA, CCD) and may also support other care-related activities through various interfaces, including evidence-based decision support, quality management and outcomes reporting. To qualify for this measure, ECDS data must be automated and accessible to the care team at the point of care (e.g., EHRs, registries and case management or disease management systems that the provider has access to during a patient interaction). Adolescent screening tool (12 17 years) Adult screening tool (18 years and older) Intake period An assessment tool that has been normalized and validated for the adolescent patient population (e.g., Patient Health Questionnaire for Adolescents [PHQ-A], Beck Depression Inventory-Primary Care Version [BDI-PC], Mood Feeling Questionnaire [MFQ], Center for Epidemiologic Studies Depression Scale [CES-D], PRIME MD- PHQ2). An assessment tool that has been normalized and validated for the adult patient population (e.g., Patient Health Questionnaire [PHQ-9], Beck Depression Inventory [BDI or BDI-II], Center for Epidemiologic Studies Depression Scale [CES-D], Depression Scale [DEPS], Duke Anxiety-Depression Scale [DADS], Geriatric Depression Scale [GDS], Cornell Scale Screening and PRIME MD-PHQ-2, Edinburgh Postnatal Depression Scale [EPDS]). January 1 through December 1 of the measurement year.

21 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Eligible Population Product lines Ages Commercial, Medicaid, Medicare (report each product line separately). 12 years of age and older as of January 1 of the measurement year. Report four age stratifications and a total rate years years years. 65+ years. Total. The total is the sum of the age stratifications. Continuous enrollment Allowable gap Anchor date Benefit Event/diagnosis The measurement year and the year prior to the measurement year. No more than one gap in continuous enrollment of up to 45 days during each year of continuous enrollment. To determine continuous enrollment for a Medicaid beneficiary for whom enrollment is verified monthly, the member may not have more than a 1-month gap in coverage (i.e., a member whose coverage lapses for 2 months [60 days] is not considered continuously enrolled). December 31 of the measurement year. Medical Members who had an outpatient visit (Depression Encounter Value Set) during the measurement year. Electronic Specification ECDS data coverage denominator ECDS data coverage numerator Denominator: Depression screening Numerator: Depression screening The eligible population. Identify all members for whom the plan has access to the electronic clinical data. This includes any provider or provider groups that can submit clinical data to the health plan via electronic file transfer (e.g., QRDA 1, C-CDA, CCD) or through thirdparty aggregator applications for purposes of quality reporting. Include members only if the electronic clinical interface is accessible by health care providers at the point of care. The eligible population: all members who meet the eligibility criteria for the ECDS data coverage numerator. Members who were screened for clinical depression using an age-appropriate standardized tool and, if the screen is positive, were provided follow-up care within 30 days of the positive result. Follow the steps below to determine numerator compliance:

22 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Step 1 Step 2 Identify all members with a result for an age-appropriate screening tool (Depression Screen Value Set) documented during the intake period. For all members with a documented screening result, identify members whose result is negative (Negative Depression Screen Value Set) using the criteria specified by the screening tool. For example, a member whose PHQ-9 score is <5 is considered to have screened negative for depression. Step 3 For all members with a documented screening result, identify members whose result is positive (Positive Depression Screen Value Set) using the criteria specified by the screening tool. For example, a member whose PHQ-9 score is 5 is considered to have screened positive for depression. Step 4 For all members from step 3, count members for whom follow-up care was provided within 30 days (inclusive) of the date of the positive screen. Follow-up must include one or more of the following in the 30-day window following the initial positive screen: Dispensed an antidepressant medication (Table AMM-C) A follow-up encounter in behavioral health (Behavioral Health Encounter Value Set), including assessment, therapy, medication management, acute care. Note: Behavioral health encounters on the same day as the positive screen count as followup care. A follow-up outpatient visit (ECDS Follow-Up Visit Value Set) with a diagnosis of depression (Depression Value Set). Note: Outpatient encounters outside behavioral health on the same day as the positive screen do not count as follow-up care. For example, a visit with a primary care provider with a diagnosis of depression or dysthymia on the same day as the positive screen does not meet the criteria for follow-up care. Follow-up with a case manager (Case Management Encounter Value Set), with documented assessment of depression symptoms (any encounter that documents the provider addressing depression symptoms). Note: Case management encounters on the same day as the positive screen do not count as follow-up care. Assessment on the same day as the positive screen, which includes documentation of additional depression assessment indicating no depression. For example, if the initial positive screen resulted from a PHQ2 score, documentation of a negative PHQ-9 counts as evidence of follow-up. Step 5 Sum the total number of members identified in step 2 with a negative screening result and members who received appropriate follow-up from step 4 to report the total numerator events.

23 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Table AMM-C: Antidepressant Medications Description Miscellaneous antidepressants Monoamine oxidase inhibitors Phenylpiperazine antidepressants Psychotherapeutic combinations SSNRI antidepressants SSRI antidepressants Tetracyclic antidepressants Tricyclic antidepressants Bupropion Isocarboxazid Phenelzine Nefazodone Prescription Vilazodone Selegiline Tranylcypromine Trazodone Amitriptyline-chlordiazepoxide Amitriptyline-perphenazine Desvenlafaxine Duloxetine Citalopram Escitalopram Maprotiline Amitriptyline Amoxapine Clomipramine Venlafaxine Fluoxetine Fluvoxamine Mirtazapine Desipramine Doxepin (>6 mg) Imipramine Fluoxetineolanzapine Paroxetine Sertraline Nortriptyline Protriptyline Trimipramine Note: NCQA will post a comprehensive list of medications and NDC codes to by November 1, Exclusions (Required) Members with either of the following: An active diagnosis of bipolar disorder (Bipolar Disorder Value Set; Bipolar Disorder ECDS Value Set) during the measurement year or the year prior to the measurement year. An active diagnosis of depression (Depression Value Set) in the year prior to the measurement year. Note: Value Sets will contain all applicable SNOMED codes. Data Elements for Reporting Organizations that submit HEDIS data to NCQA must provide the following data elements. Table ABA-1/2/3: Data Elements for Depression Screening and Follow-up Measurement year Data collection methodology (Electronic Clinical Data) Eligible population ECDS denominator ECDS numerator Required exclusions Depression Screening Numerator Reported rate Lower 95% confidence interval Upper 95% confidence interval Electronic Clinical Data For each age stratification and total For each age stratification and total Each rate, for each age stratification and total Each rate, for each age stratification and total Each rate, for each age stratification and total Each rate, for each age stratification and total Each rate, for each age stratification and total Each rate, for each age stratification and total

24 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Measure Flow Diagrams Figure 1: Electronic Data Capture Rate for Depression Care Eligible Population Is the member aged 12 or older as of January 1 st of the measurement year? YES Did the member NO Not included in eligible population Have at least one outpatient encounter during the intake period? YES Have continuous enrollment in the health plan for the measurement year and the year prior? NO Not included in eligible population YES Exclusions Did the member have a diagnosis of bipolar disorder during the measurement year or the year prior? OR Did the member have a diagnosis of depression during the year prior to the measurement year? YES Not included in denominator NO Does the health plan have direct access to the electronic clinical data as required by the measure specification? This includes any submission of clinical data directly to the health plan via electronic file transfer (e.g. QRDA 1, C-CDA, or CCD) for purposes of quality reporting. NO Not numerator compliant Numerator YES Are these data also directly accessible at the point of care by the reporting provider? NO Not numerator compliant YES Numerator compliant

25 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Figure 2: Measure Performance Eligible Population Is the member aged 12 or older as of January 1 st of the measurement year? YES Did the member NO Not included in eligible population Have continuous enrollment in the health plan for the measurement year and the year prior? YES Have at least one outpatient encounter during the intake period? NO Not included in eligible population YES Exclusions Did the member have a diagnosis of bipolar disorder during the measurement year or the year prior? OR A diagnosis of major depression or dysthymia during the year prior to the measurement year? YES Not included in denominator NO Was the member screened for clinical depression using an ageappropriate standardized tool? NO Not numerator compliant YES Numerator Did the member screen positive for clinical depression using an age-appropriate standardized tool? YES Did the member receive appropriate follow-up within 30 days of a positive result? NO NO YES Not numerator compliant Numerator compliant

26 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Depression Care for Adolescents and Adults Measure Work-Up Measure Descriptions 1. Utilization of the PHQ-9 to Monitor Depression Symptoms for Adolescents and Adults. The percentage of members 12 years of age and older with a diagnosis of major depression or dysthymia who have a PHQ-9 or PHQ-A tool administered at least once during a four-month period. 2. Depression Remission, Response or Treatment Adjustment for Adolescents and Adults. The percentage of members 12 years of age and older with a diagnosis of depression and an elevated PHQ-9 or PHQ-A score, who had evidence of response or remission within 5 7 months of the elevated PHQ-9 score or an indication of treatment adjustment within 30 days of the PHQ-9 score that showed no evidence of response or remission. 3. Depression Screening and Follow-up for Adolescents and Adults. The percentage of members 12 years of age and older who were screened for clinical depression using a standardized tool and, if screened positive, who received appropriate follow-up care. Topic Overview Importance and Prevalence Prevalence Depressive disorders are common mental disorders that occur in people of all ages. Major depressive disorder (MDD) is a leading cause of disability worldwide, affecting an estimated 120 million people (Murray et al., 2013). The lifelong prevalence is estimated to range from 10 percent 15 percent (Lepine and Briley, 2011). In the United States, 15.7 percent of people report that at some point in their lifetime they were told by a health care professional that they had depression (CDC, 2009). In 2008, the most recent year of data available, a nationally representative survey by the Substance Abuse and Mental Health Services Administration (SAMHSA, 2009) found there were 2.0 million youths (8.3 percent of the population aged 12 17) who had a major depressive episode during the past year and an estimated 1.5 million (6.0 percent) had an episode with severe impairment. Lifetime prevalence of depression and dysthymia increases from 8.4 percent for ages to 15 percent for ages (Merikangas, 2010). It has strong correlation to chronic and reoccurring depression in adulthood (Garber, 2009). Female adolescents are more likely to be diagnosed with depression than males (National Research Council and Institute of Medicine, 2009). One study found that female adolescents are also more likely to experience recurrence v (57.6 percent vs percent) (Curry et al., 2011). SAMHSA (2009) found that in 2008, 6.4 percent of adults (14.3 million) had at least one major depressive episode in the past year and more than 1 in 25 had an episode with severe impairment. The rate was highest among persons reporting two or more races (12.7 percent), while rates for single race groups were 7.0 percent among Whites, 5.2 percent among Hispanics, 4.9 percent among American Indians or Alaska Natives, 4.9 percent among Blacks and 3.6 percent among Asians. Prevalence of a major depressive episode was higher among adult females than among adult males (8.1 vs. 4.6 percent), particularly for females of child-bearing ages (SAMHSA, 2009). The high female-to-male sex ratio in the prevalence of depression, especially during the reproductive years, is one of the most replicated findings in epidemiology (Grigoriadis and Robinson, 2007). Late-life depression is also common. A systematic review and meta-analysis found the prevalence of major depression in older adults ranged from 4.6 percent to 9.3 percent (Luppa et al., 2012). There are misperceptions that depression symptoms are part of normal aging. Losses, social isolation and chronic medical problems that older patients experience can contribute to depression.

27 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Health Importance Depression an overwhelming feeling of sadness and hopelessness that can last for months or years can make people feel that life is no longer worth living. People affected by depression lose interest in activities they used to enjoy and can also be affected by physical symptoms that interfere with their ability to participate in normal daily activities. For adolescents, depression can also have a major impact, disrupting daily life at home, school or in the community. Depression is also associated with other chronic medical conditions and increased morbidity and mortality. The mortality risk for suicide in depressed patients is more than 20-fold greater than in the general population (Bostwick and Pankratz, 2000). In terms of other chronic conditions, depression is associated with a 60 percent increased risk of type 2 diabetes (Mezuk et al., 2008), and has been identified as a risk factor for development of cardiovascular disease (Van de Kooy et al., 2007). In addition, depression adversely affects the course, complications and management of other chronic medical illnesses (Katon, 2011). In adolescents, depression can also result in serious long-term morbidities such as generalized anxiety disorder and panic disorder or lead to engagement in risky behaviors such as substance use (Taylor et al., 1996; Foley et al., 1996; Friedman et al., 1996; National Research Council and Institute of Medicine., 2009). Adolescent-onset depression increases the risk of attempted suicide by five-fold in comparison to non-depressed adolescents (Garber, 2009). Most adolescents who commit suicide, the third leading cause of death among year olds, have a previous history of depression (Williams et al., 2009; National Research Council and Institute of Medicine, 2009). Depression has long been recognized as a major contributor to disease burden (Murray et al., 1997; Üstün et al., 2004). The Global Burden of Disease study of 2010 identified depression as a leading cause of disease burden in the world. Depressive disorders were the second largest contributor to years lived with disability, an indicator of the impact of disease burden (Ferrari et al., 2013). This accounts for an estimated 10 percent of Years Lived with Disability worldwide, which is three times the impact of diabetes, eight times the impact of heart disease, and forty times the impact of cancer (Murray et al., 2013). These findings underscore the need for attention to depressive disorders and the implementation of effective interventions to reduce their disease burden. Financial importance and cost effectiveness Depression has large effects on both health care costs and lost productivity. Adolescents with depression have higher medical expenditures, including those related to general and mental health care, than adolescents without a depression diagnosis (O Connor et al., 2009). For working-age adults, a recent study showed a relationship between the severity of depression symptoms and work function and found that for every 1-point increase in PHQ-9 score (a measure of depression severity), patients experienced an additional mean productivity loss of 1.65 percent. Even minor levels of depression symptoms were associated with decreases in work function (Beck et al., 2011). In a survey study, Birnbaum et al. (2011) found that major depressive disorder severity is significantly associated with increased treatment usage and costs, unemployment, disability and reduced work performance. When the results of the study were projected to the U.S. workforce, it was estimated that monthly depression-related worker productivity losses had human capital costs of nearly $2 billion. Older adults with depression or depressive symptoms have significantly higher health care costs even after adjusting for chronic medical conditions (Katon et al., 2003).

28 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Supporting Evidence for Depression Care Measures Numerous studies have demonstrated the effectiveness of screening and treatment for depression. Recent literature has focused on the care processes needed to treat and manage depression in primary care settings, where the majority of depression cases first present. Studies have found that patient outcomes improve when there is collaboration between a primary care doctor, case manager and a mental health specialist to screen for depression, monitor symptoms, provide treatment and refer to specialty care as needed (Von Korff and Goldberg, 2001; Gilbody et al., 2006; Thota et al., 2012). The following section includes information on the evidence for depression screening, tools to monitor depressive symptoms, treatment models, gaps in care and disparities. Screening and follow-up Screening for depression in adults when staff-assisted depression care supports are in place received a Grade B recommendation from the U.S. Preventive Services Task Force (USPSTF, 2009). The National Institute for Health and Clinical Excellence (NICE) guidelines recommend universal screening of adolescents for depression in primary care settings (NICE, 2005). Limited available data suggest that screening tools, feasible for use in the primary care setting, can accurately identify depressed individuals and treatment can improve depression outcomes (O Connor et al., 2009; Williams et al., 2009). The use of a standardized screening tool may help to reduce misdiagnosis, which one study suggests occurs in up to 60 percent of patients diagnosed with major depressive disorder (Mojtabai, 2013). In its review, the USPSTF found little evidence to support recommending one screening tool over another to identify depressed individuals accurately. Research has demonstrated that many brief self-administered tools are valid and reliable for identifying possible depression cases (Martin et al., 2006; Williams et al., 2002). Once a positive screen is identified, follow-up is necessary and may include further evaluation to determine or rule out a diagnosis, provide education or interventions or refer treatment with another provider. Monitoring depressive symptoms The use of standardized tools is essential for tracking depressive symptoms and monitoring patient response to treatment. Standardized instruments are useful in identifying meaningful change in clinical outcomes over time. Guidelines recommend that providers establish and maintain regular follow up with patients diagnosed with depression and use a standardized tool to track symptoms (Mitchel et al., 2013). Meta-analyses of studies in adults indicate that formally monitoring patient progress improves patient outcomes (Lambert et al., 2003; Shimokawa et al., 2010; Knaup., 2009). For adolescents, the Guideline for Adolescent Depression in Primary Care (GLAD-PC) recommends systematic and regular tracking of treatment goals and outcomes, including assessing depressive symptoms and function, monitoring for adverse events during antidepressant treatment and reassessing diagnosis and treatment if no improvement is noted after 6 8 weeks. One study found that youths with a range of symptoms improve more quickly when clinicians receive feedback from assessments every other week instead of every 3 months (Bickman et al., 2011). Existing gold standard instruments, such as the Hamilton Depression Rating Scale, can be time consuming and require a specially trained interviewer. The brief PHQ-9 questionnaire ( 2005 Pfizer) can be self-administered by the patient and has been validated for measuring depression severity and treatment response (Kroenke et al., 2001). The tool assesses the nine DSM, Fourth Edition, Text Revision (DSM-IV-TR) criterion symptoms and effects on functioning, and has been shown to be highly accurate in

29 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, discriminating patients with persistent major depression, partial remission and full remission (Gilbody et al., 2007; Lowe et al., 2004; Martin et al., 2006). Benefits of the PHQ-9 tool are numerous: it is non-proprietary and widely accepted by primary care providers and in general medical settings, it can be completed by the patient in-person or over the telephone, it is translated into many languages and it is easy for the patient to complete and the provider to score. Widespread use of the PHQ-9, within a collaborative care model, would allow organizations to systematically assess their effectiveness in helping individuals to experience remission of depressive symptoms with appropriate treatment. Interventions and treatment models to improve depression outcomes There are a number of effective treatment options available for depressive disorders, including antidepressant medications and psychotherapies. Guidelines recommend cognitive behavioral therapy and interpersonal therapy as first-line psychotherapy treatments for depression and selective serotonin reuptake inhibitors (SSRI) as firstline pharmacotherapy (APA, 2010; National Collaborating Centre for Mental Health, 2009). Clinical guidelines also recommend a stepped-care approach to depression treatment, beginning with the least-intrusive intervention and stepping up to more intensive care if the patient does not respond to or benefit from the first intervention (National Collaborating Centre for Mental Health, 2009; Mitchell et al., 2013). For mild and moderate depression, psychotherapy alone may be the preferred initial treatment, to be followed by the use of medication if symptoms persist (APA, 2010). This stepped-care approach includes providing assessment, support, psychoeducation and monitoring of symptoms as a first step, followed by psychosocial, psychological and pharmacologic interventions, and then combined treatments for those with inadequate response. High-intensity interventions, crisis and inpatient care are only used in severe cases. For adolescents, the USPSTF found adequate evidence that treatment with SSRIs, psychotherapy and combined therapy (SSRIs and psychotherapy) results in decreases of major depressive disorder symptoms. This conclusion was based on a systematic review that revealed several fair- or good-quality randomized controlled trials (RCT) (USPSTF, 2009). There are challenges to delivering guideline-recommended care in nonmental health settings such as primary care, where providers may not be as knowledgeable about depression management and there are competing demands of other medical issues (Nutting et al., 2002; Rost et al., 2000). Numerous studies have shown that a collaborative care model can address these challenges and demonstrate effectiveness for managing depression in primary care settings (Gilbody et al., 2006; Katon et al., 2008; Katon and Guico-Pabia, 2011). A recent RCT demonstrated effectiveness of the collaborative care model for adolescent depression, as well (Richardson et al., 2014). The model includes primary care providers using evidence-based approaches to depression care and a standardized tool for measuring severity of symptoms, response to treatment plan and remission. Key concepts of this approach are: Care management by a nonphysician working with the primary care physician. Planned collaborative care between physicians and mental health clinicians. Education and support of patients for self-management. Attention to patient preferences.

30 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Patients are tracked and reminded of visits with their primary care physician and monitored for treatment adherence and effectiveness. A care manager is typically used to make frequent contacts with patients, often by telephone, to provide education and self-management support and to monitor for response to treatment. If the patient does not respond to a treatment, other treatment options are explored and delivered (Solberg, 2005). This model has demonstrated improvement in treatment adherence, patient quality of life and depression outcomes. Preliminary evidence suggests the collaborative care model is also effective for depression during pregnancy and postpartum (Gjerdingen et al., 2008) and in treating late-life depression (Unutzer et al., 2002; Hunkeler et al., 2006). New models of depression treatment, such as computer-based therapy, also hold promise for expanding the reach of effective treatment. A systematic review of computer-based psychological treatments for depression and a meta-analysis of 19 RCTs support the efficacy and effectiveness of computer-based psychotherapy for depression in diverse settings and in different populations (Richards and Richardson, 2012). Health care disparities Using data from a large national survey, Gonzalez et al. (2010) found that few Americans with recent major depression receive guideline-concordant therapies, but the lowest rates of use are found among Mexican Americans and Blacks. Minority children are one-third to one-half less likely to receive mental health care as White children, despite a similar overall prevalence of disease. Moreover, of those who do receive care, these minority groups are less likely to receive complete services and are more likely to receive treatment that is inappropriate, fragmented or inadequate (Holm-Hanse, 2006). Hispanic and uninsured children have especially high rates of unmet need for mental health services, relative to other children (Kataoka et al., 2002). There are also gender disparities in receiving treatment for depression. In 2008, women who had a major depressive episode in the past year were more likely than their male counterparts to have received treatment for depression (74.2 vs percent) (SAMHSA, 2009). In terms of insurance coverage, among adults with a pastyear major depressive episode in 2008, about two-thirds of those with no insurance (64.1 percent) and commercial insurance (69.8 percent) received treatment for depression in the past year, compared with higher rates for those with Medicaid (83.1 percent) and other health insurance (83.5 percent), including Medicare and VA benefits (SAMHSA, 2009). Gaps in care There are significant quality concerns along the continuum of depression care (Katon and Guico-Pabia, 2011): under-diagnosis (Goldman et al., 1999), under-treatment (Kessler et al., 2003), inappropriate treatment (Mojtabai and Olfson, 2011), lack of follow-up and monitoring (Katon and Seelig, 2008). Quality gaps are more pronounced among ethnic and racial minorities (Gonzalez et al., 2010) and individuals with multiple chronic conditions (Katon et al., 2004). In a large representative survey study, only one-third of those with depression reported receiving mental health services in a given year and only about half had any type of health service use (Wang et al., 2005).

31 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, For adolescents, only 25 percent of those diagnosed with depression actually receive treatment; among those who go undetected, 20 percent develop recurrent or chronic depression (O Connor et al., 2009; Garber et al., 2009). Provider perception of diagnosis and treatment may contribute to low screening and treatment rates. In a systematic review, Williams et al (2009) found that while a majority of studies have shown high rates of treatment for depression upon diagnosis, one survey of pediatricians found that only 25 percent believed it was their responsibility to treat depression in adolescents. 86 percent expressed concern with prescribing medications; 90 percent expressed concern with counseling. Other studies in the review revealed that physicians who do screen for depression report they do not systematically use a standardized tool or the DSM-IV criteria. References AAP Committee on Pediatric Emergency Medicine Technical Report Pediatric and Adolescent Mental Health Emergencies in the Emergency Medical Services System. Pediatrics 127(5):e APA Practice Guideline for the Treatment for Patients with Major Depressive Disorder, 3rd Ed. Arlington, VA: Workgroup on Major Depressive Disorder, American Psychiatric Association. Beck, A., A.L Crain, L.I. Solberg, et al Severity of Depression and Magnitude of Productivity Loss. Annals of Family Medicine 9: Bickman, L., S.D. Kelley, C. Breda, A.R. de Andrade, M. Riemer Effects of Routine Feedback to Clinicians on Mental Health Outcomes of Youths: Results of a Randomized Trial. Psychiatric Services 62(12): Birnbaum, H. G., R.C. Kessler, D. Kelley, R. Ben Hamadi, V.N. Joish, P.E. Greenberg Employer Burden of Mild, Moderate, and Severe Major Depressive Disorder: Mental Health Services Utilization and Costs, and Work Performance. Depression and Anxiety 27(1): Bostwick, J.M., V.S.Pankratz Affective Disorders and Suicide Risk: a Reexamination. American Journal of Psychiatry 157: Centers for Disease Control and Prevention Anxiety and Depression Effective Treatments Exist: People with Depression and Anxiety Should Seek Help as Early as Possible to Reduce Health Effects and Improve Quality of Life. Based on 2006 Behavior Risk Factor Surveillance System. Cheung, A.H., R.A. Zuckerbrot, P.S. Jensen, K. Ghalib, D. Laraque, R.E.K. Stein GLAD-PC Steering Group. Guidelines for Adolescent Depression in Primary Care (GLAD-PC): II. Treatment and Ongoing management. Pediatrics 120(5):e Child Mind Institute Major Depressive Disorder. Curry, J., S. Silva, P. Rohde, G. Ginsburg, C. Kratochvil, A. Simons, J. March, et al Recovery and Recurrence Following Treatment for Adolescent Major Depression. Archives of General Psychiatry 68(3): Ferrari, A. J., F.C. Charlson, R.E. Norman, S.B. Patten, G. Freedman, C.J.L. Murray, T. Vos, H.A. Whiteford Burden of Depressive Disorders by Country, Sex, Age, and Year: Findings from the Global Burden of Disease Study PLoS Medicine 10(11):e Foley, H.A., C.O. Carlton, R.J Howell The Relationship of Attention Deficit Hyperactivity Disorder and Conduct Disorders to Juvenile Delinquency: Legal Implications. Bulletin of the American Academy of Psychiatry Law 24: Friedman, R.M., J.W. Katz-Levey, R.W Manderschied, D.L. Sondheimer Prevalence of Serious Emotional Disturbance in Children and Adolescents. In: Manderscheid, R.W., and M.A. Sonnenschein (eds.). Mental Health, United States.Rockville, MD: Center for Mental Health Services, Garber, J. et al Prevention of Depression in At-Risk Adolescents: A Randomized Controlled Trial. Journal of the American Medical Association 301(21):

32 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Gilbody, S., D. Richards, S. Brealey, C. Hewitt Screening for Depression in Medical Settings with the Patient Health Questionnaire (PHQ): A Diagnostic Meta-Analysis. Journal of General Internal Medicine 22(11): Gilbody, S., P. Bower, J. Fletcher, D. Richards, A.J. Sutton Collaborative Care for Depression: A Cumulative Meta-Analysis and Review of Longer-Term Outcomes. Archives of Internal Medicine 166(21): Gjerdingen, D., W. Katon, D.E. Rich Stepped Care Treatment of Postpartum Depression: A Primary Care-Based Management Model. Women's Health 18(1): González, H.M., W.A. Vega, D.R. Williams, W. Tarraf, B.T. West, H.W. Neighbors Depression Care in the United States: Too Little For Too Few. Archives of General Psychiatry 67(1):37. Grigoriadis, S. and G.E. Robinson Gender Issues in Depression. Annals of Clinical Psychiatry 19(4): Holm-Hansen, C Racial and Ethnic Disparities in Children s Mental Health. Saint Paul, MN: Wilder Research. %20Children's%20Mental%20Health/Racial%20and%20Ethnic%20Disparities%20in%20Children%E2%80 %99s%20Mental%20Health,%20Full%20Report.pdf. Hunkeler, E.M., W. Katon, L. Tang, J.W. Williams, K. Kroenke, E.B. Lin, L.H. Harpole et al Long Term Outcomes from the IMPACT Randomized trial for Depressed Elderly Patients in Primary Care. British Medical Journal 332(7536): Kataoka, S.H., L. Zhang, K.B. Wells Unmet Need for Mental Health Care Among US Children: Variation by Ethnicity and Insurance Status. American Journal of Psychiatry 159(9): Katon, W.J Epidemiology and Treatment of Depression in Patients with Chronic Medical Illness. Dialogues in Clinical Neuroscience 13(1):7. Katon, W., and C.J. Guico-Pabia Improving Quality of Depression Care Using Organized Systems of Care: A Review of the Literature. The Primary Care Companion to CNS Disorders, 13(1). Katon W.J., G. Simon, J. Russo, et al Quality of Depression Care in a Population-Based Sample of Patients with Diabetes and Major Depression. Medical Care, 42(12): Katon W.J., L.E Russo, J. Unutzer Increased Medical Costs of a Population-Based Sample of Depressed Elderly Patients. Archives of General Psychiatry 60: Katon, W.J., and M. Seelig Population-Based Care of Depression: Team Care Approaches to Improving Outcomes. Journal of Occupational and Environmental Medicine 50(4): Kessler, R.C., P. Berglund, O. Demler, et al The Epidemiology of Major Depressive Disorder: Results from the National Comorbidity Survey Replication (NCS-R). Journal of the American Medical Association 289(23): Kessler R.C., W.T. Chiu, O. Demler, E.E. Walters Prevalence, Severity, and Comorbidity of Twelve- Month DSM-IV Disorders in the National Comorbidity Survey Replication (NCS-R). Archives of General Psychiatry 62(6): Knaup C., M. Koesters, D. Schoefer, T. Becker, B. Puschner Effect of Feedback of Treatment Outcome in Specialist Mental Healthcare: Meta-Analysis. British Journal of Psychiatry 195(1): Kroenke, K., R.L. Spitzer, J.B.W. Williams The PHQ 9. Journal of General Internal Medicine 16(9): Lambert, M.J., J.L. Whipple, E.J. Hawkins, D.A. Vermeersch, S.L. Nielsen, D.W. Smart Is It Time for Clinicians to Routinely Track Patient Outcome? A Meta-Analysis. Clinical Psychology: Science and Practice. 10(3): Lépine, J.P., M. Briley The Increasing Burden of Depression. Neuropsychiatric Disease and Treatment. 7(suppl 1):3-7. Löwe, B., K. Kroenke, W. Herzog, K. Gräfe Measuring Depression Outcome with a Brief Self-Report Instrument: Sensitivity to Change of the Patient Health Questionnaire (PHQ-9). Journal of Affective Disorders 81(1): Luppa, M., C. Sikorski, T. Luck, L. Ehreke, A. Konnopka, B. Wiese, S. Weyerer, H-H. König, S.G. Riedel- Heller Age-and Gender-Specific Prevalence of Depression in Latest-Life Systematic Review and Meta-Analysis. Journal of Affective Disorders 136(3): Martin, A., W. Rief, A. Klaiberg, E. Braehler Validity of the Brief Patient Health Questionnaire Mood Scale (PHQ-9) in the General Population. General Hospital Psychiatry 28(1):71-77.

33 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Merikangas, K.R., J.P. He, M. Burstein, et al Lifetime Prevalence of Mental Disorders in U.S. Adolescents: Results from the National Comorbidity Survey Replication Adolescent Supplement (NCS- A). Journal of the American Academy of Child and Adolescent Psychiatry 49(10): Mezuk, B., W.W. Eaton, S. Albrecht, S.H. Golden Depression and Type 2 Diabetes Over the Lifespan: A Meta-Analysis. Diabetes Care 31: Mitchell, J., M. Trangle, B. Degnan, T. Gabert, B. Haight, D. Kessler, N. Mack, E. Mallen, H. Novak, D. Rossmiller, L. Setterlund, K. Somers, N. Valentino, S. Vincent Institute for Clinical Systems Improvement. Adult Depression in Primary Care. Updated September Mojtabai, R Clinician-Identified Depression in Community Settings: Concordance with Structured- Interview Diagnoses. Psychotherapy and Psychosomatics, 82(3): Mojtabai, R., M. Olfson Proportion of Antidepressants Prescribed Without a Psychiatric Diagnosis is Growing. Health Affairs 30(8): Murray, C.J.L, and A.D. Lopez Global Mortality, Disability, and the Contribution of Risk Factors: Global Burden of Disease Study. The Lancet 349(9063): Murray, C.J.L., T. Vos, R. Lozano, M. Naghavi, A.D. Flaxman, C. Michaud, M. Ezzati, et al Disability- Adjusted Life Years (DALYs) for 291 Diseases and Injuries in 21 regions, : a Systematic Analysis for the Global Burden of Disease Study The Lancet 380(9859): National Collaborating Centre for Mental Health Depression: The Treatment and Management of Depression in Adults. London (UK): National Institute for Health and Clinical Excellence (NICE); 2009 Oct. 64 p. (Clinical guideline; no. 90). National Institute for Health and Clinical Excellence (NICE) Clinical Guideline 28: Depression in Children and Young People: Identification and Management in Primary, Community, and Secondary Care. London, UK: NHS. NICE, Depression: Treatment and Management of Depression in Adults Including Adults with Chronic Physical Health Problems. London: National Institute for Health and Clinical Excellence. Nutting, P.A., K. Rost, M. Dickinson, J.J. Werner, P. Dickinson, J.L. Smith, B. Gallovic Barriers to Initiating Depression Treatment in Primary Care Practice. Journal of General Internal Medicine 17(2): O'Connor, E.A., E.P. Whitlock, T.L. Beil, B.N. Gaynes Screening for Depression in Adult Patients in Primary Care Settings: A Systematic Evidence Review. Annals of Internal Medicine 151(11): Richards, D., and T. Richardson Computer-Based Psychological Treatments for Depression: A Systematic Review and Meta-Analysis. Clinical Psychology Review 32(4): Richardson, L.P., E. Ludman, E. McCauley, J. Lindenbaum, C. Larison, C. Zhou, G. Clarke, D. Brent, W. Katon Collaborative Care for Adolescents with Depression in Primary Care: a Randomized Clinical Trial. Journal of the American Medical Association 312(8): Rost, K., P. Nutting, J. Smith, J.C. Coyne, L. Cooper-Patrick, L. Rubenstein The Role of Competing Demands in the Treatment Provided Primary Care Patients with Major Depression. Archives of Family Medicine 9(2):150. Shimokawa K., M.J. Lambert, D.W. Smart Enhancing Treatment Outcome of Patients At Risk of Treatment Failure: Meta-Analytic and Mega-Analytic Review of a Psychotherapy Quality Assurance System. Journal of Consulting Clinical Psychology 78(3): Simon G.E Evidence Review: Efficacy and Effectiveness of Antidepressant Treatment in Primary Care. General Hospital Psychiatry 24(4): Solberg, L.I., M.A. Trangle, A.P. Wineman Follow-Up and Follow-Through of Depressed Patients in Primary Care: The Critical Missing Components Of Quality Care. The Journal of the American Board of Family Practice 18(6): Substance Abuse and Mental Health Services Administration Results from the 2008 National Survey on Drug Use and Health: National Findings (Office of Applied Studies, NSDUH Series H-36, HHS Publication No. SMA ). Rockville, MD. 2k8NSDUH/2k8results.cfm#Ch8 Taylor, E. et al Hyperactivity and Conduct Problems as Risk Factors for Adolescent Development. Journal of the American Academy of Child and Adolescent Psychiatry 35: Thota, A.B., T.A. Sipe, G.J. Byard, C.S. Zometa, R.A. Hahn, L.R. McKnight-Eily, D.P. Chapman et al Collaborative Care to Improve the Management of Depressive Disorders: A Community Guide Systematic Review and Meta-Analysis. American Journal of Preventive Medicine 42(5):

34 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Unützer, J., W. Katon, C.M. Callahan, J.W. Williams Jr, E. Hunkeler, L. Harpole, M. Hoffing et al Collaborative Care Management of Late-Life Depression in the Primary Care Setting: A Randomized Controlled Trial. Journal of the American Medical Association 288(22): U.S. Preventive Services Task Force Screening and Treatment for Major Depressive Disorder in Children and Adolescents: US Preventive Services Task Force Recommendation Statement. Pediatrics 23: Üstün, T.B., J.L. Ayuso-Mateos, S. Chatterji, C. Mathers, C.J.L. Murray Global Burden of Depressive Disorders in the Year The British Journal of Psychiatry 184(5): Van der Kooy, K., H. van Hout, H. Marwijk, H. Marten, C. Stehouwer, A. Beekman Depression and the Risk for Cardiovascular Diseases: Systematic Review and Meta-Analysis. International Journal of Geriatric Psychiatry 2: Von Korff, M., D. Goldberg Improving Outcomes in Depression. British Medical Journal 323:948 9 Wang, P.S., M. Lane, M. Olfson, H.A. Pincus, K.B. Wells, R.C. Kessler Twelve Month Use of Mental Health Services in the United States. Archives of General Psychiatry 62(6): Williams Jr, J. W., P.H. Noël, J.A. Cordes, G. Ramirez, M. Pignone Is This Patient Clinically Depressed? Journal of the American Medical Association 287(9): Williams, S.B., E.A. O'Connor, M. Eder, E.P. Whitlock Screening for Child and Adolescent Depression in Primary Care Settings: A Systematic Evidence Review for the US Preventive Services Task Force. Pediatrics 123(4):e716-e735.

35 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Proposed New Risk Adjusted Measures for HEDIS 2016: Inpatient Hospital Utilization Emergency Department Utilization NCQA seeks comments on the following proposed new risk-adjusted utilization measures for inclusion in the HEDIS 2016 measurement set: 1. Inpatient Hospital Utilization. For members 18 years of age and older, the number of acute inpatient discharges during the measurement year (observed) and the predicted probability of inpatient discharges (expected). Reported in the following categories: Surgery and Medicine. 2. Emergency Department Utilization. For members 18 years of age and older, the number of emergency department (ED) visits during the measurement year (observed) and the predicted probability of ED visits (expected). The aim of applying a risk adjustment strategy to these utilization measures is to allow better comparison of inpatient and ED use across health plans and to create an even playing field by removing the effect of select patient characteristics and health status differences on the reported results. NCQA investigated the appropriateness of developing these risk adjusted HEDIS measures by building from existing, unadjusted measures: Inpatient Utilization General Hospital/Acute Care and Ambulatory Care (ED Visits Only). Since 1993, these measures have reported the unadjusted total discharges per member month/ year from acute inpatient care and ED outpatient services across health plan members of all ages. Risk modeling and testing were undertaken to assess whether risk adjustment might improve the measures ability to highlight quality differences, such as the impact of effective care coordination and other system interventions, in reducing inpatient admissions and ED overutilization. The proposed risk adjusted specifications are intended to enhance our understanding of variation in utilization and comparability of inpatient and ED utilization among plans. NCQA used a comprehensive approach when assessing appropriate strategies for reporting adjusted utilization in HEDIS. We conducted a number of stakeholder interviews with health plans and risk adjustment experts, most of whom support the application of a risk adjustment strategy to HEDIS utilization measures. We then employed a large research database of Medicare Advantage and commercial plan members to model several variations of risk adjustment. Test results reveal that risk adjustment is a desirable refinement and demonstrate that the proposed risk adjustment strategy is both accurate and reliable. NCQA s advisory panels agree that the results support the reliability of the risk adjustment model and that the measures can help identify opportunities for quality improvement. Supporting documents for the proposed measure include the draft measure specifications. NCQA acknowledges the contributions of the Risk Adjustment Expert Work Group, the Geriatric Measurement Advisory Panel and the Technical Measurement Advisory Panel.

36 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, SUMMARY OF CHANGES TO HEDIS 2016 First-year measure. Inpatient Hospital Utilization (IHU) Description The number of acute inpatient discharges during the measurement year. Data are reported in the following categories: Number of inpatient surgery discharges. Number of inpatient medicine discharges. Number of total inpatient discharges. Expected count of inpatient surgery discharges Expected count of inpatient medicine discharges. Expected count of total discharges. Definitions Classification period The year prior to the measurement year. Eligible Population Product lines Ages Continuous enrollment Allowable gap Anchor date Benefit Event/diagnosis Commercial, Medicare (report each product line separately). 18 and older as of December 31 of the measurement year. The measurement year and the year prior to the measurement year. No more than one gap in enrollment of up to 45 days during each year of continuous enrollment. December 31 of the measurement year. Medical. None.

37 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Administrative Specification Denominator Numerator The eligible population. For organizations that use MS-DRGs: Identify all acute inpatient stays with a discharge date during the measurement year for the following categories: Surgery (Surgery MS-DRG Value Set). Medicine (Medicine MS-DRG Value Set). Total Inpatient (the sum of Surgery and Medicine). If the organization does not use MS-DRGs, use the following steps to identify inpatient discharges. Step 1 Step 2 Identify all acute inpatient stays with a discharge date during the measurement year. Eliminate discharges with: A principal diagnosis of mental health of chemical dependency (Mental and Behavioral Disorders Value Set). A principal diagnosis of live-born infant (Deliveries Infant Record Value Set). A maternity-related principal diagnosis (Maternity Diagnosis Value Set). A maternity-related stay (Maternity Value Set). Step 3 Report total inpatient, using all discharges identified after completing steps 1 and 2. Step 4 Report surgery. From discharges remaining after removing required exclusions (step 2) from total inpatient (step 1); identify surgery (Surgery Value Set). Step 5 Report medicine. Categorize as medicine the discharges remaining after removing surgery (step 4). Risk Adjustment Determination For each member in the eligible population, use the following steps to identify risk adjustment categories based on presence of comorbidity, age and gender. Comorbidities Step 1 Identify all diagnoses for encounters during the classification period. Include the following when identifying encounters: Outpatient visits (Outpatient Value Set). Observation visits (Observation Value Set). Nonacute inpatient encounters (Nonacute Inpatient Value Set). Acute inpatient encounters (Acute Inpatient Value Set). ED visits (ED Value Set).

38 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Step 2 Assign each diagnosis to one comorbid Clinical Condition (CC) category using Table CC Comorbid. Exclude all diagnoses that cannot be assigned to a comorbid CC category. For members with no qualifying diagnoses from face-to-face encounters, skip to the Risk Adjustment Weighting section. All digits must match exactly when mapping diagnosis codes to the comorbid CCs. Step 3 Determine HCCs for each comorbid CC identified. Refer to Table HCC Rank. For each member s comorbid CC list, match the comorbid CC code to the comorbid CC code in the table, and assign: The ranking group. The rank. The HCC. For comorbid CCs that do not match to Table HCC Rank, use the comorbid CC as the HCC and assign a rank of 1. Note: One comorbid CC can map to multiple HCCs; each HCC can have one or more comorbid CCs. Step 4 Assess each ranking group separately and select only the highest ranked HCC in each ranking group using the Rank column (1 is the highest rank possible). Drop all other HCCs in each ranking group, and de-duplicate the HCC list if necessary. Note: Refer to the Plan All-Cause Readmissions (PCR) measure for a Comorbid CC calculation example. Step 5 Identify combination HCCs listed in Table HCC Comb. Some combinations suggest a greater amount of risk when observed together. For example, when diabetes and CHF are present, an increased amount of risk is evident. Additional HCCs are selected to account for these relationships. Compare each stay s list of unique HCCs to those in the HCC column in Table HCC Comb and assign any additional HCC conditions. For fully nested combinations (e.g., the diabetes/chf combination is nested in the diabetes/ CHF/renal combination), use only the more comprehensive pattern. In this example, only the diabetes/chf/renal combination is counted. For overlapping combinations (e.g., the CHF, COPD combination overlaps the CHR/renal/ diabetes combination), use both sets of combinations. In this example, both CHF/COPD and CHF/renal/diabetes combinations are counted. Based on the combinations, a member can have none, one or more of these added HCCs. Example Note: Refer to the PCR measure for a combination HCC calculation example.

39 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Risk Adjustment Weighting The calculation of risk-adjusted outcomes (counts of discharges) use a two-step model. In the first step, logistic regression is used to estimate the probability of having any discharge in the measurement year. In the second step, a Poisson regression model is used to predict the count of discharges among those patients who had at least one discharge in the measurement year. The results from each model are then combined to predict for each patient, how many discharges a member may have in the measurement year given the member s age, gender and comorbidities. Separate risk adjustment weights will be provided for the logistic and Poisson regression models. For each member in the eligible population, use the following steps to identify risk adjustment weights for each product line based on presence of comorbidity, age and gender. Steps 1-5 are performed for the logistic regression model, steps 6-10 are performed for the Poisson regression model. Then using the results from each model (step 5 and step 10, respectively), proceed through step 11. Note: The final weights table will be released on November 2, Complete steps 1 5 for the logistic regression model. Step 1 For each member with a comorbidity HCC Category, link the weights. For Commercial: Use Table XX. For Medicare: Use Table XX. Step 2 Link the age and gender weights for each member. For Commercial: Use Table XX. For Medicare: Use Table XX. Step 3 Identify the base risk weight. For Commercial: Use Table XX. For Medicare: Use Table XX. Step 4 Step 5 Sum all weights associated with the member (i.e., comorbidities, age, gender and base risk weight) Use the formula below to calculate the predicted probability of discharges based on the sum of the weights for each member for each numerator (inpatient surgery, inpatient medicine, total inpatient) (step 4). Calculate the predicted probability of each member having at least one discharge in the measurement year. This comes from the sum of weights from the logistic regression model using the formula below: Predicted probability of any discharge = Complete steps 6 10 for the Poisson regression model. ee ( Logistic Regression Weights For Each Member) 1+ee ( Logistic Regression Weights For Each Member) Step 6 For each member with a comorbidity HCC Category, link the weights. For Commercial: Use Table XX. For Medicare: Use Table XX. Step 7 Link the age and gender weights for each member. For Commercial: Use Table XX.

40 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, For Medicare: Use Table XX. Step 8 Step 9 Step 10 Identify the base risk weight. For Commercial: Use Table XX. For Medicare: Use Table XX. Sum all weights associated with the member (i.e., comorbidities, age, gender and base risk weight). Use the formula below to calculate the predicted unconditional count of discharges based on the sum of the weights for each member (step 9). Calculate the predicted unconditional count of events. These predicted counts are not adjusted for the likelihood of having any discharges. ( Poisson Regression Weights For Each Member) Predicted unconditional count of discharges = ee Step 11 Calculate the final member-level expected count of discharges using the formula below: Expected count of discharges = Predicted probability of any discharge Predicted unconditional count of discharges. Reporting: Denominator Count the number of members in the eligible population for each age and gender group and the overall total. Enter these values into the reporting table (Table IHU-A-3). Reporting: Risk Adjustment Step 1 Step 2 For each numerator (surgery, medicine, total) calculate the sum of expected counts of discharges (step 11) across all members within each age and gender group and the overall total. Round to two decimal places using the.5 rule and enter these values into the reporting table. Note: Do not take the average of the cells in the reporting table (Table IHU-B-3, Table IHU-C-3, Table IHU-D-3) Note Organizations may not use risk assessment protocols to supplement diagnoses for calculation of the risk adjustment scores for this measure. The IHU measurement model was developed and tested using only claims-based diagnoses; and diagnoses from additional data sources would affect the validity of the models as they are currently implemented in the specifications.

41 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Reporting: Numerator Count the number of acute inpatient discharges for each numerator (Surgery, Medicine, and total) within the measurement year for each age strata and enter these values into the reporting table. Table IHU-A-3: Number of Members in the Eligible Population Age Sex Members Male Female Male Female Male Female Male Female Male Female Male Female Total Total Total Total Total Total Total

42 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Table IHU-B-3: Inpatient Discharges by Age and Risk Adjustment: Surgery Age Sex Inpatient Discharges Inpatient Discharges/1,000 Members Expected Count of Discharges O/E Ratio (Observed Discharges/ Expected Count) Male Female Total Male Female Total Male Female Total Male Female Total Male Female Total Male Female Total Total

43 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Table IHU-C-3: Inpatient Discharges by Age and Risk Adjustment: Medicine Age Sex Inpatient Discharges Inpatient Discharges/1,000 Members Expected Count of Discharges O/E Ratio (Observed Discharges/ Expected Count) Male Female Total Male Female Total Male Female Total Male Female Total Male Female Total Male Female Total Total

44 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Table IHU-D-3: Inpatient Discharges by Age and Risk Adjustment: Total Age Sex Inpatient Discharges Inpatient Discharges/1,000 Members Expected Count of Discharges O/E Ratio (Observed Discharges/Expected Count) Male Female Total Male Female Total Male Female Total Male Female Total Male Female Total Male Female Total Total

45 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, SUMMARY OF CHANGES TO HEDIS 2016 First-year measure. Emergency Department Utilization (EDU) Description The number of emergency department (ED) visits during the measurement year. Data are reported as follows: Number of ED visits. Expected count of ED visits. Definitions Classification Period The year prior to the measurement year. Eligible Population Product lines Ages Continuous enrollment Allowable gap Anchor date Benefit Event/diagnosis Commercial, Medicare (report each product line separately). 18 and older as of December 31 of the measurement year. The measurement year and the year prior to the measurement year. No more than one gap in enrollment of up to 45 days during each year of continuous enrollment. December 31 of the measurement year. Medical. None. Administrative Specification Denominator Numerator Step 1 The eligible population Count each visit to an ED once, regardless of the intensity or duration of the visit. Count multiple ED visits with the same date of service as one visit. Do not include ED visits that result in an inpatient admission. Identify eligible ED visits using either of the following: An ED Visit (ED Value Set) A procedure code (ED procedure Code Value Set) with an ED place of service code (ED POS Value Set)

46 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Step 2: Eliminate claims and encounters that indicate the encounter was for mental health or chemical dependency using any of the following: A principal diagnosis of mental health or chemical dependency (Mental and Behavioral Disorders Value Set). Psychiatry (Psychiatry Value Set). Electroconvulsive Therapy (Electroconvulsive Therapy Value Set). Alcohol or drug rehabilitation or detoxification (AOD Rehab and Detox Value Set). Risk Adjustment Determination For each member in the eligible population, use the following steps to identify risk adjustment categories based on presence of comorbidity, age and gender. Comorbidities Step 1 Step 2 Identify all diagnoses for encounters during the classification period. Include the following when identifying encounters: Outpatient visits (Outpatient Value Set). Observation visits (Observation Value Set). Nonacute inpatient encounters (Nonacute Inpatient Value Set). Acute inpatient encounters (Acute Inpatient Value Set). ED visits (ED Value Set). Assign each diagnosis to one comorbid Clinical Condition (CC) category using Table CC Comorbid. Exclude all diagnoses that cannot be assigned to a comorbid CC category. For members with no qualifying diagnoses from face-to-face encounters, skip to the Risk Adjustment Weighting section. All digits must match exactly when mapping diagnosis codes to the comorbid CCs. Step 3 Determine HCCs for each comorbid CC identified. Refer to Table HCC Rank. For each member s comorbid CC list, match the comorbid CC code to the comorbid CC code in the table, and assign: The ranking group. The rank. The HCC. For comorbid CCs that do not match to Table HCC Rank, use the comorbid CC as the HCC and assign a rank of 1. Note: One comorbid CC can map to multiple HCCs; each HCC can have one or more comorbid CCs.

47 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Step 4 Example Step 5 Assess each ranking group separately and select only the highest ranked HCC in each ranking group using the Rank column (1 is the highest rank possible). Drop all other HCCs in each ranking group, and de-duplicate the HCC list if necessary. Note: Refer to the Plan All-Cause Readmissions (PCR) measure for a Comorbid CC calculation example. Identify combination HCCs listed in Table HCC Comb. Some combinations suggest a greater amount of risk when observed together. For example, when diabetes and CHF are present, an increased amount of risk is evident. Additional HCCs are selected to account for these relationships. Compare each stay s list of unique HCCs to those in the HCC column in Table HCC Comb and assign any additional HCC conditions. For fully nested combinations (e.g., the diabetes/chf combination is nested in the diabetes/ CHF/renal combination), use only the more comprehensive pattern. In this example, only the diabetes/chf/renal combination is counted. For overlapping combinations (e.g., the CHF, COPD combination overlaps the CHR/renal/ diabetes combination), use both sets of combinations. In this example, both CHF/COPD and CHF/renal/diabetes combinations are counted. Example Based on the combinations, a member can have none, one or more of these added HCCs. Note: Refer to the PCR measure for a combination HCC calculation example. Risk Adjustment Weighting The calculation of risk-adjusted outcomes (counts of ED visits) uses a two-step model. In the first step, logistic regression is used to estimate the probability of having any ED visit in the measurement year. In the second step, a Poisson regression model is used to predict the count of ED visits among those patients who had at least one ED visit in the measurement year. The results from each model are then combined to predict for each patient, how many ED visits a member may have in the measurement year given the member s age, gender and comorbidities. Separate risk adjustment weights will be provided for the logistic and Poisson regression models. For each member in the eligible population, use the following steps to identify risk adjustment weights for each product line based on presence of comorbidity, age and gender. Steps 1-5 are performed for the logistic regression model, steps 6-10 are performed for the Poisson regression model. Then using the results from each model (step 5 and step 10, respectively), proceed through step 11. Note: The final weights table will be released on November 2, Complete steps 1 5 for the logistic regression model. Step 1 Step 2 For each member with a comorbidity HCC Category, link the weights. For Commercial: Use Table XX. For Medicare: Use Table XX. Link the age and gender weights for each member. For Commercial: Use Table XX. For Medicare: Use Table XX.

48 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Step 3 Identify the base risk weight. For Commercial: Use Table XX. For Medicare: Use Table XX. Step 4 Step 5 Sum all weights associated with the member (i.e., comorbidities, age, gender and base risk weight). Use the formula below to calculate the predicted probability of ED visits based on the sum of the weights for each numerator (step 4). Calculate the predicted probability of each member having at least one ED visit in the measurement year. This comes from the sum of weights from the logistic regression model using the formula below: Predicted probability of any ED visit = Complete steps 6 10 for the Poisson regression model. ee ( Logistic Regression Weights For Each Member) 1+ee ( Logistic Regression Weights For Each Member) Step 6 For each member with a comorbidity HCC Category, link the weights. For Commercial: Use Table XX. For Medicare: Use Table XX. Step 7 Link the age and gender weights for each member. For Commercial: Use Table XX. For Medicare: Use Table XX. Step 8 Identify the base risk weight. For Commercial: Use Table XX. For Medicare: Use Table XX. Step 9 Step 10 Sum all weights associated with the member (i.e., comorbidities, age, gender and base risk weight). Use the formula below to calculate the predicted unconditional count of ED visits based on the sum of the weights for each member (step 9). Calculate the predicted unconditional count of events. These predicted counts are not adjusted for the likelihood of having any ED visits. ( Poisson Regression Weights For Each Member) Predicted unconditional count of ED visits = ee Step 11 Calculate the final member-level predicted count of ED visits using the formula below: Expected count of ED visits = Predicted probability of any ED visit Predicted unconditional count of ED visits.

49 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Reporting: Denominator Count the number of members in the eligible population for each age and gender combination. Enter these values into the reporting table (Table EDU-A-3). Reporting: Risk Adjustment Step 1 Step 2 Calculate the sum of expected counts of ED visits (step 11) across all members within each age and gender group and the overall total. Round to two decimal places using the.5 rule and enter these values into the reporting table. Note: Do not take the average of the cells in the reporting table (Table EDU-B-3). Note Organizations may not use risk assessment protocols to supplement diagnoses for calculation of the risk adjustment scores for this measure. The EDU measurement model was developed and tested using only claims-based diagnoses; diagnoses from additional data sources would affect the validity of the models as they are currently implemented in the specifications. Reporting: Numerator Count the number of emergency department (ED) visits within the measurement year for each age and gender combination and enter these values into the reporting table. Table EDU-A-3: Number of Members in the Eligible Population Age Sex Members Male Female Total Male Female Total Male Female Total Male Female Total Male Female Total Male 85+ Female Total Total

50 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Table EDU-B-3: Number of ED visits by Age and Risk Adjustment Age Sex ED Visits ED Visits/ 1,000 Members Expected ED Visits O/E Ratio (Observed ED Visits/Expected Count) Male Female Total Male Female Total Male Female Total Male Female Total Male Female Total Male Female Total Total

51 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Proposed New Measure for HEDIS : Statin Therapy for Patients With Cardiovascular Disease NCQA seeks comments on the proposed new measure for inclusion in the HEDIS 2016 measurement set: Statin Therapy for Patients With Cardiovascular Disease. The percentage of males years of age and females years of age during the measurement year, who were identified as having clinical atherosclerotic cardiovascular disease (ASCVD) and were dispensed at least moderate-intensity statin therapy that they remained on for at least 80 percent of the treatment period. Two rates are reported: 1. Received Statin Therapy. The percentage of members who were identified as having clinical ASCVD and were dispensed at least moderate intensity statin therapy during the measurement year. 2. Statin Adherence 80 percent. The percentage of members who were identified as having clinical ASCVD and were dispensed at least moderate-intensity statin therapy that they remained on for at least 80 percent of the treatment period. This measure represents an important area for quality improvement in patients with cardiovascular disease by assessing the use of statin therapy at an appropriate intensity and adherence to reduce the risk for cardiovascular events. The measure is based on 2013 blood cholesterol guidelines from the American College of Cardiology and the American Heart Association (ACC/AHA). 2 Convincing evidence estimates the benefit of statin therapy and adherence to reduce the risk for cardiovascular events: Moderate intensity statin therapy lowers low-density lipoprotein cholesterol (LDL-C) by 30 percent to less than 50 percent, on average. High-intensity statin therapy lowers LDL-C by 50 percent or more, on average. 2 Every 25 percent increase in adherence to statin therapy results in a 3.8 mg/dl reduction in LDL-C levels. 3 Every 10 mg/dl reduction in LDL-C levels results in a 10 percent reduction in overall cardiovascular risk. 4 NCQA tested this measure in a large research database of commercially insured and Medicare Advantage individuals to assess importance, feasibility, validity and overall performance. We tested multiple aspects of the specifications including denominator identification and age ranges, exclusions, statin dosage intensities, statin dispensing events and adherence to statin medications. Testing results revealed that the methods used to identify the denominator are appropriate. The age limit for females captures the patient population to benefit from statin therapy, while accounting for the risk of pregnancy. NCQA s advisory panels agreed with the specified denominator identification methods and age limits. 1 HEDIS is a registered trademark of the National Committee for Quality Assurance (NCQA). 2 Stone, N.J., J. Robinson, A.H. Lichtenstein, C.N. Bairey Merz, D.M. Lloyd-Jones, C.B. Blum, P. McBride, R.H. Eckel, J.S. Schwartz, A.C. Goldberg, S.T. Shero, G.D. Smith, Jr, D. Levy, K. Watson, P.W.F. Wilson ACC/AHA Guideline on the Treatment of Blood Cholesterol to Reduce Atherosclerotic Cardiovascular Risk in Adults. Journal of the American College of Cardiology doi: /j.jacc Ho, P.M., C.L. Bryson, J.S. Rumsfeld Medication adherence: its importance in cardiovascular outcomes. Circulation 119(23): C.R. Rahilly-Tierney, E.V. Lawler, R.E. Scranton, J.M. Gaziano Cardiovascular benefit of magnitude of lowdensity lipoprotein cholesterol reduction. Circulation 120:

52 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, The prevalence of conditions that are contraindications for statin therapy was very low in the measure s denominator. Our advisory panels recommended pregnancy, ESRD, cirrhosis and rhabdomyolysis as exclusions for this measure for face validity. The measure also proposes to exclude women trying to become pregnant and will use claims for clomiphene and in vitro fertilization to identify these patients. Our advisory panels debated excluding patients with claims for myalgia, myositis or myopathy. They recognized that although muscle pain or weakness is a common side effect of statin therapy, indicating statin intolerance, patients experiencing those issues should not necessarily stop receiving treatment; lower dosage intensity or alternative statins may be prescribed instead. However, there is currently no method for using administrative data to accurately identify patients who experience intolerance to statins. With these considerations in mind, we seek comments specific to excluding patients with claims for myalgia, myositis or myopathy. Testing results also revealed low rates of patients taking the recommended statin dosage intensity and poor rates of adherence to statin therapy. NCQA s advisory panels strongly support a measure to assess at least moderate-intensity statins to improve quality care and reduce the risk for cardiovascular events in patients with established disease. Furthermore, our panels recommend aligning with the accepted standard of 80 percent proportion of days covered, to measure high medication adherence. We request comments on these issues: Patients on statin therapy often experience muscle pain and weakness as symptoms of statin intolerance. However, there is currently no method for accurate identification of patients who experience intolerance to statins using administrative data. Although the codes for myalgia, myositis or myopathy are nonspecific, it is possible that claims for these conditions could serve as a proxy for statin intolerance. NCQA seeks comments on the following options for consideration: 1. Exclude patients with claims for myalgia, myositis and myopathy. 2. Exclude patients with claims for only myositis and myopathy. Do not exclude patients with claims for myalgia because it is the least severe of the conditions. 3. Do not exclude patients with claims for myalgia, myositis or myopathy. Supporting documents for the proposed measure include the draft measure specification and associated measure rationale work-up. NCQA acknowledges the contributions of the Cardiovascular Measurement Advisory Panel, the Technical Measurement Advisory Panel and the HEDIS Coding Panel and the Pharmacy Panel.

53 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Statin Therapy for Patients With Cardiovascular Disease SUMMARY OF CHANGES TO HEDIS 2016 First-year measure. Description The percentage of males years of age and females years of age during the measurement year who were identified as having clinical atherosclerotic cardiovascular disease (ASCVD) and were dispensed at least moderate-intensity statin therapy that they remained on for at least 80 percent of the treatment period. Two rates are reported: 1. Received Statin Therapy. The percentage of members who were identified as having clinical ASCVD and were dispensed at least moderate-intensity statin therapy during the measurement year. 2. Statin Adherence 80 Percent. The percentage of members who were identified as having clinical ASCVD and were dispensed at least moderate-intensity statin therapy that they remained on for at least 80 percent of the treatment period. Definitions Clinical ASCVD IPSD Treatment period PDC Calculating number of days covered for multiple prescriptions Refer to Eligible Population Event/Diagnosis for member identification instructions. Index prescription start date. The earliest prescription dispensing date for any statin medication of at least moderate intensity during the measurement year. The period of time beginning on the IPSD through the last day of the measurement year. Proportion of days covered. The number of days the member is covered by at least one statin medication prescription of appropriate intensity, divided by the number of days in the treatment period. If multiple prescriptions for different medications are dispensed on the same day, calculate the number of days covered by a statin medication (for the numerator) using the prescriptions with the longest days supply. For multiple different prescriptions dispensed on different days with overlapping days supply, count each day in the treatment period only once toward the numerator. If multiple prescriptions for the same medication are dispensed on the same day or on different days, sum the days supply and use the total to calculate the number of days covered by a statin medication (for the numerator). For example, three prescriptions for the same medication are dispensed on the same day, each with a 30-day supply. Sum the days supply for a total of 90 days covered by a statin. Subtract any days supply that extends beyond December 31 of the measurement year. Use the drug ID provided by the NDC to determine if the prescriptions are the same or different.

54 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Eligible Population: Rate 1 Received Statin Therapy Product line Age Continuous enrollment Allowable gap Anchor date Benefit Event/Diagnosis Step 1: Commercial, Medicaid, Medicare (report each product line separately). Males years as of December 31 of the measurement year. Females years as of December 31 of the measurement year. The measurement year and the year prior to the measurement year. No more than one gap in enrollment of up to 45 days during each year of continuous enrollment. To determine continuous enrollment for a Medicaid beneficiary for whom enrollment is verified monthly, the member may not have more than a 1-month gap in coverage (i.e., a member whose coverage lapses for 2 months [60 days] is not considered continuously enrolled). December 31 of the measurement year. Medical during the measurement year and the year prior. Pharmacy during the measurement year. Follow the steps below to identify the eligible population. Members are identified for the eligible population in two ways: by event or by diagnosis. The organization must use both methods to identify the eligible population, but a member only needs to be identified by one method to be included in the measure. Event. Any of the following during the year prior to the measurement year meet criteria: MI. Discharged from an inpatient setting with an MI (MI Value Set). Use both facility and professional claims to identify MI. CABG. Discharged from an inpatient setting with a CABG (CABG Value Set). Use both facility and professional claims to identify CABG. PCI. Members who had PCI (PCI Value Set) in any setting. Other revascularization. Members who had any other revascularization procedures (Other Revascularization Value Set) in any setting. Diagnosis. Identify members as having ischemic vascular disease (IVD) who met at least one of the following criteria during both the measurement year and the year prior to the measurement year. Criteria need not be the same across both years. At least one outpatient visit (Outpatient Value Set) with an IVD diagnosis (IVD Value Set), or At least one acute inpatient encounter (Acute Inpatient Value Set) with an IVD diagnosis (IVD Value Set). Step 2: Required exclusions Exclude members who meet any of the following criteria: Pregnancy (Pregnancy Value Set) during the measurement year or year prior to the measurement year. In vitro fertilization (IVF Value Set) in the measurement year or year prior to the measurement year. Dispensed at least one prescription for clomiphene (Table XXX-X) during the measurement year or the year prior to the measurement year.

55 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, ESRD (ESRD Value Set) during the measurement year or the year prior to the measurement year. Table XXX-X: Medications to Identify Exclusions Cirrhosis (Cirrhosis Value Set) during the measurement year or the year prior to the measurement year. Myalgia, myositis, myopathy, or rhabdomyolysis (Muscular Pain and Disease Value Set) during the measurement year. Description Estrogen agonists Clomiphene Prescription Note: An NDC list will be available on Administrative Specification: Rate 1 Received Statin Therapy Denominator Numerator The Rate 1 eligible population. The number of members who had at least one dispensing event for a statin of at least moderate dosage intensity (Table XXX) during the measurement year. Table XXX: High and Moderate-Intensity Statin Prescriptions Description Prescription High-intensity statin therapy Atorvastatin mg Rosuvastatin mg Moderate-intensity statin therapy Atorvastatin mg Rosuvastatin 5 10 mg Simvastatin mg Pravastatin mg Lovastatin 40 mg Fluvastatin XL 80 mg Fluvastatin 40 mg bid Pitavastatin 2 4 mg Eligible Population: Rate 2 Statin Adherence 80 Percent Product line Age Commercial, Medicaid, Medicare (report each product line separately). Males years as of December 31 of the measurement year. Females years as of December 31 of the measurement year. Continuous enrollment Allowable gap Anchor date Benefit The measurement year and the year prior to the measurement year. No more than one gap in enrollment of up to 45 days during each year of continuous enrollment. To determine continuous enrollment for a Medicaid beneficiary for whom enrollment is verified monthly, the member may not have more than a 1-month gap in coverage (i.e., a member whose coverage lapses for 2 months [60 days] is not considered continuously enrolled). December 31 of the measurement year. Medical during the measurement year and the year prior. Pharmacy during the measurement year. Event/Diagnosis All members who meet the numerator criteria for Rate 1.

56 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Administrative Specification: Rate 2 Statin Adherence 80 Percent Denominator Numerator The Rate 2 eligible population. The number of members who achieved a PDC of at least 80% during the treatment period. Follow the steps below to identify numerator compliance. Step 1 Step 2 Step 3 Step 4 Step 5 Identify the IPSD. The IPSD is the earliest dispensing event for any medication in Table XXX during the measurement year. To determine the treatment period, calculate the number of days from the IPSD (inclusive) to the end of the measurement year. Count the days covered by at least one prescription for statin medication during the treatment period. To ensure the measure does not give credit for supply that extends beyond the measurement year, subtract any days supply that extends beyond December 31 of the measurement year. Calculate the member s PDC using the following equation. Round (using the.5 rule) to two decimal places. Total Days Covered by a Statin Medication in the Treatment Period (step 3) Total Days in Treatment Period (step 2) Sum the number of members whose PDC is 80% for the treatment period.

57 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Statin Therapy for Patients With Cardiovascular Disease Measure Work-Up Measure Description The percentage of males years of age and females years of age during the measurement year who were identified as having clinical atherosclerotic cardiovascular disease (ASCVD) and were dispensed at least moderate-intensity statin therapy that they remained on for at least 80 percent of the treatment period. Two rates are reported: 1. Received Statin Therapy. The percentage of members who were identified as having clinical ASCVD and were dispensed at least moderate-intensity statin therapy during the measurement year. 2. Statin Adherence 80 Percent. The percentage of members who were identified as having clinical ASCVD and were dispensed at least moderate-intensity statin therapy that they remained on for at least 80 percent of the treatment period. Topic Overview Importance and Prevalence Cardiovascular disease is the leading cause of death in the United States. The death rate due to cardiovascular disease fell by 39 percent between 2001 and However, the public health burden remains significant. More than 85 million American adults have one or more types of cardiovascular disease (Mozaffarian et al., 2015). It is estimated that by 2030 more than 43 percent of Americans will have a form of cardiovascular disease (Heidenreich et al., 2011). National initiatives to improve cardiovascular health include the Million Hearts initiative to prevent 1 million heart attacks and strokes by 2017 (Million Hearts, 2011) and the American Heart Association (AHA) goal to reduce deaths from cardiovascular disease and stroke by 20 percent by 2020 (Mozaffarian et al., 2015). Data from the National Health and Nutrition Examination Survey (NHANES) estimate that more than 15 million American adults 20 and older have coronary heart disease. Coronary heart disease is more prevalent in men than in women (7.6 percent vs. 5.0 percent). Slight differences also exist based on race/ethnicity. The prevalence of coronary heart disease is highest in non-hispanic White men (7.8 percent) and lowest in non- Hispanic White women (4.6 percent). Just over 7 percent of non-hispanic Black men and 7 percent of non- Hispanic Black women have coronary heart disease. In the Hispanic population, 6.7 percent of men and nearly 6 percent of women have coronary heart disease (Mozaffarian et al., 2015). Data from the Framingham Heart Study estimate that the incidence of coronary heart disease is 10 years ahead in men (Thom, 2001). In addition, the incidence of cardiovascular events, such as myocardial infarction and sudden death, is 20 years ahead in men (Thom, 2001). In 2011, coronary heart disease was an underlying cause in 1 of 7 deaths in the United States. Coronary heart disease death rates per 100,000 were highest in males (161.5 for Black males and for White males). Deaths due to coronary heart disease per 100,000 were 99.7 for Black females and 80.1 for White females (CDC/NCHS, 2014). Atherosclerosis is a systemic disease process that occurs when plaque builds up within the walls of arteries. Plaque consists of fat, cholesterol, calcium, inflammatory cells and scar tissue that can harden overtime and narrow arteries. The narrowing of arteries reduces the flow of oxygen to organs and throughout the body, which results in most cardiovascular events, including heart attack and stroke (NHLBI, 2014). Coronary heart disease occurs when plaque builds up in arteries that supply oxygen to the heart (NHLBI, 2014). Chest pain or discomfort due to the reduced flow of oxygen rich blood to the heart is called angina pectoris. More than 8 million adults (3.3 percent) 20 and older have angina in the United States (Mozaffarian et al.,2015). The prevalence of angina is higher in women than in men between ages 40 and 74 (Ford, 2003). Plaque buildup can lead to peripheral arterial disease, which results when plaque builds up in arteries that

58 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, supply oxygen to the legs, arms and pelvis (NHLBI, 2014). Nearly 7 million adults 40 years of age and older have peripheral artery disease. The prevalence is higher in older adults, non-hispanic Blacks and women (Mozaffarian et al.,2015, Eraso, 2012 and Ostchega, 2007). A myocardial infarction (heart attack) occurs when oxygen rich blood is suddenly blocked from reaching the heart. More than 7 million adults 20 and older have had a myocardial infarction; the rate is twice as high in men than in women (Mozaffarian et al.,2015). Data show that 15 percent of people with myocardial infarction will die from it (Mozaffarian et al.,2015). Financial importance and cost-effectiveness In 2011, the total cost of cardiovascular disease and stroke in the United States was estimated to be $320 billion. This total includes direct costs such as the cost of physicians and other health professionals, hospital services, prescribed medications and home health care, as well as indirect costs due to loss of productivity from premature mortality. Interventions to address cardiovascular disease are increasing: since 2000, the number of inpatient cardiovascular operations and procedures increased by 28 percent, from 5,939,000 to 7,588,000 (Mozaffarian et al., 2015). By 2030, direct medical costs for cardiovascular disease are projected to increase to nearly $918 billion (Heidenreich, 2011). Evidence Supporting Statin Therapy Statins (HMG CoA reductase inhibitors) are a class of drugs that lower blood cholesterol. Statins work in the liver by preventing the formation of cholesterol, thus lowering the amount of cholesterol in the blood (AHA, 2014). Statins are most effective in lowering low-density lipoprotein cholesterol (LDL-C). The amount of cholesterol lowering effect is based on statin intensity, which is classified as either high, moderate or low intensity. Table 1. Statin Therapy Dosage Intensities High-Intensity Statin Therapy Moderate-Intensity Statin Therapy Low-Intensity Statin Therapy Daily dose lowers LDL C by approximately 50 percent on average Atorvastatin mg Rosuvastatin mg Daily dose lowers LDL C by approximately 30 percent to <50 percent on average Atorvastatin mg Rosuvastatin 5 10 mg Simvastatin mg Pravastatin mg Lovastatin 40 mg Fluvastatin XL 80 mg Fluvastatin 40 mg bid Pitavastatin 2 4 mg Daily dose lowers LDL C by <30 percent on average Simvastatin 10 mg Pravastatin mg Lovastatin 20 mg Fluvastatin mg Pitavastatin 1 mg Statins are among the most commonly prescribed medications in the United States, accumulating $17 billion in sales in 2012 (Consumer Reports, 2014). According to recent blood cholesterol treatment guidelines from the American College of Cardiology and American Heart Association (ACC/AHA), statins of moderate or high intensity are recommended for adults with established clinical ASCVD. Many studies support the use of statins to reduce ASCVD events in primary and secondary prevention. One meta-analysis of data from 170,000 patients in 26 randomized controlled trials found that intensive statin therapy reduces major vascular events by 15 percent (CTT, 2010). The study also found a 13 percent reduction in coronary death or nonfatal myocardial infarction, a 19 percent reduction in coronary revascularization and a 16 percent reduction in ischemic stroke (CTT, 2010). Another systematic review and meta-analysis estimates that long term statin therapy reduces the risk for ASCVD events by 25 percent 45 percent (Law, 2003).

59 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Safety considerations and contraindications Statin therapy is a first-line treatment for lowering blood cholesterol. While statins are considered safe for most patients, there are safety concerns to consider before prescribing and throughout treatment. Statins are contraindicated for women who are pregnant or breastfeeding and should not be used in women of childbearing potential, unless they are using effective forms of contraception (Stone et al., 2013). Evidence also shows that statin therapy should be avoided in patients with ESRD. Conclusions from a review of clinical trials, review articles and treatment guidelines found that statin therapy in ESRD patients fails to significantly alter the course of cardiovascular events (Nemerovski, 2013). The most common side effect of statin therapy is muscle pain or weakness, which can occur in varying forms of severity. However, the extent of muscle pain due to statin therapy is unclear (Thompson, 2003 and Parker et al., 2013). Statin therapy should not be used in patients with rhabdomyolysis, the most severe form of muscle symptoms (Stone et al., 2013). Clinicians can discontinue or adjust statin therapy in patients that develop mild to moderate muscle symptoms to assess other muscle related conditions and determine a tolerated statin intensity (Stone et al., 2013). Statins are cleared in the liver and can cause elevated liver biochemistries. This presents a concern for patients with existing liver disease. Research suggests that patients with decompensated cirrhosis and acute liver failure should not receive statin therapy due to the unlikely benefit of cholesterol lowering (Tandra, 2009). Statin adherence The ACC/AHA guidelines state adherence to both medication and lifestyle regimens are required for ASCVD risk reduction (Stone et al., 2013). This measure uses the proportion of days covered (PDC) to assess adherence. According to the Pharmacy Quality Alliance, a PDC threshold of 80 percent is considered highly adherent for most classes of chronic medications (Nau, 2012). The impact of adherence on statin efficacy has been shown: each 25 percent increase in statin adherence is associated with a ~3.8 mg/dl reduction in lowdensity lipoprotein cholesterol (Ho, 2009). Non-adherence to statin therapy can result in an increased risk for mortality. One study found a 12 percent 25 percent increase in the risk for mortality with non-adherence to statins after an acute myocardial infarction (Rasmussen, 2007). Research shows that adherence to statin medications is poor in the United States. In a randomized trial of medication coverage, 50 percent of patients in the control group (usual coverage) stopped using statin medications within one year of starting treatment (Choudhry, 2011). NCQA seeks to improve statin adherence in patients with cardiovascular disease and thereby reduce the risk for cardiovascular related mortality. Gaps in care A recent cohort study analyzed data from the National Cardiovascular Data Registry Practice Innovation and Clinical Excellence registry. The study identified more than 1 million patients that would benefit from statin therapy, according to the updated ACC/AHA guidelines. More than 91 percent of the patients studied had ASCVD. The study found that more than 32 percent of patients were not receiving statin therapy; more than 22 percent of patients were on non-statin therapies for cholesterol management (Maddox, 2014). NCQA s testing found similar results in a research database of commercial and Medicare Advantage health plans. NCQA reviewed statin dose intensities

60 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, and found that among patients with ASCVD, only 6 percent 9 percent were on high-intensity statins. We also found low adherence to statin therapy. Results highlight gaps in care for patients with cardiovascular disease and the need for improvement. Alignment with the new blood cholesterol guidelines will improve quality of care for patients with cardiovascular disease. Health care disparities Health disparities among genders exist when comparing the use of statins for secondary prevention of cardiovascular disease. One study found that although women with cardiovascular disease had higher LDL-C levels than men, they were less likely to receive any statin therapy (Virani et al., 2015). In another study, a meta-analysis found that among patients prescribed a statin medication, women were 10 percent more likely to be nonadherent. Non-White patients were 53 percent more likely to be nonadherent to statin therapy than White patients (Lewey, 2013). These gender-based and racially-based disparities signal gaps in quality that could relate to higher cardiovascular mortality rates in some groups, compared with mortality rates in White men. References American Heart Association (AHA) Drug therapy for cholesterol. rol/drug-therapy-for-cholesterol_ucm_305632_article.jsp. Accessed January 11, Centers for Disease Control and Prevention/National Center for Health Statistics (NCHS) Mortality multiple cause micro-data files, Public-use data file and documentation. NHLBI tabulations. Accessed July 3, Cholesterol Treatment Trialists (CTT) Collaboration Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170,000 participants in 26 randomised trials. Lancet 376(9753): doi: /s (10) Choudhry, N.K., J. Avorn, R.J. Glynn, et al Full coverage for preventive medications after myocardial infarction. New England Journal of Medicine. 365(22): Consumer Reports Are you taking the right treatment for your high cholesterol? March. Eraso, L.H., E. Fukaya, E.R. Mohler 3rd, et al Peripheral arterial disease, prevalence and cumulative risk factor profile analysis. Eur J Prev Cardiol. 21: Ford, E.S., W.H. Giles Changes in prevalence of nonfatal coronary heart disease in the United States from Ethn Dis. 13: Heidenreich, P.A., J.G. Trogdon, O.A. Khavjou, et al Forecasting the future of cardiovascular disease in the United States: a policy statement from the American Heart Association. Circulation.123: Ho, P.M., C.L. Bryson, J.S. Rumsfeld Medication adherence: its importance in cardiovascular outcomes. Circulation 119(23): Law, M.R., N.J. Wald, A.R. Rudnicka Quantifying effects of statins on low density lipoprotein cholesterol, ischaemic heart disease, and stroke: systematic review and meta-analysis. BMJ. 326(7404):1423. Lewey, J., W.H. Shrank, A.D. Bowry, et al Gender and racial disparities in adherence to statin therapy: a meta-analysis. American Heart Journal. 165(5): doi: /j.ahj Maddox, T.M., W.B. Borden, F. Tang, et al Implications of the 2013 ACC/AHA cholesterol guidelines for adults in contemporary cardiovascular practice: insights from the NCDR Pinnacle registry. Journal of American College of Cardiology. 64(21): doi: /j.acc Million Hearts The initiative. Accessed January Mozaffarian, D., E.J. Benjamin, A.S. Go, et al Heart disease and stroke statistics 2015 update: a report from the American Heart Association. Circulation. 131:e29-e322. doi: /CIR

61 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, National Heart, Lung, and Blood Institute (NHLBI) What is Atherosclerosis? Accessed January Nau, D.P Proportion of Days Covered (PDC) as a Preferred Method of Measuring Medication Adherence. Pharmacy Quality Alliance (PQA). Accessed November Nemerovski, C.W., J. Lekura, P.T. Mehta, C.L. Moore Safety and efficacy of statins in patients with end-stage renal disease. The Annals of Pharmacotherapy. 47(10): doi: / Ostchega, Y., R. Paulose-Ram, C.F. Dillon, Q. Gu, J.P. Hughes Prevalence of peripheral arterial disease and risk factors in persons aged 60 and older: data from the National Health and Nutrition Examination Survey J Am Geriatr Soc. 55: Parker, B.A., J.A. Capizzi, A.S. Grimaldi, et al Effect of statins on skeletal muscle function. Circulation. 127(1): doi /CIRCULATIONAHA Rasmussen, J.N., A. Chong, D.A. Alter Relationship between adherence to evidence-based pharmacotherapy and long-term mortality after acute myocardial infarction. Journal of the American Medical Association. 297(2): Stone, N.J., J. Robinson, A.H. Lichtenstein, et al ACC/AHA Guideline on the Treatment of Blood Cholesterol to Reduce Atherosclerotic Cardiovascular Risk in Adults. J Am Coll Cardiol. 63(25 Pt B): doi: /j.jacc Epub 2013 Nov 12. Tandra, S. and R. Vuppalanchi Use of statins in patients with liver disease. Current Treatment Options in Cardiovascular Medicine. 11(4): Thom, T.J., W.B. Kannel, H. Silbershatz, R.B. D Agostino Sr Cardiovascular disease in the United States and prevention approaches. In Hurst s the Heart, edited by V. Fuster, R.W. Alexander, R.A. O Rourke, R. Roberts, S.B. King 3rd, H.J.J. Wellens, th ed. New York, NY: McGraw-Hill. Thompson, P.D., P. Clarkson, R.H. Karas Statin-associated myopathy. Journal of the American Medical Association. 289(13): Virani, S.S., L.D. Woodard, D.J. Ramsey, et al Gender disparities in evidence-based statin therapy in patients with cardiovascular disease. American Journal of Cardiology. 115(1): doi: /j.amjcard

62 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Recommendations for Treatment of Blood Cholesterol to Reduce Atherosclerotic Cardiovascular Risk in Adults Statin Treatment Organization, Guideline Date Age Population Other risk factors Recommendation Type/ Grade years High-intensity statin therapy should be initiated or continued as firstline therapy, unless contraindicated. I A American College of Cardiology/American Heart Association (2013) years >75 years Clinical ASCVD If high-intensity statin therapy is contraindicated or when characteristics predisposing to statin-associated adverse effects are present, moderate-intensity statin should be used as the second option if tolerated It is reasonable to evaluate the potential for ASCVD risk-reduction benefits and for adverse effects, drug-drug interactions and to consider patient preferences, when initiating a moderate- or highintensity statin. It is reasonable to continue statin therapy in those who are tolerating it. I A IIa B

63 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Grading System Key: ACC/AHA Classification of Recommendation and Level of Evidence Size of Treatment Effect Class I Benefit >>> Risk Procedure/treatment should be performed/ administered Class IIa Benefit >> Risk Additional studies with focused objectives needed It is reasonable to perform procedure/administer treatment Class IIb Benefit Risk Procedure/treatment may be considered Class II No Benefit or Class III Harm Estimate of Certainty (Precision) of Treatment Effect Level A Multiple populations evaluated Level B Limited populations evaluated Level C Very limited populations evaluated Recommendation that procedure or treatment is useful/effective Sufficient evidence from multiple randomized trials or meta-analyses Recommendation the procedure or treatment is useful/effective Evidence from single randomized trial or nonrandomized studies Recommendation that procedure or treatment is useful/effective Only expert opinion, case studies, or standard of care Recommendation in favor of treatment or procedure being useful/effective Some conflicting evidence from multiple randomized trials or meta-analyses Recommendation in favor of treatment or procedure being useful/effective Some conflicting evidence from single randomized trial or nonrandomized studies Recommendation in favor of treatment or procedure being useful/effective Only diverging expert opinion, case studies, or standard of care Recommendation s usefulness/efficacy less well established Greater conflicting evidence from multiple randomized trials or meta-analyses Recommendation s usefulness/efficacy less well established Greater conflicting evidence from single randomized trial or nonrandomized studies Recommendation s usefulness/efficacy less well established Only diverging expert opinion, case studies, or standard of care Recommendation that procedure or treatment is not useful/effective and may be harmful Sufficient evidence from multiple randomized trials or meta-analyses Recommendation that procedure or treatment is not useful/effective and may be harmful Evidence from single randomized trial or nonrandomized studies Recommendation that procedure or treatment is not useful/effective and may be harmful Only expert opinion, case studies, or standard of care References for Recommendations Stone, N.J., J. Robinson, A.H. Lichtenstein, et al ACC/AHA Guideline on the Treatment of Blood Cholesterol to Reduce Atherosclerotic Cardiovascular Risk in Adults. J Am Coll Cardiol. 63(25 Pt B): doi: /j.jacc Epub 2013 Nov 12.

64 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Proposed New Measures for HEDIS : Statin Therapy for Patients With Diabetes NCQA seeks comments on the proposed new measure for inclusion in the HEDIS 2016 measurement set: Statin Therapy for Patients With Diabetes. The percentage of patients years of age with diabetes who do not have clinical atherosclerotic cardiovascular disease (ASCVD) and were dispensed a statin of any dosage intensity that they remained on for at least 80 percent of the treatment period. Two rates are reported: 1. Received Statin Therapy. The percentage of patients who were identified as having diabetes and were dispensed a statin of any dosage intensity during the measurement year. 2. Statin Adherence 80 Percent. The percentage of patients who were identified as having diabetes and were dispensed a statin of any dosage intensity that they remained on for at least 80 percent of the treatment period. This measure represents an important area for quality improvement in patients with diabetes by assessing the use of statin therapy and adherence for primary prevention of cardiovascular disease. The measure is based on 2013 blood cholesterol guidelines from the American College of Cardiology and the American Heart Association (ACC/AHA) and 2015 guidelines from the American Diabetes Association (ADA). 2,3 Convincing evidence estimates the benefit of statin therapy and adherence to reduce the risk for cardiovascular events: Every 25 percent increase in adherence to statin therapy results in a 3.8 mg/dl reduction in LDL-C levels. 4 Every 10 mg/dl reduction in LDL-C levels results in a 10 percent reduction in overall cardiovascular risk. 5 NCQA tested this measure in a large research database of commercially insured and Medicare Advantage individuals to assess importance, feasibility, validity and overall performance. We tested multiple aspects of the specification including denominator identification and age ranges, exclusions, statin dosage intensities, statin dispensing events and adherence to statin medications. Testing results revealed that the methods used to identify patients in the denominator are appropriate. NCQA s advisory panels agreed with the specified denominator identification methods. The prevalence of conditions that are contraindications for statin therapy was very low in the measure s denominator. Our advisory panels recommended pregnancy, end stage renal disease, cirrhosis and rhabdomyolysis as exclusions for this measure for face validity. The measure also proposes to exclude women trying to become pregnant and will use claims for clomiphene and In vitro fertilization to identify these patients. Our advisory panels debated excluding patients with claims for myalgia, myositis or myopathy. They recognized that although muscle pain or weakness is a common side effect of statin therapy, indicating statin intolerance, patients experiencing these issues should not necessarily stop receiving treatment; lower dosage intensity or alternative statins may be prescribed instead. However, there is currently no method for using 1 HEDIS is a registered trademark of the National Committee for Quality Assurance (NCQA). 2 Stone, N.J., J. Robinson, A.H. Lichtenstein, C.N. Bairey Merz, D.M. Lloyd-Jones, C.B. Blum, P. McBride, R.H. Eckel, J.S. Schwartz, A.C. Goldberg, S.T. Shero, D.G. Smith, Jr., D. Levy, K. Watson, P.W.F. Wilson ACC/AHA Guideline on the Treatment of Blood Cholesterol to Reduce Atherosclerotic Cardiovascular Risk in Adults. Journal of the American College of Cardiology doi: /j.jacc American Diabetes Association (ADA) Standards of Medical Care in Diabetes 2015: Cardiovascular disease and risk management. Diabetes Care 38(Suppl. 1): S49 S57. doi: /dc15-S011 4 Ho, P.M., C.L. Bryson, J.S. Rumsfeld Medication adherence: its importance in cardiovascular outcomes. Circulation 119(23): Rahilly-Tierney, C.R., E.V. Lawler, R.E. Scranton, J.M. Gaziano Cardiovascular benefit of magnitude of lowdensity lipoprotein cholesterol reduction. Circulation 120:

65 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, administrative data to accurately identify patients who experience intolerance to statins. With these considerations in mind, we seek comments specific to excluding patients with claims for myalgia, myositis or myopathy. Testing results also revealed low rates of patients with diabetes taking statins and poor rates of adherence to statin therapy. Although the guidelines for patients with diabetes recommend moderate or high-intensity statin therapy, our advisory panels strongly support measuring any dosage intensity statin therapy for primary prevention of cardiovascular disease in patients with diabetes. NCQA requests specific comment on statin dosage intensity for this measure. NCQA recommends an 80 percent adherence threshold to align with accepted standards for being highly adherent. We request comments on these issues: 1. Patients on statin therapy often experience muscle pain and weakness as symptoms of statin intolerance. However, there is currently no method for accurate identification of patients who experience intolerance to statins using administrative data. Although the codes for myalgia, myositis or myopathy are nonspecific, it is possible that claims for these conditions could serve as a proxy for statin intolerance. NCQA seeks comments on the following options for consideration: a. Exclude patients with claims for myalgia, myositis and myopathy. b. Exclude patients with claims for only myositis and myopathy. Do not exclude patients with claims for myalgia because it is the least severe of the conditions. c. Do not exclude patients with claims for myalgia, myositis or myopathy. 2. The ACC/AHA recommends moderate intensity statin therapy for patients with diabetes. However, our advisory panels recommended recognizing all intensities of statin therapy as numerator compliant for this measure because it is primary prevention of cardiovascular disease in a population that has potentially heterogeneous levels of risk. We request feedback on the proposed numerator, which will recognize patients who were dispensed a statin of any dosage intensity as numerator compliant. Supporting documents for the proposed measure include the draft measure specification and associated measure rationale work-up. NCQA acknowledges the contributions of the Diabetes Measurement Advisory Panel, the Technical Measurement Advisory Panel, the HEDIS Coding Panel and the Pharmacy Panel.

66 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Statin Therapy for Patients With Diabetes SUMMARY OF CHANGES TO HEDIS 2016 First-year measure. Description The percentage of members years of age during the measurement year with diabetes who do not have clinical atherosclerotic cardiovascular disease (ASCVD) who were dispensed a statin of any dosage intensity that they remained on for at least 80 percent of the treatment period. Two rates are reported: 1. Received Statin Therapy. The percentage of members who were identified as having diabetes and were dispensed a statin of any dosage intensity during the measurement year. 2. Statin Adherence 80 Percent. The percentage of members who were identified as having diabetes and were dispensed a statin of any dosage intensity that they remained on for at least 80% of the treatment period. Definitions IPSD Treatment period PDC Calculating number of days covered for multiple prescriptions Index prescription start date. The earliest prescription dispensing date for any statin medication of at least moderate intensity during the measurement year. The period of time beginning on the IPSD through the last day of the measurement year. Proportion of days covered. The number of days the member is covered by at least one statin medication prescription of appropriate intensity, divided by the number of days in the treatment period. If multiple prescriptions for different medications are dispensed on the same day, calculate number of days covered by a statin medication (for the numerator) using the prescriptions with the longest days supply. For multiple different prescriptions dispensed on different days with overlapping days supply, count each day within the treatment period only once toward the numerator. If multiple prescriptions for the same medication are dispensed on the same or different day, sum the days supply and use the total to calculate the number of days covered by a statin medication (for the numerator). For example, three prescriptions for the same medication are dispensed on the same day, each with a 30-day supply, sum the days supply for a total of 90 days covered by a statin. Subtract any days supply that extends beyond December 31 of the measurement year. Use the drug ID provided by the NDC to determine if the prescriptions are the same or different. Eligible Population: Rate 1 Received Statin Therapy Product lines Ages Commercial, Medicaid, Medicare (report each product line separately) years as of December 31 of the measurement year.

67 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Continuous enrollment Allowable gap Anchor date Benefit Event/ diagnosis Step 1 The measurement year and the year prior to the measurement year. No more than one gap in enrollment of up to 45 days during the measurement year. To determine continuous enrollment for a Medicaid beneficiary for whom enrollment is verified monthly, the member may not have more than a 1-month gap in coverage (i.e., a member whose coverage lapses for 2 months [60 days] is not considered continuously enrolled). December 31 of the measurement year. Medical during the measurement year and the year prior. Pharmacy during the measurement year. Follow the steps below to identify the eligible population. There are two ways to identify members with diabetes: by claim/encounter data and by pharmacy data. The organization must use both methods to identify the eligible population, but a member only needs to be identified by one method to be included in the measure. Members may be identified as having diabetes during the measurement year or the year prior to the measurement year. Claim/encounter data. Members who met any of the following criteria during the measurement year or the year prior to the measurement year (count services that occur over both years): At least two outpatient visits (Outpatient Value Set), observation visits (Observation Value Set), ED visits (ED Value Set) or non-acute inpatient encounters (Non-acute Inpatient Value Set) on different dates of service, with a diagnosis of diabetes (Diabetes Value Set). Visit type need not be the same for the two visits. At least one acute inpatient encounter (Acute Inpatient Value Set) with a diagnosis of diabetes (Diabetes Value Set). Pharmacy data. Members who were dispensed insulin or hypoglycemics/ antihyperglycemics on an ambulatory basis during the measurement year or the year prior to the measurement year (Table CDC-A).

68 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Table CDC-A: Prescriptions to Identify Members with Diabetes Description Prescription Alpha-glucosidase inhibitors Acarbose Miglitol Amylin analogs Pramlinitide Antidiabetic combinations Alogliptin-metformin Alogliptin-pioglitazone Glimepiride-pioglitazone Glimepiride-rosiglitazone Glipizide-metformin Glyburide-metformin Linagliptin-metformin Metformin-pioglitazone Metformin-repaglinide Metformin-rosiglitazone Insulin Insulin aspart Insulin aspart-insulin aspart protamine Insulin detemir Insulin glargine Insulin glulisine Insulin isophane human Insulin isophane-insulin regular Insulin lispro Insulin lispro-insulin lispro protamine Insulin regular human Metformin-saxagliptin Metformin-sitagliptin Sitagliptin-simvastatin Meglitinides Nateglinide Repaglinide Glucagon-like peptide-1 Exenatide Liraglutide Albiglutide (GLP1) agonists Sodium glucose Canagliflozin Dapagliflozin cotransporter 2 (SGLT2) inhibitor Sulfonylureas Chlorpropamide Glipizide Tolazamide Glimepiride Glyburide Tolbutamide Thiazolidinediones Pioglitazone Rosiglitazone Dipeptidyl peptidase-4 (DDP-4) inhibitors Alogliptin Linagliptin Saxagliptin Sitaglipin Step 2: Required exclusions Exclude members who meet any of the following criteria: Members with cardiovascular disease are identified in two ways: by event or by diagnosis. The organization must use both methods to identify the eligible population, but a member only needs to be identified by one method to be included in the measure. Event. Any of the following during the year prior to the measurement year meet criteria: MI. Discharged from an inpatient setting with an MI (MI Value Set). Use both facility and professional claims to identify MI. CABG. Discharged from an inpatient setting with a CABG (CABG Value Set). Use both facility and professional claims to identify CABG. PCI. Members who had PCI (PCI Value Set) in any setting. Other revascularization. Members who had any other revascularization procedure (Other Revascularization Value Set) in any setting. Diagnosis. Identify members as having ischemic vascular disease (IVD) who met at least one of the following criteria during both the measurement year and the year prior to the measurement year. Criteria need not be the same across both years. At least one outpatient visit (Outpatient Value Set) with an IVD diagnosis (IVD Value Set), or At least one acute inpatient encounter (Acute Inpatient Value Set) with an IVD diagnosis (IVD Value Set).

69 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Pregnancy (Pregnancy Value Set) during the measurement year or year prior to the measurement year. In vitro fertilization (IVF Value Set) in the measurement year or year prior to the measurement year. Dispensed at least one prescription for clomiphene (Table XXX-X) during the measurement year or the year prior to the measurement year. ESRD (ESRD Value Set) during the measurement year or the year prior to the measurement year. Cirrhosis (Cirrhosis Value Set) during the measurement year or the year prior to the measurement year. Myalgia, myositis, myopathy, or rhabdomyolysis (Muscular Pain and Disease Value Set) during the measurement year. Table XXX-X: Medications to Identify Exclusions Description Estrogen agonists Prescription Clomiphene Note: An NDC list will be available on Administrative Specification: Rate 1 Received Statin Therapy Denominator Numerator The Rate 1 eligible population. The number of members who had at least one dispensing event for a statin of any dosage intensity (Table XXX) during the measurement year. Table XXX: High, Moderate and Low-Intensity Statin Prescriptions Description Prescription High-intensity statin therapy Atorvastatin mg Rosuvastatin mg Moderate-intensity statin therapy Atorvastatin mg Rosuvastatin 5 10 mg Simvastatin mg Pravastatin mg Lovastatin 40 mg Fluvastatin XL 80 mg Fluvastatin 40 mg bid Pitavastatin 2 4 mg Low-intensity statin therapy Simvastatin 10 mg Pravastatin mg Lovastatin 20 mg Fluvastatin mg Pitavastatin 1 mg

70 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Eligible Population: Rate 2 Statin Adherence 80 Percent Product lines Ages Continuous enrollment Allowable gap Anchor date Benefit Commercial, Medicaid, Medicare (report each product line separately) years as of December 31 of the measurement year. The measurement year and the year prior to the measurement year. No more than one gap in enrollment of up to 45 days during the measurement year. To determine continuous enrollment for a Medicaid beneficiary for whom enrollment is verified monthly, the member may not have more than a 1-month gap in coverage (i.e., a member whose coverage lapses for 2 months [60 days] is not considered continuously enrolled). December 31 of the measurement year. Medical during the measurement year and the year prior. Pharmacy during the measurement year. Event/Diagnosis All members who meet the numerator criteria for Rate 1. Administrative Specification: Rate 2 Statin Adherence 80 Percent Denominator The Rate 2 eligible population. Numerator The number of members in the denominator who achieved a PDC of at least 80% during the treatment period. Follow the steps below to identify numerator compliance. Step 1 Step 2 Step 3 Step 4 Step 5 Identify the IPSD. The IPSD is the earliest dispensing event for any medication in Table XXX during the measurement year. To determine the treatment period, calculate the number of days from the IPSD (inclusive) to the end of the measurement year. Count the days covered by at least one prescription for statin medication during the treatment period. To ensure the measure does not give credit for supply that extends beyond the measurement year, subtract any days supply that extends beyond December 31 of the measurement year. Calculate the member s PDC using the following equation. Round (using the.5 rule) to two decimal places. Total Days Covered by a Statin Medication in the Treatment Period (step 3) Total Days in Treatment Period (step 2) Sum the number of members whose PDC is 80% for the treatment period.

71 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Exclusion (optional) Identify members who do not have a diagnosis of diabetes (Diabetes Value Set), in any setting, during the measurement year or year prior to the measurement year and who had a diagnosis of gestational diabetes or steroid-induced diabetes (Diabetes Exclusions Value Set), in any setting, during the measurement year or the year prior to the measurement year.

72 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Statin Therapy for Patients With Diabetes Measure Work-Up Measure Description The percentage of members years of age during the measurement year with diabetes who do not have clinical atherosclerotic cardiovascular disease (ASCVD) who were dispensed a statin of any dosage intensity that they remained on for at least 80 percent of the treatment period. Two rates are reported: 1. Received Statin Therapy. The percentage of members who were identified as having diabetes and were dispensed a statin of any dosage intensity during the measurement year. 2. Statin Adherence 80 Percent. The percentage of members who were identified as having diabetes and were dispensed a statin of any dosage intensity that they remained on for at least 80% of the treatment period. Topic Overview Importance and Prevalence Diabetes is a complex group of diseases marked by high blood sugar due to the body s inability to make or use insulin. Diabetes can lead to serious complications (CDC, 2014). Twenty nine million (9.3 percent) of Americans had diabetes in 2012 and 1.7 million adults were newly diagnosed with diabetes (ADA, 2014). Patients with diabetes have elevated cardiovascular risk, thought to be due in part to elevations in unhealthy cholesterol levels. Having unhealthy cholesterol levels places patients at a significant risk for developing atherosclerotic cardiovascular disease (ASCVD) (ADA, 2015). Primary prevention for cardiovascular disease is an important aspect of diabetes management. The risk of an adult with diabetes developing cardiovascular disease is two to four times higher than that of an adult without diabetes (AHA, 2012). In addition to being at a higher risk for developing cardiovascular disease, patients with diabetes tend to have worse survival after the onset of cardiovascular disease (Stone et al., 2013). The Centers for Disease Control and Prevention estimates that adults with diabetes are 1.7 times more likely to die from cardiovascular disease than adults without diabetes (CDC, 2014). Cardiovascular disease is the leading cause of death in the United States. Between 2001 and 2011, the death rate due to cardiovascular disease fell by 39 percent. However, the public health burden remains significant. More than 85 million American adults have one or more types of cardiovascular disease (Mozaffarian et al., 2015). It is estimated that, by 2030, more than 43 percent of Americans will have a form of cardiovascular disease (Heidenreich et al., 2011). By 2020, the American Heart Association (AHA) aims to improve cardiovascular health for all Americans by 20 percent, based on seven quantifiable metrics, and reduce deaths from cardiovascular disease and stroke by 20 percent (Mozaffarian et al., 2015). Financial importance and costeffectiveness The total cost of diabetes care in the United States was $245 billion in This is a 41 pecent increase from $175 billion in The cost of care to treat patients with diabetes includes direct costs ($176 billion) from office visits, hospital care and medications. Indirect costs to treat patients with diabetes are estimated to be $69 billion and includes costs for absenteeism, reducted productivity, unemployment due to disability and loss of productivity due to premature mortality. Research also shows that more than 1 in 10 dollars spent on health care in the United States are spent on the care of patients with diabetes and its complications. (ADA, 2013)

73 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Evidence Supporting Statin Therapy for Diabetes Numerous studies have demonstrated the efficacy of statins in reducing cardiovascular risk. The use of statins for primary prevention of cardiovascular disease in patients with diabetes, based on their age and other risk factors, is recommended by guidelines from the American Diabetes Association (ADA) and the American College of Cardiology/American Heart Association (ACC/AHA). The following sections include information on reducing cardiovascular risk for these patients, as well as gaps in care and disparities. Cardiovascular risk for patients with diabetes Statins (3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors) are a class of drugs that decrease low-density lipoprotein cholesterol (LDL-C) levels. Statins can decrease LDL-C levels by as much as 50% and could have additional benefit on high-density lipoprotein cholesterol (HDL-C) and triglyceride levels (Spratt, 2009). The amount of cholesterol lowering effect is based on statin intensity, which is classified as either high, moderate or low intensity (Table 1). Table 1. Statin Therapy Dosage Intensities High-Intensity Statin Therapy Moderate-Intensity Statin Therapy Low-Intensity Statin Therapy Daily dose lowers LDL C by approximately 50 percent on average Atorvastatin mg Rosuvastatin mg Daily dose lowers LDL C by approximately 30 percent to <50 percent on average Atorvastatin mg Rosuvastatin 5 10 mg Simvastatin mg Pravastatin mg Lovastatin 40 mg Fluvastatin XL 80 mg Fluvastatin 40 mg bid Pitavastatin 2 4 mg Daily dose lowers LDL C by <30 percent on average Simvastatin 10 mg Pravastatin mg Lovastatin 20 mg Fluvastatin mg Pitavastatin 1 mg Cholesterol lowering medications, such as statins, are among the most commonly prescribed drugs in America, accumulating $17 billion in sales in In the United States, 22 percent of adults (45 and older) take statins (Consumer Reports, 2014). Evidence shows statin use decreases cardiovascular mortality in patients with established cardiovascular disease, and total mortality rates. Primary and secondary prevention trial data strongly support starting lipid-lowering therapy with a statin in most patients with type 2 diabetes (Spratt, 2009). In a systematic review and meta-analysis of 12 studies conducted to evaluate the clinical benefit of lipid-lowering drug treatment in primary and secondary prevention, researchers found statins were equally effective in patients with and without diabetes (Costa et al, 2006). However, after adjusting for baseline risk, patients with diabetes had greater benefit in both the primary and secondary prevention of death due to coronary artery disease, nonfatal myocardial infarction, revascularization and stroke. Another meta-analysis by the American College of Physicians on lipid-lowering therapy for type 2 diabetes patients found a 22 percent reduction of cardiovascular events with primary prevention and a 24 percent reduction for secondary prevention (Spratt, 2009). Existing evidence cited by the ACC/AHA and ADA guidelines for statin therapy in patients with diabetes (but without ASCVD) focuses exclusively on the use of moderate intensity statin therapy versus placebo for primary prevention. These studies did not include low intensity statins. In contrast, cited studies for secondary prevention of cardiovascular disease in patients with diabetes examine moderate versus high intensity statins, and again did not include low intensity statins. There is

74 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, currently a lack of data comparing all statin intensities (low, moderate and high) to estimate differences in risk reduction for primary prevention in patients with diabetes. The guidelines clearly outline that the benefits of moderate or high intensity statins for patients with established cardiovascular disease outweigh potential harms and side effects. Conversely, patients with diabetes who do not have established cardiovascular disease have heterogeneous levels of risk. Because of this, there are concerns that statin use at only high or moderate intensity potentially exposes patients with diabetes to overly aggressive treatment. To ally these concerns, the measure includes the option of low intensity statins in recognition of providers weighing patient specific clinical considerations and enagement in shared decision making. Safety considerations and contraindicatio ns Statin therapy is a first-line treatment for lowering blood cholesterol. While statins are considered safe for most patients, there are safety concerns to consider before prescribing and throughout treatment. Statins are contraindicated for women who are pregnant or breastfeeding and should not be used in women of childbearing potential, unless they are using effective forms of contraception (Stone et al., 2013). Evidence also shows that statin therapy should be avoided in patients with ESRD. Conclusions from a review of clinical trials, review articles and treatment guidelines found that statin therapy in ESRD patients fails to significantly alter the course of cardiovascular events (Nemerovski, 2013). The most common side effect of statin therapy is muscle pain or weakness, which can occur in varying forms of severity. However, the extent of muscle pain due to statin therapy is unclear (Thompson, 2003 and Parker et al., 2013). Statin therapy should not be used in patients with rhabdomyolysis, the most severe form of muscle symptoms (Stone et al., 2013). Clinicians can discontinue or adjust statin therapy in patients that develop mild to moderate muscle symptoms to assess other muscle related conditions and determine a tolerated statin intensity. (Stone et al., 2013). Statins are cleared in the liver and can cause elevated liver biochemistries. This presents a concern for patients with existing liver disease. Research suggests that patients with decompensated cirrhosis and acute liver failure should not receive statin therapy due to the unlikely benefit of cholesterol lowering (Tandra, 2009). Statin adherence The ACC/AHA guidelines state, adherence to both medication and lifestyle regimens are required for ASCVD risk reduction (Stone et al., 2013). This measure uses the proportion of days covered (PDC) to assess adherence. According to the Pharmacy Quality Alliance, a PDC threshold of 80 percent is considered highly adherent for most classes of chronic medications (Nau, 2012). The impact of adherence on statin efficacy has been shown: each 25 percent increase in statin adherence is associated with a ~3.8 mg/dl reduction in low-density lipoprotein cholesterol (Ho, 2009). Nonadherence to statin therapy can result in an increased risk for morbidity and mortality. One study found a 12 percent 25 percent increase in the risk for mortality with non-adherence to statins after an acute myocardial infarction (Rasmussen, 2007). It can be extremely difficult to obtain high levels of adherence when prescribing medication as part of a primary prevention strategy, such as statins for lowering cholesterol levels to reduce cardiovascular risk (Mitka, 2010). Generally, long-term statin adherence for diabetes patients is poor, especially among those who have few other cardiovascular risk factors (Donnelly et al., 2008). Research shows that adherence to statin medications is poor in the United States. In a randomized trial of medication coverage, 50 percent of patients in the control group

75 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, (usual coverage) stopped using statin medications within one year of starting treatment (Choudhry, 2011). NCQA seeks to improve statin adherence in patients with diabetes and reduce cardiovascular risk and related mortality. Gaps in care Health care disparities NCQA tested this measure in a research database of commercial and Medicare Advantage health plans. In the diabetes population (40 75 years of age), 46 percent of commercial patients and 71 percent of Medicare patients were dispensed any statin. NCQA also tested statin adherence and found low adherence rates in the commercial population, with 25 percent achieving 80 percent adherence. A recent study on the primary prevention of cardiovascular disease found that minorities and elderly patients were less likely to be prescribed a statin (Fleetcroft, 2014). For patients with diabetes, research has demonstrated that fewer women than men tend to be dispensed a statin; this disparity puts women at greater risk for cardiovascular disease (Butalia, 2014). A study focused on veterans with diabetes found that female patients were less likely than males to receive lipid-lowering therapy, even though women tended to have higher LDL levels (Vimalananda, 2011). Disparities also exist in the adherence to statin therapy. One meta-analysis found that among patients prescribed a statin medication, women were 10 percent more likely to be nonadherent. Non-White patients were 53 percent more likely to be nonadherent to statin therapy than White patients (Lewey, 2013). References American Diabetes Association (ADA) Economic Costs of Diabetes in the U.S. in Diabetes Care. 36(4): doi: /dc ADA Statistics About Diabetes. Accessed January ADA Standards of Medical Care in Diabetes-2015: Cardiovascular disease and risk management. Diabetes Care. 38(Suppl. 1): S49 S57. doi: /dc15-S011 American Heart Association (AHA) Cardiovascular Disease & Diabetes. Last modified July 5. Diabetes_UCM_313865_Article.jsp Accessed January Butalia, S., A.M. Lewin, S.H. Simpson, et al Sex-based disparities in cardioprotective medication use in adults with diabetes. Diabetol Metab Syndr. 6(1):117. doi: / ecollection Centers for Disease Control and Prevention (CDC) National Diabetes Statistics Report: Estimates of Diabetes and Its Burden in the United States, Atlanta, GA: U.S. Department of Health and Human Services. Choudhry, N.K., J. Avorn, R.J. Glynn, et al Full coverage for preventive medications after myocardial infarction. New England Journal of Medicine. 365(22): Costa, J., M. Borges, C. David, A. Vaz Carneiro Efficacy of lipid lowering drug treatment for diabetic and non-diabetic patients: meta-analysis of randomised controlled trials. BMJ. 332(7550): Epub 2006 Apr 3. Consumer Reports Are you taking the right treatment for your high cholesterol? March. Donnelly, L.A., A.S.F. Doney, A.D. Morris, et al Long-Term Adherence to Statin Treatment in Diabetes. 25(7): Fleetcroft, R., P. Schofield, M. Ashworth Variations in statin-prescribing for primary cardiovascular disease prevention: cross-sectional analysis. BMC Health Services Research. 14:414. doi: / Heidenreich, P.A., J.G. Trogdon, O.A. Khavjou, et al Forecasting the future of cardiovascular disease in the United States: a policy statement from the American Heart Association. Circulation. 123(8): doi: /CIR.0b013e31820a55f5. Epub 2011 Jan 24. Ho, P.M., C.L. Bryson, J.S. Rumsfeld Medication adherence: its importance in cardiovascular outcomes. Circulation 119(23): Lewey, J., W.H. Shrank, A.D. Bowry, et al Gender and racial disparities in adherence to statin therapy: a meta-analysis. Am Heart J 165(5):665-78, 678.e1. doi: /j.ahj Epub 2013 Mar 26

76 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Mitka, M Improving Medication Adherence Promises Great Payback, but Poses Tough Challenge. JAMA 303(9): Mozaffarian, D., E.J. Benjamin, A.S. Go, et al Heart disease and stroke statistics 2015 update: a report from the American Heart Association. Circulation 131:e29-e322. doi: /CIR Nau, D.P Proportion of Days Covered (PDC) as a Preferred Method of Measuring Medication Adherence. Pharmacy Quality Alliance (PQA). Accessed November Nemerovski, C.W., J. Lekura, P.T. Mehta, C.L. Moore Safety and efficacy of statins in patients with end-stage renal disease. The Annals of Pharmacotherapy 47(10): doi: / Parker, B.A., J.A. Capizzi, A.S. Grimaldi, et al Effect of statins on skeletal muscle function. Circulation 127(1): doi /CIRCULATIONAHA Rasmussen, J.N., A. Chong, D.A. Alter Relationship between adherence to evidence-based pharmacotherapy and long-term mortality after acute myocardial infarction. Journal of the American Medical Association 297(2): Spratt, K.A Managing Diabetic Dyslipidemia: Aggressive Approach. J Am Osteopath Assoc 109(5 Suppl):S Stone, N.J., J. Robinson, A.H. Lichtenstein, et al ACC/AHA Guideline on the Treatment of Blood Cholesterol to Reduce Atherosclerotic Cardiovascular Risk in Adults. J Am Coll Cardiol. 63(25 Pt B): doi: /j.jacc Epub 2013 Nov 12. Tandra, S. and R. Vuppalanchi Use of statins in patients with liver disease. Current Treatment Options in Cardiovascular Medicine 11(4): Thompson, P.D., P. Clarkson, R.H. Karas Statin-associated myopathy. Journal of the American Medical Association 289(13): Vimalananda, V.G., D.R. Miller, M. Palnati, et al Gender Disparities in Lipid-Lowering Therapy Among Veterans with Diabetes. Womens Health Issues. 21(4):S176-S181. doi: /j.whi

77 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Recommendations for Statin Therapy for Patients with Diabetes Organization, Guideline, Date American College of Cardiology/ American Heart Association (2013) American Diabetes Association (2015) Age Population Other Risk Factors years Diabetes mellitus years <40 or >75 years Diabetes mellitus and a 7.5% estimated 10- year ASCVD risk Diabetes mellitus <40 years No CVD risk factors Diabetes CVD risk factor(s) 1 Diabetes Overt CVD 2 Diabetes years No CVD risk factors Diabetes CVD risk factor(s) Diabetes Overt CVD Diabetes >75 years No CVD risk factors Diabetes CVD risk factor(s) Diabetes Overt CVD Diabetes Recommendation Moderate-intensity statin therapy should be initiated or continued High-intensity statin therapy is reasonable unless contraindicated It is reasonable to evaluate the potential for ASCVD benefits and for adverse effects, for drug-drug interactions, and to consider patient preferences when deciding to initiate, continue, or intensify statin therapy. None Moderate or High High Moderate High High Moderate Moderate or high High Type/ Grade I A IIa B IIa C C A A B A B B A 1 CVD risk factors include LDL cholesterol $100 mg/dl (2.6 mmol/l), high blood pressure, smoking, and overweight and obesity. 2 Overt CVD includes those with previous cardiovascular events or acute coronary syndromes.

78 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Grading System Key: ACC/AHA Classification of Recommendation and Level of Evidence Size of Treatment Effect Class I Benefit >>> Risk Procedure/treatment should be performed/ administered Class IIa Benefit >> Risk Additional studies with focused objectives needed It is reasonable to perform procedure/administer treatment Class IIb Benefit Risk Procedure/treatment may be considered Class II No Benefit or Class III Harm Estimate of Certainty (Precision) of Treatment Effect Level A Multiple populations evaluated Level B Limited populations evaluated Level C Very limited populations evaluated Recommendation that procedure or treatment is useful/effective Sufficient evidence from multiple randomized trials or meta-analyses Recommendation the procedure or treatment is useful/effective Evidence from single randomized trial or nonrandomized studies Recommendation that procedure or treatment is useful/effective Only expert opinion, case studies, or standard of care Recommendation in favor of treatment or procedure being useful/effective Some conflicting evidence from multiple randomized trials or meta-analyses Recommendation in favor of treatment or procedure being useful/effective Some conflicting evidence from single randomized trial or nonrandomized studies Recommendation in favor of treatment or procedure being useful/effective Only diverging expert opinion, case studies, or standard of care Recommendation s usefulness/efficacy less well established Greater conflicting evidence from multiple randomized trials or meta-analyses Recommendation s usefulness/efficacy less well established Greater conflicting evidence from single randomized trial or nonrandomized studies Recommendation s usefulness/efficacy less well established Only diverging expert opinion, case studies, or standard of care Recommendation that procedure or treatment is not useful/effective and may be harmful Sufficient evidence from multiple randomized trials or meta-analyses Recommendation that procedure or treatment is not useful/effective and may be harmful Evidence from single randomized trial or nonrandomized studies Recommendation that procedure or treatment is not useful/effective and may be harmful Only expert opinion, case studies, or standard of care

79 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, American Diabetes Association Level of Evidence A B Description Clear evidence from well-conducted, generalizable randomized controlled trials that are adequately powered, including: Evidence from a well-conducted multicenter trial Evidence from a meta-analysis that incorporated quality ratings in the analysis Compelling nonexperimental evidence; i.e., all or none rule developed by the Centre for Evidence-Based Medicine at the University of Oxford Supportive evidence from well-conducted randomized controlled trials that are adequately powered, including: Evidence from a well-conducted trial at one or more institutions Evidence from a meta-analysis that incorporated quality ratings in the analysis Supportive evidence from well-conducted cohort studies: Evidence from a well-conducted prospective cohort study or registry Evidence from a well-conducted meta-analysis of cohort studies Supportive evidence from a well-conducted case-control study C E Supportive evidence from poorly controlled or uncontrolled studies: Evidence from randomized clinical trials with one or more major or three or more minor methodological flaws that could invalidate the results Evidence from observational studies with high potential for bias (such as case series with comparison with historical controls) Evidence from case series or case reports Conflicting evidence with the weight of evidence supporting the recommendation Expert consensus or clinical experience References for Recommendations Stone, N.J., J. Robinson, A.H. Lichtenstein, et al ACC/AHA Guideline on the Treatment of Blood Cholesterol to Reduce Atherosclerotic Cardiovascular Risk in Adults. J Am Coll Cardiol. 63(25 Pt B): doi: /j.jacc Epub 2013 Nov 12. American Diabetes Association (ADA) Standards of Medical Care in Diabetes-2015: Cardiovascular disease and risk management. Diabetes Care 38(Suppl. 1): S49 S57. doi: /dc15-S011

80 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Proposed New Measure for HEDIS : Hospitalization for Potentially Preventable Complications (HPC) NCQA seeks comments on the following proposed new measure for inclusion in the HEDIS 2016 measurement set: Hospitalization for Potentially Preventable Complications. This measure assesses, for members 67 years of age and older, the number of acute inpatient discharges during the measurement year with a diagnosis of an ambulatory care sensitive condition (ACSC) per 1,000 members. Data are reported in the following categories: Number of hospitalizations for acute ACSC. Number of hospitalizations for chronic ACSC. Total number of hospitalizations for all ACSC. Expected count of hospitalization for acute ACSC. Expected count of hospitalization for chronic ACSC. Expected count of hospitalization for all ACSC. ACSCs are acute or chronic health conditions that can be managed or treated in an outpatient setting. Research shows that for these conditions, appropriate and timely care can often prevent the development of complications requiring hospitalization. This measure is based on the NQF-endorsed Prevention Quality Indicators (PQI), developed by the Agency for Healthcare Research and Quality (AHRQ), and assesses the quality of ambulatory care to prevent complications of specific chronic and acute conditions that result in hospitalization. The measure will provide important information to health plans, providers, consumers and other stakeholders about how well a system of care helps prevent hospitalization for older adults with chronic and acute conditions. NCQA field-tested the measure in a large Medicare claims data set to determine if there is sufficient variation between health plans to demonstrate a quality gap and to assess the reliability of the measure (Table 1). Although the rate of observed hospitalization for chronic and acute complications was relatively low, the measure showed significant variation across health plans, suggesting a quality gap. Plans in the 90th percentile had hospitalization rates that were 3 5 times higher than plans in the 10th percentile. Results differed slightly for adults 85 and older, who had a higher average rate of hospitalization for acute conditions than for chronic conditions. Table 1. Rate of Hospitalization per 1,000 Beneficiaries, 2011 Medicare Advantage Data Age Average Min 10th 25th 50th 75th 90th Max Chronic Acute Chronic + Acute Chronic Acute Chronic + Acute Chronic Acute Chronic + Acute Note: Lower rates indicate better performance for this measure. 1 HEDIS is a registered trademark of the National Committee for Quality Assurance (NCQA).

81 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, To test the measure s reliability, we looked at plan performance across two time periods to see if there were similar rates over time (supporting the idea that the measure score represents a true signal) or if there was significant variation from year to year or between age strata (suggesting that the measure score represents random noise). Analysis demonstrated high rates of correlation across years for the and age groups and an acceptable rate of correlation for the 85+ age group, suggesting that the rates are reliable. NCQA s advisory panels concluded that results demonstrate measure reliability and opportunities for improvement. We request comments on the following: We are considering excluding members with long-term residence of 100 days or more in a nursing facility during the year. If the exclusion is required, what data sources could health plans use to determine whether their members reside in nursing homes? Supporting documents for the proposed measure include draft measure specifications and associated measure rationale work-up. NCQA acknowledges the contributions of the Geriatric Measurement Advisory Panel and the Technical Measurement Advisory Panel.

82 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Hospitalization for Potentially Preventable Complications (HPC) SUMMARY OF CHANGES TO HEDIS 2016 First-year measure. Description For members 67 years of age and older, the number of acute inpatient discharges during the measurement year with a diagnosis of an ambulatory care sensitive condition (ACSC) per 1,000 members. Data are reported in the following categories: Number of hospitalizations for acute ACSC. Number of hospitalizations for chronic ACSC. Total number of hospitalizations for all ACSC. Expected count of hospitalization for acute ACSC. Expected count of hospitalization for chronic ACSC. Expected count of hospitalization for all ACSC. Definitions ACSC Ambulatory care sensitive condition. An acute or chronic health condition that can be managed or treated in an outpatient setting. The ambulatory care conditions included in this measure are: Chronic ACSC Diabetes short-term complications. Diabetes long-term complications. Uncontrolled diabetes. Lower-extremity amputation among patients with diabetes. COPD. Asthma. Hypertension. Heart failure. Acute ACSC Bacterial pneumonia. Urinary tract infection. Cellulitis. Pressure ulcer. Classification period The year prior to the measurement year.

83 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Eligible Population Product lines Ages Continuous enrollment Allowable gap Anchor date Benefit Event/diagnosis Medicare. 67 years and older as of December 31 of the measurement year. The measurement year and the year prior to the measurement year. No more than one gap in enrollment of up to 45 days during each year of continuous enrollment. December 31 of the measurement year. Medical. None. Administrative Specification Report each rate separately and as a combined rate. The total rate is the sum of the two numerators. Denominator The eligible population. Numerators Chronic ACSC Step 1 Step 2 Step 3 Follow the steps below to identify the numerator. Identify all acute inpatient stays with a discharge date during the measurement year. Acute-to-acute transfers: Keep the original discharge and drop the transfer s discharge. For the remaining acute inpatient discharges, identify discharges with any of the following: Primary diagnosis for diabetes short-term complications (ketoacidosis, hyperosmolarity or coma; Diabetes Short Term Complications Value Set). Primary diagnosis for diabetes with long-term complications (renal, eye, neurological, circulatory or unspecified complications; Diabetes Long Term Complications Value Set). Primary diagnosis for uncontrolled diabetes (Uncontrolled Diabetes Value Set). Discharges with any procedure code for lower extremity amputation (Lower Extremity Amputation Procedures Value Set) and any diagnosis for diabetes (Diabetes Diagnosis Value Set). Exclude any discharge with a diagnosis for traumatic amputation of the lower extremity (Traumatic Amputation of Lower Extremity Value Set) or toe amputation procedure (Toe Amputation Value Set). Primary diagnosis of COPD (COPD Diagnosis Value Set), excluding any discharge with a diagnosis for cystic fibrosis and anomalies of the respiratory system (Cystic Fibrosis and Respiratory System Anomalies Value Set). Primary diagnosis for asthma (Asthma Diagnosis Value Set), excluding any discharge with a diagnosis for cystic fibrosis and anomalies of the respiratory system (Cystic Fibrosis and Respiratory System Anomalies Value Set).

84 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Primary diagnosis for acute bronchitis (Acute Bronchitis Diagnosis Value Set) and diagnosis for COPD (COPD Diagnosis Value Set). Exclude any discharge with a diagnosis for cystic fibrosis and anomalies of the respiratory system (Cystic Fibrosis and Respiratory System Anomalies Value Set). Primary diagnosis for heart failure (Heart Failure Diagnosis Value Set), excluding any discharges with a cardiac procedure (Cardiac Procedure Value Set). Primary diagnosis for hypertension (Hypertension Value Set), excluding any discharge with a cardiac procedure (Cardiac Procedure Value Set) or diagnosis of Stage I-IV kidney disease (Stage I-IV Kidney Disease Value Set) with a dialysis procedure (Dialysis Value Set). Note: For criteria that include multiple events, codes must be on the same claim. Acute ACSC Step 1 Step 2 Step 3 Follow the steps below to identify the numerator. Identify all acute inpatient stays with a discharge date during the measurement year. Acute-to-acute transfers: Keep the original discharge and drop the transfer discharge. For the remaining acute inpatient discharges, identify discharges with the any of the following: Primary diagnosis of bacterial pneumonia (Bacterial Pneumonia Value Set), excluding any discharge with a diagnosis of sickle cell anemia, HB-S disease (Sickle Cell Anemia and HB-S Disease Value Set) or procedure or diagnosis for immunocompromised state (Immunocompromised State Value Set). Primary diagnosis of urinary tract infection (Urinary Tract Infection Value Set), excluding any discharge with a diagnosis of kidney/urinary tract disorder (Kidney/Urinary Tract Disorder Value Set) or procedure or diagnosis for immunocompromised state (Immunocompromised State Value Set). Primary diagnosis of cellulitis (Cellulitis Value Set). Primary diagnosis of pressure ulcer (Pressure Ulcer Value Set). Note: For criteria that include multiple events, codes must be on the same claim. Total ACSC Count of inpatient stays with a discharge date during the measurement year for a chronic or acute ACSC. Sum the events from the Chronic ACSC and Acute ACSC numerators to obtain a total ACSC rate. Exclusions Residence in a skilled nursing facility for 100 days or more during the measurement year.

85 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Risk Adjustment Determination For each member in the eligible population, use the following steps to identify risk adjustment categories based on presence of comorbidity, age and gender. Comorbidities Step 1 Step 2 Identify all diagnoses for encounters during the classification period. Include the following when identifying encounters: Outpatient visits (Outpatient Value Set). Observation visits (Observation Value Set). Nonacute inpatient encounters (Nonacute Inpatient Value Set). Acute inpatient encounters (Acute Inpatient Value Set). ED visits (ED Value Set). Assign each diagnosis to one comorbid Clinical Condition (CC) category using Table CC Comorbid. Exclude all diagnoses that cannot be assigned to a comorbid CC category. For members with no qualifying diagnoses from face-to-face encounters, skip to the Risk Adjustment Weighting section. All digits must match exactly when mapping diagnosis codes to the comorbid CCs. Step 3 Determine HCCs for each comorbid CC identified. Refer to Table HCC Rank. For each member s comorbid CC list, match the comorbid CC code to the comorbid CC code in the table, and assign: The ranking group. The rank. The HCC. For comorbid CCs that do not match to Table HCC Rank, use the comorbid CC as the HCC and assign a rank of 1. Note: One comorbid CC can map to multiple HCCs; each HCC can have one or more comorbid CCs. Step 4 Example Step 5 Assess each ranking group separately and select only the highest ranked HCC in each ranking group using the Rank column (1 is the highest rank possible). Drop all other HCCs in each ranking group, and de-duplicate the HCC list if necessary. Refer to the Plan All-Cause Readmissions (PCR) measure for a Comorbid CC calculation example. Identify combination HCCs listed in Table HCC Comb. Some combinations suggest a greater amount of risk when observed together. For example, when diabetes and CHF are present, an increased amount of risk is evident. Additional HCCs are selected to account for these relationships. Compare each stay s list of unique HCCs to those in the HCC column in Table HCC Comb and assign any additional HCC conditions.

86 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, For fully nested combinations (e.g., the diabetes/chf combination is nested in the diabetes/chf/renal combination), use only the more comprehensive pattern. In this example, only the diabetes/chf/renal combination is counted. For overlapping combinations (e.g., the CHF/COPD combination overlaps the CHF/ renal/diabetes combination), use both sets of combinations. In this example, both CHF/COPD and CHF/renal/diabetes combinations are counted. Based on the combinations, a member can have none, one or more of these added HCCs. Example Refer to the PCR measure for a combination HCC calculation example. Risk Adjustment Weighting The calculation of risk-adjusted outcomes (counts of discharges) uses a two-step model. In the first step, logistic regression is used to estimate the probability of any discharge in the measurement year. In the second step, a Poisson regression model is used to predict the count of discharges among members who had at least one discharge in the measurement year. The results from each step are combined to predict how many discharges a member may have in the measurement year, given the member s age, gender and comorbidities. Separate risk adjustment weights will be provided for the logistic and Poisson regression models. For each member in the eligible population, use the following steps to identify risk adjustment weights for each type of ACSC based on presence of comorbidity, age and gender. Steps 1-5 are performed for the logistic regression model, steps 6-10 are performed for the Poisson regression model. Then using the results from each model (step 5 and step 10, respectively), proceed through step 11. Note: The final weights table will be released on November 2, Complete steps 1 5 for the logistic regression model. Step 1 Step 2 Step 3 Step 4 Step 5 For each member with a comorbidity HCC Category, link the weights. For Chronic ACSC: Use Table HPC-Chronic-ComorbHCC-Weights. For Acute ACSC: Use Table HPC-Acute-ComorbHCC-Weights. For Total ACSC: Use Table HPC-Total-ComorbHCC-Weights. Link the age and gender weights for each member. For Chronic ACSC: Use Table HPC-Chronic-ComorbHCC-OtherWeights. For Acute ACSC: Use Table HPC-Acute-ComorbHCC-OtherWeights. For Total ACSC: Use Table HPC-Total-ComorbHCC-OtherWeights. Identify the base risk weight. For Chronic ACSC: Use Table HPC-Chronic-ComorbHCC-OtherWeights. For Acute ACSC: Use Table HPC-Acute-ComorbHCC-OtherWeights. For Total ACSC: Use Table HPC-Total-ComorbHCC-OtherWeights. Sum all weights associated with the member (i.e., comorbidities, age, gender, base risk weight) for each numerator (Chronic ACSC, Acute ACSC, Total ACSC). Use the formula below to calculate the predicted probability of having at least one discharge in the measurement year, based on the sum of the weights for each member, for each numerator (Step 4).

87 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Calculate the predicted probability of each member having at least one discharge in the measurement year. This comes from the sum of weights from the logistic regression model using the formula below: Predicted probability of any discharge = Complete steps 6 10 for the Poisson regression model. ee ( Logistic Regression WeightsForEachMember) 1+ee ( Logistic Regression WeightsForEachMember) Step 6 Step 7 Step 8 Step 9 Step 10 For each member with a comorbidity HCC Category, link the weights. For Chronic ACSC: Use Table HPC-Chronic-ComorbHCC-Weights. For Acute ACSC: Use Table HPC-Acute-ComorbHCC-Weights. For Total ACSC: Use Table HPC-Total-ComorbHCC-Weights. Link the age and gender weights for each member. For Chronic ACSC: Use Table HPC-Chronic-ComorbHCC-OtherWeights. For Acute ACSC: Use Table HPC-Acute-ComorbHCC-OtherWeights. For Total ACSC: Use Table HPC-Total-ComorbHCC-OtherWeights. Identify the base risk weight. For Chronic ACSC: Use Table HPC-Chronic-ComorbHCC-OtherWeights. For Acute ACSC: Use Table HPC-Acute-ComorbHCC-OtherWeights. For Total ACSC: Use Table HPC-Total-ComorbHCC-OtherWeights. Sum all weights associated with the member (i.e., comorbidities, age, gender, base risk weight) for each numerator (Chronic ACSC, Acute ACSC, Total ACSC). Use the formula below to calculate the predicted unconditional count of discharges based on the sum of the weights for each member for each numerator (step 9). Calculate the predicted unconditional count of events. These predicted counts are not adjusted for the likelihood of having any discharges. ( Poisson Regression WeightsForEachMember) Predicted unconditional count of discharges = ee Complete step 11 for the logistic regression model and for the Poisson regression model. Step 11 Calculate the final member-level expected count of discharges using the formula below: Expected count of discharges = Predicted probability of any discharge Predicted unconditional count of discharges Reporting: Denominator Count the number of members in the eligible population for each age and gender group and the overall total. Enter these values into the reporting table (Table HPC-A-3). Reporting: Risk Adjustment Step 1 Step 2 For each numerator, calculate the sum of expected counts of discharges (step 11) across all members within each age and gender group and the overall total. Round to four decimal places using the.5 rule and enter these values into the reporting table. Reporting: Numerator

88 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Count the number of acute inpatient discharges for each numerator within the measurement year for each age and gender combination and enter these values into the reporting table. Note Organizations may not use risk assessment protocols to supplement diagnoses for calculating risk adjustment scores for this measure. The HPC measurement model was developed and tested using only claims-based diagnoses; diagnoses from additional data sources would affect the validity of the models as they are currently implemented in the specifications. Table HPC-A-3: Number of Members in the Eligible Population Age Sex Members Total Male Female Male Female Male Female Male Female Total: Total: Total: Total: Table HPC-B-3: Hospitalization for Potentially Preventable Complication Rates by Age and Risk Adjustment: Chronic Conditions Age Total Sex Chronic ACSC Discharges Chronic ACSC Discharges/1,000 Members Expected Count of Discharges O/E Ratio (Observed Discharges/ Expected Count) Male Female Total: Male Female Total: Male Female Total: Male Female Total:

89 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Table HPC-C-3: Hospitalization for Potentially Preventable Complication Rates by Age, Gender and Risk Adjustment: Acute Conditions Age Total Sex Acute ACSC Discharges Acute ACSC Discharges/1,000 Members Predicted Count of Discharges O/E Ratio (Observed Discharges/ Expected Count) Male Female Total: Male Female Total: Male Female Total: Male Female Total: Table HPC-D-3: Hospitalization for Potentially Preventable Complication Rates by Age, Gender and Risk Adjustment: Total Age Total Sex Total ACSC Discharges Total ACSC Discharges/1,000 Members Predicted Count of Discharges O/E Ratio (Observed Discharges/ Expected Count) Male Female Total: Male Female Total: Male Female Total: Male Female Total:

90 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Hospitalization for Potentially Preventable Complications (HPC) Measure Work-Up Measure Description For members 67 years of age and older, the number of acute inpatient discharges during the measurement year with a diagnosis of an ambulatory care sensitive condition (ACSC) per 1,000 members. Data are reported in the following categories: Number of hospitalizations for acute ACSC. Number of hospitalizations for chronic ACSC. Total number of hospitalizations for all ACSC. Expected count of hospitalization for acute ACSC. Expected count of hospitalization for chronic ACSC. Expected count of hospitalization for all ACSC. Topic Overview Importance and Prevalence In 2011, approximately 3 out of every 10 Medicare beneficiaries were admitted to the hospital (DHHS 2013). A number of studies have suggested that certain hospitalizations can be prevented by optimal outpatient care. These conditions, called ambulatory care sensitive conditions (ACSC) and the evidence behind the conditions, is described in greater detail below. Financial importance and costeffectiveness Hospital and inpatient care is the largest component of total health care costs for older adults (26 percent of Medicare spending, approximately $140 billion dollars annually) (KFF 2012). Hospitalization also poses risks for older adults, who frequently develop serious conditions as a result of hospitalization (e.g., delirium, infection, decline in functional ability) (Gillick 1982; Covinsky 2011). Reducing the rate of hospitalization for older adults will improve patient health, reduce costs and improve quality of life. Supporting Evidence Development of ambulatory care sensitive conditions ACSC were designed to evaluate the potential impact of differences in socioeconomic status and resources on hospitalization rates. An early study by Billings et al. (1993) aimed to improve the understanding of the causes of variation in hospital use and evaluate the effectiveness of programs designed to improve access to care. His team used a modified Delphi approach to define three basic categories for grouping all causes of hospital admission: 1. Conditions for which the provision of timely and effective outpatient care is likely to have little impact on the need for hospital admission. 2. Conditions for which timely and effective outpatient care can help to reduce the risks of hospitalization by either preventing the onset of an illness or condition, controlling an acute episodic illness or condition or managing a chronic disease or condition.

91 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Referral-sensitive surgeries, defined as high-cost/high-technology surgical procedures for which impediments to access or referral to specialty care may reduce the chances of having the surgery. Their analysis then focused on the rates of hospitalization among adults under 65 for conditions identified as being potentially responsive to timely and effective outpatient management. They found adults in low-income areas had substantially higher admission rates for ACSC than in high-income areas. The authors suggested that adults in low income areas are more likely to be affected by access problems, given higher rates of the uninsured and less experience in navigating the complexities of the fragmented health care delivery system. This lack of adequate access to ambulatory care and potentially low performance of outpatient care delivery systems was partially responsible for the higher rates of hospitalization for ACSC. Other factors contributing to hospitalization included disease prevalence, patient lifestyle (alcohol/substance abuse), and possible differences in physician decision making (Billings 1993). Since this early study, many more studies have examined the effect of income, insurance and access on ACSC hospitalization and many more diagnoses have been classified in various research studies as potentially ACSC hospitalizations. Across studies, the list of potentially ACSC now includes over 100 conditions. Research on ACSC in Medicare populations We identified two studies that looked at hospitalization for ACSC in the Medicare population. In 2001, McCall et al. evaluated the feasibility of measuring hospitalization for ACSC for Medicare + Choice (MC) programs (now called Medicare Advantage). The authors suggested that information about ACSC hospitalization could be used by health plans to evaluate their providers processes of care and to develop case management strategies to reduce rates of ACSC hospitalizations. ACSC can also be used as identifying events to improve the adequacy of primary care for potentially vulnerable populations. To meet this focus, Health Care Financing Administration (HCFA) had each condition reviewed by two clinical consultants to ensure that selected ACSCs were appropriate for the elderly population. ACSC selected for the study included chronic (asthma/chronic obstructive pulmonary disease (COPD), congestive heart failure, seizure disorder, diabetes, and hypertension); acute (hypoglycemia, urinary tract infections, cellulitis, dehydration, hypokalemia, gastric and duodenal ulcer, bacterial pneumonia, and severe ear/nose/throat infections); and preventable (influenza and malnutrition). The study found that the oldest-old (age 85 and older) experience statistically significant higher rates of ACSC admissions and are more likely to die during an ACSC admission than younger Medicare beneficiaries. The study also found lower overall rates of hospitalization in the MC population than in the Medicare FFS population. On average, MC adjusted hospitalization rates were about one-third lower than comparable FFS rates. The rates of hospital discharges for MC in comparison to FFS may be explained by better management of patient conditions, utilization controls or healthier MA enrollees. Ultimately, the study found the measurement of ACSC in the MC population to be feasible; however, they suggested further studies may want to limit the scope of conditions to the most frequently occurring in the Medicare population (e.g., congestive heart failure, pneumonia and asthma/copd), as many of the other conditions evaluated in this study did not occur with enough frequency to produce statistically reliable estimates at the MA organization level for the majority of MA organizations (McCall 2001).

92 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, In 2004, CMS contracted with RTI to explore trends in hospitalization for ACSC among a sample of Medicare FFS beneficiaries. Specifically, they were interested in a set of ACSC that had previously been shown to be increasing in the Medicare population. Using a comprehensive literature review and expert review process they narrowed a list of 48 possible acute and chronic ACSC down to eleven conditions selected for inclusion in the analysis: (1) cellulitis, (2) asthma, (3) chronic obstructive pulmonary disease (COPD), (4) congestive heart failure (CHF), (5) dehydration, (6) pneumonia, (7) septicemia, (8) stroke, (9) urinary tract infection (UTI), (10) acute diabetic events and (11) lower limb peripheral vascular disease (PVD). During the study period ( ), all cause hospitalization rates increased by 6 percent in the Medicare FFS population. Multivariate modeling of the trend in ACSC hospitalizations from showed that changes in sociodemographic characteristics and health status among elderly Medicare FFS beneficiaries explained a substantial proportion of the observed positive trend in ACSC hospitalization rates for CHF, COPD and PVD among Medicare beneficiaries. Poverty appeared to have the strongest relationship with rate of ACSC hospitalization (McCall 2004). The authors concluded that, since rates of ACSC hospitalization are strongly influenced by the health status of the Medicare population, interventions employed to reduce hospitalization for ACSCs may have to be tailored to the specific underlying condition to be effective. Prior-year hospitalization for the ACSCs appeared to be a strong proxy for severity of disease; the authors suggested that targeting hospitalized Medicare beneficiaries or those that have been hospitalized in the prior year for disease management programs may be a reasonable strategy to reduce future hospitalizations. The authors concluded that further exploration is necessary to understand the factors that contribute to ACSC hospitalization that may be beyond the control of the health plan or provider such as population aging or onset of new co-morbid conditions (McCall 2004). Development of prevention quality indicators In 2001, the Agency for Healthcare Research and Quality s (AHRQ) Evidence-Based Practice Center (EPC) at the University of California San Francisco (UCSF) and Stanford University developed the Prevention Quality Indicators (PQI) based on the original Healthcare Cost and Utilization Project (HCUP) Quality Indicators developed in the early 1990s (Davies 2001). They reviewed the evidence on ACSC to date and used a multi-stakeholder review process using three questions to assess face validity of the indicators: 1. Have clinical trials demonstrated that specific outpatient therapies can reduce the risk of hospitalization? 2. Have observational studies shown associations between specific outpatient therapies and the risk of hospitalization? 3. Is there general consensus that hospitalizations for these conditions are often avoidable or preventable, if the patient has timely access to high-quality outpatient care? They selected 16 ACSCs to be used as area-level quality indicators (dehydration, bacterial pneumonia, urinary tract infection, perforated appendix, angina, asthma, COPD, CHF, diabetes short term complications, uncontrolled diabetes, diabetes long term complications, lower extremity amputation in diabetics, hypertension, low birth weight, pediatric asthma and pediatric gastroenteritis). In general, the AHRQ, UCSF and Stanford research team ( AHRQ team ) found little published evidence for individual indicators, presumably due to the common usage of indicators within sets. Most studies have examined sets of ACSC conditions, without providing data stratified by indicator. In general across studies the AHRQ team found condition prevalence, race and

93 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, socioeconomic status were independent predictors of the rate of hospitalization for ACSC in the general population. At the individual condition level, self-reported health status, functional limitations, several chronic diseases and a chronic disease risk score are associated with preventable hospitalizations among Medicare beneficiaries. Income was found to be a much less powerful predictor of hospitalization for chronic ACSC among Medicare beneficiaries after adjusting for health factors (Davies 2001). Although many studies have been published about the association between access to care and ACSC hospitalization, AHRQ found few studies that tested true measures of access to care, as opposed to socioeconomic status; one of these which used survey day from the Medicare Current Beneficiaries Survey (MCBS) is described above (McCall 2004). Bindman and colleagues found that patient reported difficulty in receiving medical care when needed explained 50 percent of the variability in hospitalization rates for 5 chronic medical conditions. Having a regular source of care and a higher primary care physician/population ratio were also independently associated with avoidable hospitalization rates (Bindman 1995). Other studies have shown that the physician-to-population ratio for family and general physicians is associated with avoidable hospitalization. Beneficiaries in fair or poor health are at increased risk if they live in a primary-care shortage area. Relationships between access indicators (e.g., patient-reported access, regular source of care, primary care physician-to-population ratio) and hospitalization for ACSC did not hold in two separate studies of rural zip codes, suggesting that avoidable hospitalization rates are invalid indicators of access in rural areas. Almost all chronic ACSCs and several of the acute conditions have practice guidelines. Studies have shown that access to ambulatory care and adherence to evidence-based treatment guidelines can reduce patient complication rates of existing disease, many of which result in hospitalization. Since many of the hospital admission rates for ACSC are correlated, it is likely that a common underlying factor influences rates (AHRQ 2007). For specific evidence on each PQI, refer to the AHRQ Guide to Prevention Quality Indicators: e_v31.pdf Expanding PQI use for performance measures Recently, AHRQ convened a multi-stakeholder panel of experts to review the evidence for all AHRQ PQI and assess the appropriateness of using PQI for quality improvement, public reporting and pay-for-performance (Davies 2009). The group used a Delphi and Nominal Panel method to solicit feedback from panel members on the face validity of the PQI for different settings and uses. Overall, panelists rated most indicators as appropriate for many settings and use. The lowestrated indicators were perforated appendix, dehydration, bacterial pneumonia, UTI and angina. Panel members had major concerns regarding use for these measures in either pay-for-performance or comparative reporting at the payer level. The following text summarizes the qualitative recommendations of the panelists regarding each of the conditions and pathways for payers and providers to influence hospitalization (Davies 2009). Diabetes-related indicators: Payer and provider organizations may be able to reduce hospitalization for diabetes by enhancing coverage for medication supplied for blood glucose monitoring and care coordination for diabetes patients. Ongoing patient education and promotion of self-management might also reduce rates of hospitalization for diabetes. Perforated appendix: Panelists did not feel this indicator was necessarily reflective of high quality outpatient care since most appendicitis patients present directly to the emergency room. The panelists felt that time to presentation, which is the highest predictor of appendicitis, was not in the health system s

94 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, control. They also expressed concerns that older adults tend to present atypical symptoms of appendicitis and therefore may be more difficult to diagnosis. COPD and asthma: Panelists cited increased reimbursement for smoking cessation programs, medication, access to pulmonary rehabilitation and oxygen therapy as affecting rates of hospitalization. Additionally, patient education and improved care coordination could reduce rates of hospitalization for COPD and asthma. Panelists expressed concern that this rate may reflect some level of social hospitalization for situations where the provider feels the support in the home environment is insufficient for recovery. Hypertension: Payer and provider organizations may be able to reduce hypertension-related hospitalizations through enhanced coverage of preventive primary care visits, patient education and anti-hypertensive medication. Improved rates of blood pressure screening may also reduce rates of hospitalization. Congestive heart failure: Panelists cited enhanced coverage of medications, access to primary care and patient education as the main mechanisms for mitigating hospitalization, in addition to outreach to at-risk patients through teleconferencing and home visits. Dehydration: In general, panelists expressed concern about the state of evidence linking payer and provider organization intervention to reduction of admission for dehydration. They cited that many older adults do not present in a timely manner to the outpatient setting and patients are rarely sent home from ambulatory care with hypovolemia. Bacterial pneumonia: Panelists agreed that payers could influence hospitalization for bacterial pneumonia by ensuring access to immunizations and antibiotics, but were uncertain about the degree to which increased access could reduce hospitalization in high-risk populations. UTI: Some panelists expressed concern about the lack of evidence directly linking care in the outpatient setting to hospitalization for UTI. Others suggested that enhanced coverage of antibiotics and careful attention to inappropriate use of Foley/suprapubic catheters could impact rates of hospitalization. Angina without cardiac procedure: Panelists were divided with regard to the degree to which payers and providers could influence hospitalization for angina. Panelists expressed concern that many individuals with angina are directed to the emergency room, where thresholds for admission for chest pain are low due to the fear of possible legal action. Lower extremity amputation: Minor problems in the lower extremities can be treated in outpatient care, limiting progression of the disease. Payer organizations may be able to enhance coverage of medication, supplies for diabetes self-management and promote care coordination. There was a concern that patient factors such as diet, income and geographic limitations may limit the control the health care system has on admission rates. Health care disparity and gaps in care ACSC admission rates are higher in the U.S. among low-income persons, African- Americans, Hispanics, Medicaid beneficiaries and the uninsured (Gaskin and Hoffman 2000; O Neil 2010; Chang 2009; Vest 2010; Johnson 2012). The National Health Care Disparities report examined rates of potentially avoidable hospitalization using the AHRQ PQI measures for specific race/ethnic groups and area income quartile. The report found the highest rate of PQI hospitalization for Black adults and the lowest rates for Asian/Pacific Islander Adults. Geographic areas with the lowest income had the highest rates of PQI hospitalizations and areas with the highest income had the lowest rates of potentially avoidable hospitalization (AHRQ 2013). It is important to note that

95 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, none of these studies reflect the rapidly changing landscape of coverage expansion envisioned by the Affordable Care Act. Benefits envisioned by use of this measure The composites were designed to assess quality of care for ambulatory care sensitive conditions. Not all complications that result in hospitalization are preventable; therefore, the goal of this measure is not to obtain a rate of zero hospitalizations, but to illuminate the success of the community health care system at managing chronic disease and outpatient care for acute conditions. These composites can act as indicators to help flag potential health care quality problems in both chronic disease management and acute care. The conditions covered in the composites align with the AHRQ PQI measures and use the same value sets. Summary of controversy / contradictory evidence The complexity of the relationship between socioeconomic status and PQI rates makes it difficult to delineate how much of the observed relationships are due to access-tocare difficulties in potentially underserved populations and how much is due to other patient characteristics that are unrelated to quality of care and vary by socioeconomic status. For some indicators, patient preferences and hospital capabilities for inpatient or outpatient care might explain variations in hospitalization. In addition, environmental conditions (such as poor air quality) that are not under the direct control of the health care system can substantially influence some of the PQIs (AHRQ 2007). The evidence related to potentially avoidable hospital admissions is limited for each indicator because many of the indicators have been developed and subsequently evaluated as integral components of sets. Relatively little is known about which components represent the strongest measures of access and quality (AHRQ 2007). Specific Guideline Recommendations It is NCQA policy to use guidelines that are evidence-based, applicable to physicians and other health care providers and developed by a national specialty organization or government agency. This measure of utilization is not based on clinical guidelines; however, the appropriate treatment plans for these conditions are supported by clinical guidelines. Results should be used in conjunction with other use-of-service data and clinical quality-of-care data to evaluate trends in patient care settings. Year-to-year trends and risk-adjusted comparison among plans should be evaluated, rather than a cross-sectional, observed rate of hospitalization. References AHRQ Guide to Prevention Quality Indicators. Agency for Healthcare Research and Quality, Rockville, MD. Accessed July 31, Available at: AHRQ National Health Care Disparities Report. Rockville, MD. Accessed September 1, Available at: Billings, J., L. Zeitel, J. Lukomnik, T.S. Carey, A.E. Blank, & L. Newman Impact of socioeconomic status on hospital use in New York City. Health Affairs, 12(1): Bindman, A.B., K. Grumbach, D. Osmond, et al Preventable hospitalizations and access to health care. The Journal of the American Medical Association, 274(4): Chang, C. F., and R.A. Pope Potentially avoidable hospitalizations in Tennessee: analysis of prevalence disparities associated with gender, race, and insurance. Public Health Reports, 124(1):127. Covinsky, K.E., E. Pierluissi, & C.B. Johnston Hospitalization-associated disability. The Journal of the American Medical Association, 306(16): Davies, S.M., J. Geppert, M. McClellan, et al Refinement of the HCUP Quality Indicators. Rockville (MD): Agency for Healthcare Research and Quality (US); 2001 May. (Technical Reviews, No. 4.) Accessed July 31, Available from:

96 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Davies, S.M., K.M. McDonald, E. Schmidt, E. Schultz, J. Geppert, & P.S. Romano Expanding Use of the Prevention Quality Indicators: Report of Clinical Expert Review Panel. Report prepared for AHRQ. Accessed July 31, Available at: 20Summary% 20Report.pdf Department of Health and Human Services (DHHS) Health Indicators Warehouse: CMS Report by Indicator Utilization Report. Accessed July 31, Available at: Gaskin, D.J., and C.Hoffman Racial and Ethnic Differences in Preventable Hospitalizations across 10 States. Medical Care Research and Review, 57(Suppl 1): Gillick, M. R., N.A. Serrell, & L.S. Gillick Adverse consequences of hospitalization in the elderly. Social science & medicine, 16(10): Johnson, P. J., N. Ghildayal, A.C. Ward, B.C. Westgard, L.L. Boland, & J.S. Hokanson Disparities in potentially avoidable emergency department (ED) care: ED visits for ambulatory care sensitive conditions. Medical Care, 50(12): Kaiser Family Foundation (KFF) Medicare Spending and Financing Fact Sheet. Accessed July 31, Available at: McCall, N.T., E. Brody, L. Mobley, & S. Subramanian Investigation of Increasing Rates of Hospitalization for Ambulatory Care Sensitive Conditions Among Medicare Fee-for-Service Beneficiaries. Prepared for the Centers for Medicare and Medicaid Services. Accessed July 31, Available at: Reports/Reports/downloads/McCall_2004_3.pdf McCall, N.T., J. Harlow, & D. Dayhoff Rates of Hospitalization for Ambulatory Care Sensitive Conditions in the Medicare+Choice Population. Health Care Financing Review, 22(3): O'Neil, S.S., T. Lake, A. Merrill, A. Wilson, D.A. Mann, & L.M. Bartnyska Racial Disparities in Hospitalizations for Ambulatory Care Sensitive Conditions. American journal of preventive medicine, 38(4): Vest, J. R., L.D. Gamm, B.A. Oxford, M.I. Gonzalez, & K.M. Slawson Determinants of preventable readmissions in the United States: a systematic review. Implementation Science, 5(1):88.

97 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Proposed Retirement for HEDIS : Use of Appropriate Medications for People With Asthma (ASM) Proposed Changes to Existing Measures for HEDIS 2016: Medication Management for People With Asthma (MMA) Asthma Medication Ratio (AMR) NCQA seeks comments on proposed changes to the HEDIS Asthma measure set: Retire Use of Appropriate Medications for People With Asthma. Expand the upper age limit for Medication Management for People With Asthma and Asthma Medication Ratio to include health plan members up to 85 years of age; add Medicare as a reporting product line. The proposed retirement of Use of Appropriate Medications for People With Asthma measure results from consistently high HEDIS performance rates and little variation in plan performance for both commercial and Medicaid plans. Additionally, Medication Management for People With Asthma is a more effective way of assessing asthma medication management. The Medication Management for People With Asthma and Asthma Medication Ratio measures evaluate the effectiveness of asthma management in members 5 64 years of age. Stakeholders expressed interest in expanding the measures to include older adults. NCQA tested the feasibility of expanding the eligible population to ages 65 and older and whether it would provide meaningful and valid HEDIS performance information: specifically, because the frequency of clinical measure exclusions increases with age, would health plans have a sufficient denominator? Based on test findings, a sufficient number of adults 65 and older remain after applying exclusions to the older commercial and Medicare populations, which supports including these adults in the HEDIS Asthma measure set. Supporting documents for this measure include the current measure specifications and measure work-up. NCQA acknowledges the contributions of the Respiratory Measurement Advisory Panel and the Geriatric Measurement Advisory Panel. 1 HEDIS is a registered trademark of the National Committee for Quality Assurance (NCQA).

98 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Asthma Medication Ratio (AMR) SUMMARY OF CHANGES TO HEDIS 2016 Expanded age range up to 85 years Added the Medicare product line Description The percentage of members years of age who were identified as having persistent asthma and had a ratio of controller medications to total asthma medications of 0.50 or greater during the measurement year. Definitions Oral medication dispensing event One prescription of an amount lasting 30 days or less. To calculate dispensing events for prescriptions longer than 30 days, divide the day s supply by 30 and round down to convert. For example, a 100-day prescription is equal to three dispensing events (100/30 = 3.33, rounded down to 3). The organization should allocate the dispensing events to the appropriate year based on the date on which the prescription is filled. Multiple prescriptions for different medications dispensed on the same day are counted as separate dispensing events. If multiple prescriptions for the same medication are dispensed on the same day, sum the day s supply and divide by 30. Use the Drug ID to determine if the prescriptions are the same or different. Refer to the definition of Oral medication dispensing event in ASM for examples. Inhaler dispensing event When identifying the eligible population, use the definition below to count inhaler dispensing events. All inhalers (i.e., canisters) of the same medication dispensed on the same day count as one dispensing event. Medications with different Drug IDs dispensed on the same day are counted as different dispensing events. For example, if a member received three canisters of Medication A and two canisters of Medication B on the same date, it would count as two dispensing events. Allocate the dispensing events to the appropriate year based on the date when the prescription was filled. Use the Drug ID field in the NDC list to determine if the medications are the same or different. Injection dispensing event Units of medication Injections count as one dispensing event. Multiple dispensing events of the same or different medication count as separate dispensing events. Allocate the dispensing events to the appropriate year based on the date when the prescription was filled. When identifying medication units for the numerator, count each individual medication, defined as an amount lasting 30 days or less, as one medication unit. One medication unit equals one inhaler canister, one injection, or a 30-day or less supply of an oral medication. For example, two inhaler canisters of the same medication dispensed on the same day count as two medication units and only one dispensing event.

99 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Eligible Population Product lines Ages Commercial, Medicaid, Medicare (report each product line separately) years by December 31 of the measurement year. Report five four age stratifications and a total rate: 5 11 years years years years years. Total. The total is the sum of the age stratifications. Continuous enrollment Allowable gap Anchor date Benefits Event/ diagnosis Step 1 Step 2 The measurement year and the year prior to the measurement year. No more than one gap in enrollment of up to 45 days during each year of continuous enrollment. To determine continuous enrollment for a Medicaid beneficiary for whom enrollment is verified monthly, the member may not have more than a 1-month gap in coverage during each year of continuous enrollment year. December 31 of the measurement year. Medical. Pharmacy during the measurement year. Follow the steps below to identify the eligible population for the measure. Identify members as having persistent asthma who met at least one of the following criteria during both the measurement year and the year prior to the measurement year. Criteria need not be the same across both years. At least one ED visit (ED Value Set), with a principal diagnosis of asthma (Asthma Value Set). At least one acute inpatient encounter (Acute Inpatient Value Set), with a principal diagnosis of asthma (Asthma Value Set). At least four outpatient visits (Outpatient Value Set) or observation visits (Observation Value Set), on different dates of service, with any diagnosis of asthma (Asthma Value Set) and at least two asthma medication dispensing events (Table ASM-C). Visit type need not be the same for the four visits. At least four asthma medication dispensing events (Table ASM-C). A member identified as having persistent asthma because of at least four asthma medication dispensing events, where leukotriene modifiers or antibody inhibitors were the sole asthma medication dispensed in that year, must also have at least one diagnosis of asthma (Asthma Value Set), in any setting, in the same year as the leukotriene modifier or antibody inhibitor (i.e., the measurement year or the year prior to the measurement year).

100 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Step 3: Required exclusions Exclude members who met any of the following criteria: Members who had any diagnosis from any of the following value sets, any time during the member s history through December 31 of the measurement year: Emphysema Value Set. Other Emphysema Value Set. COPD Value Set. Obstructive Chronic Bronchitis Value Set. Chronic Respiratory Conditions Due to Fumes/Vapors Value Set. Cystic Fibrosis Value Set. Acute Respiratory Failure Value Set. Members who have no asthma controller or reliever medications dispensed (Table AMR-A) during the measurement year. Administrative Specification Denominator Numerator Step 1 Step 2 Step 3 Step 4 The eligible population. The number of members who have a medication ratio of 0.50 or greater during the measurement year. For each member, count the units of controller medications (Table AMR-A) dispensed during the measurement year. Refer to the definition of Units of medications. For each member, count the units of reliever medications (Table AMR-A) dispensed during the measurement year. Refer to the definition of Units of medications. For each member, sum the units calculated in step 1 and step 2 to determine units of total asthma medications. For each member, calculate the ratio of controller medications to total asthma medications using the following formula. Units of Controller Medications (step 1) Units of Total Asthma Medications (step 3) Step 5 Sum the total number of members who have a ratio of 0.50 or greater in step 4.

101 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Table AMR-A: Asthma Controller and Reliever Medications Description Asthma Controller Medications Prescriptions Antiasthmatic combinations Dyphylline-guaifenesin Guaifenesin-theophylline Antibody inhibitors Omalizumab Inhaled steroid combinations Budesonide-formoterol Fluticasone-salmeterol Mometasone-formoterol Inhaled corticosteroids Beclomethasone Budesonide Ciclesonide Flunisolide Fluticasone CFC free Mometasone Triamcinolone Leukotriene modifiers Montelukast Zafirlukast Zileuton Mast cell stabilizers Cromolyn Methylxanthines Description Short-acting, inhaled beta-2 agonists Aminophylline Dyphylline Albuterol Levalbuterol Theophylline Asthma Reliever Medications Prescriptions Metaproterenol Note: NCQA will post a comprehensive list of medications and NDC codes to by November 3, Data Elements for Reporting Organizations that submit HEDIS data to NCQA must provide the following data elements. Table ASM-1/2: Data Elements for Use of Appropriate Medications for People With Asthma Measurement year Data collection methodology (Administrative) Eligible population Number of required exclusions Numerator events by administrative data Reported rate Lower 95% confidence interval Upper 95% confidence interval Administrative For each age stratification and total For each age stratification and total For each age stratification and total For each age stratification and total For each age stratification and total For each age stratification and total

102 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Medication Management for People With Asthma (MMA) SUMMARY OF CHANGES TO HEDIS 2016 Expanded age range up to 85 years Added the Medicare product line Description The percentage of members years of age during the measurement year who were identified as having persistent asthma and were dispensed appropriate medications that they remained on during the treatment period. Two rates are reported: 1. The percentage of members who remained on an asthma controller medication for at least 50% of their treatment period. 2. The percentage of members who remained on an asthma controller medication for at least 75% of their treatment period. Definitions IPSD Treatment period PDC Oral medication dispensing event Index prescription start date. The earliest prescription dispensing date for any asthma controller medication during the measurement year. The period of time beginning on the IPSD through the last day of the measurement year. Proportion of days covered. The number of days that a member is covered by at least one asthma controller medication prescription, divided by the number of days in the treatment period. One prescription of an amount lasting 30 days or less. To calculate dispensing events for prescriptions longer than 30 days, divide the days supply by 30 and round down to convert. For example, a 100-day prescription is equal to three dispensing events (100/30 = 3.33, rounded down to 3). The organization should allocate the dispensing events to the appropriate year based on the date when the prescription is filled. Multiple prescriptions for different medications dispensed on the same day count as separate dispensing events. If multiple prescriptions for the same medication are dispensed on the same day, sum the day s supply and divide by 30. Use the Drug ID to determine if the prescriptions are the same or different. Refer to the Oral medication dispensing event definition in ASM for examples. Inhaler dispensing event When identifying the eligible population, use the definition below to count inhaler dispensing events. All inhalers (i.e., canisters) of the same medication dispensed on the same day count as one dispensing event. Medications with different Drug IDs dispensed on the same day are counted as different dispensing events. For example, if a member received three canisters of Medication A and two canisters of Medication B on the same date, it would count as two dispensing events.

103 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Allocate the dispensing events to the appropriate year based on the date when the prescription was filled. Use the Drug ID field in the NDC list to determine if the medications are the same or different. Injection dispensing event Calculating number of days covered for multiple prescriptions Injections count as one dispensing event. Multiple dispensing events of the same medication or a different medication count as separate dispensing events. Allocate the dispensing events to the appropriate year based on the date when the prescription was filled. If multiple prescriptions for different medications are dispensed on the same day, calculate number of days covered by a controller medication (for the numerator) using the prescriptions with the longest day s supply. For multiple different prescriptions dispensed on different days with overlapping day s supply, count each day within the treatment period only once toward the numerator. If multiple prescriptions for the same medication are dispensed on the same or different day, sum the day s supply and use the total to calculate the number of days covered by a controller medication (for the numerator). For example, three controller prescriptions for the same medication are dispensed on the same day, each with a 30- day supply, sum the day s supply for a total of 90 days covered by a controller. Subtract any day s supply that extends beyond December 31 of the measurement year. Use the drug ID provided by the NDC to determine if the prescriptions are the same or different. Eligible Population Product lines Ages Commercial, Medicaid, Medicare (report each product line separately) years by December 31 of the measurement year. Report five four age stratifications and a total rate: 5 11 years years years years years. Total The total is the sum of the age stratifications. Continuous enrollment Allowable gap Anchor date Benefits The measurement year and the year prior to the measurement year. No more than one gap in enrollment of up to 45 days during each year of continuous enrollment. To determine continuous enrollment for a Medicaid beneficiary for whom enrollment is verified monthly, the member may not have more than a 1-month gap in coverage during each year of continuous enrollment year. December 31 of the measurement year. Medical. Pharmacy during the measurement year.

104 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Event/diagnosis Step 1 Follow the steps below to identify the eligible population for the measure. Identify members as having persistent asthma who met at least one of the following criteria during both the measurement year and the year prior to the measurement year. Criteria need not be the same across both years. At least one ED visit (ED Value Set), with a principal diagnosis of asthma (Asthma Value Set). At least one acute inpatient encounter (Acute Inpatient Value Set), with a principal diagnosis of asthma (Asthma Value Set). At least four outpatient visits (Outpatient Value Set) or observation visits (Observation Value Set) on different dates of service, with any diagnosis of asthma (Asthma Value Set) and at least two asthma medication dispensing events (Table ASM-C). Visit type need not be the same for the four visits. At least four asthma medication dispensing events (Table ASM-C). Step 2 Step 3: Required exclusions A member identified as having persistent asthma because of at least four asthma medication dispensing events, where leukotriene modifiers or antibody inhibitors were the sole asthma medication dispensed in that year, must also have at least one diagnosis of asthma (Asthma Value Set), in any setting, in the same year as the leukotriene modifier or antibody inhibitor (i.e., measurement year or year prior to the measurement year). Exclude members who met any of the following criteria: Members who had any diagnosis from any of the following value sets, any time during the member s history through December 31 of the measurement year: Emphysema Value Set. Other Emphysema Value Set. COPD Value Set. Obstructive Chronic Bronchitis Value Set. Chronic Respiratory Conditions Due to Fumes/Vapors Value Set. Cystic Fibrosis Value Set. Acute Respiratory Failure Value Set. Members who have no asthma controller medications (Table ASM-D) dispensed during the measurement year. Administrative Specification Denominator The eligible population. Numerators Medication compliance 50% Medication compliance 75% The number of members who achieved a PDC of at least 50% for their asthma controller medications (Table ASM-D) during the measurement year. The number of members who achieved a PDC of at least 75% for their asthma controller medications (Table ASM-D) during the measurement year. Follow the steps below to identify numerator compliance.

105 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Step 1 Step 2 Step 3 Step 4 Identify the IPSD. The IPSD is the earliest dispensing event for any asthma controller medication (Table ASM-D) during the measurement year. To determine the treatment period, calculate the number of days from the IPSD (inclusive) to the end of the measurement year. Count the days covered by at least one prescription for an asthma controller medication (Table ASM-D) during the treatment period. To ensure that day s supply that extends beyond the measurement year is not counted, subtract any day s supply that extends beyond December 31 of the measurement year. Calculate the member s PDC using the following equation. Round (using the.5 rule) to two decimal places. Total Days Covered by a Controller Medication in the Treatment Period (step 3) Total Days in Treatment Period (step 2) Medication compliance 50% Medication compliance 75% Sum the number of members whose PDC is 50% for their treatment period. Sum the number of members whose PDC is 75% for their treatment period. Data Elements for Reporting Organizations that submit HEDIS data to NCQA must provide the following data elements. Table ASM-1/2: Data Elements for Use of Appropriate Medications for People With Asthma Administrative Measurement year Data collection methodology (Administrative) Eligible population For each age stratification and total Number of required exclusions For each age stratification and total Numerator events by administrative data For each age stratification and total Reported rate For each age stratification and total Lower 95% confidence interval For each age stratification and total Upper 95% confidence interval For each age stratification and total

106 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Medication Management for People with Asthma and Asthma Medication Ratio Measures Work-Up Medication Management for People With Asthma Measure Description The percentage of members 5 85 years of age who were identified as having persistent asthma and were dispensed appropriate medications, which they remained on during the treatment period. Two rates are reported: 1. The percentage of members who remained on an asthma controller medication for at least 50 percent of their treatment period. 2. The percentage of members who remained on an asthma controller medication for at least 75 percent of their treatment period. Asthma Medication Ratio Measure Description The percentage of members over 5-85 years of age who were identified as having persistent asthma and had a ratio of controller medications to total asthma medications of 0.50 or greater during the measurement year. Topic Overview Importance and Prevalence Health importance Asthma is one of the most prevalent chronic diseases. In 2010, 25.7 million Americans had asthma: 7 million children, 15.6 million adults under 65 and 3.1 million adults 65 and older (Akinbami et al. 2012). Asthma has become more common over the past decade, occurring in 7.3 percent of the population in 2001, compared with 8.4 percent in 2010 (Akinbami et al. 2012). Asthma is responsible for more than 3,000 deaths in the U.S. annually (American Lung Association 2012b). In 2010, approximately 17.8 million clinical visits (hospital, outpatient, emergency department, and physician offices) were attributed to asthma (Centers for Disease Control and Prevention [CDC] 2014). The incidence rate, and subsequently the number of asthma-related health visits, is expected to increase by an additional 100 million globally by 2025 (World Health Organization 2007). Appropriate medication adherence could ameliorate the severity of many asthmarelated symptoms (Akinbami et al. 2009). According to the Asthma Regional Council of New England, two-thirds of adults and children who display asthma symptoms are considered not well controlled or very poorly controlled as defined by clinical practice guidelines (Stillman 2010). Pharmacologic therapy is used to prevent and control asthma symptoms, improve quality of life, reduce frequency and severity of asthma exacerbations and reverse airflow obstruction (National Heart, Lung, and Blood Institute [NHLBI]/National Asthma and Education Prevention Program [NAEPP] 2007). Utilization, outcome by age, race/ ethnicity, gender The National Health Interview Survey (NHIS) examined asthma prevalence among a range of subgroups from the late 1980s to 2006 (CDC 2009). Survey results showed that children consistently demonstrated higher prevalence and hospitalization rates than adults (CDC 2009). Asthma is among the leading causes of hospitalization for children (American Lung Association 2012b). It disproportionately affects a higher percentage of boys than girls (CDC 2009).

107 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, In addition, children of low-income families experience more urgent care visits, hospitalizations and mortality due to asthma (CDC 2009). In terms of racial/ethnic disparities, asthma prevalence rates are highest for non-hispanic African-American children (14.2 percent) and lowest for Asian children (7.1 percent), with the greatest amount of variability among Hispanic subgroups (CDC 2009). For the adult subgroup, African-Americans have a higher prevalence of asthma than Whites; non-hispanic African-Americans have higher rates of asthma than Hispanics and non-hispanic Whites (CDC 2009). African Americans are also more likely to be hospitalized or die as a result of asthma-based complications (CDC 2009). Women consistently outrank men in prevalence of asthma (CDC 2009), and historically have had higher hospital discharge rates and higher mortality rates due to asthma (CDC 2009). Asthma in the older adult population Asthma prevalence in older adults is comparable to other age groups; clinical practice guidelines suggest the same treatments for all asthma patients over 12 years old. One study found that asthma is more likely to be uncontrolled in older adults (Melani 2013). Asthma in older adults tends to be more severe than asthma developed earlier in life, which can be exacerbated by other comorbid conditions (Reed 2010). Factors such as patients attribution of symptoms to aging and confusing symptoms with other chronic conditions or comorbidities prevent proper recognition and diagnosis of asthma in the older population (Melani 2013). Non-adherence to medication is also an issue, increasing the risk of adverse events, ED visits and hospitalizations, as well as cost (Melani 2013). In adults over 40 years of age, COPD becomes more common and distinguishing asthma from COPD becomes problematic (Global Initiative for Asthma [GINA] 2012). Many patients have symptoms of both asthma and COPD. One study found that from , approximately 7 percent of people 65 and older were estimated to have asthma and 9 percent had COPD. 3 percent were estimated to have co-occurring current asthma and COPD (Oraka et al. 2012). The prevalence of asthma decreases and the prevalence of COPD increases with advancing age (Oraka et al. 2012). The authors concluded that although asthma affects a substantial proportion of the elderly population, increased diagnosis of COPD may overshadow correct diagnosis and treatment of asthma. Financial importance and costeffectiveness Asthma accounted for more than $50 billion spent on health care in the United States in 2007, an increase of almost $2 billion from 2002 (CDC 2011). Inpatient hospitalization accounted for over 50 percent of overall asthma-related costs (Bahadori et al. 2009). In addition to the direct financial burden, asthma is also a leading cause of absenteeism and productivity an estimated 14.2 million missed workdays for adults and more than 14 million missed school days for children (Akinbami et al. 2009). Studies have shown that the indirect costs of asthma are a growing financial burden on patients and result in significant additional costs (Bahadori et al. 2009). Appropriate medication management has the potential to prevent a significant proportion of asthma-related costs (Akinbami et al. 2009). The Asthma Regional Council supports this, stating that proper management could potentially save at least 25 percent ($5 billion) of total national asthma costs, nationally by reducing health care costs (American Lung Association 2012a). The Children s Health Fund s Childhood Asthma Initiative examined patients enrolled in an asthma intervention program. Treatment that aligned with clinical guidelines reduced the severity of symptoms and asthma-related events (Berger 2010). Subsequent savings attributed to improved clinical outcomes totaled nearly $4.2 million, or $4,525 per patient, translating to a significant reduction in federally subsidized and private, insurance-based costs for this population (Berger 2010).

108 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Supporting Evidence for Management of Asthma The goals of therapy for adults and children with asthma are to reduce symptoms and impairment (e.g., coughing, breathlessness, ED visit, hospitalizations, progressive loss of lung function); and to minimize the risk of adverse effects from medication (NHLBI/NAEPP 2007). Numerous randomized control trials have found that a step approach to asthma management improves outcomes (NHLBI/NAEPP 2007; GINA 2014). It involves adjusting medication (increasing doses if necessary, decreasing doses when possible) throughout a cycle of assessment, treatment and review (NHLBI/NAEPP 2007; GINA 2014). Clinical practice guidelines contain different recommendations for children with asthma than for adults with asthma. Guidelines generally reference the age groups 0 4 years, 5 11 years and 12 years and older (Table 1). Asthma Medications Controller medications These medications reduce airway inflammation, control symptoms and reduce future risks such as exacerbations and decline in lung function (GINA 2014). Inhaled corticosteroids (ICS) are the most effective long-term-control medications because they alleviate the underlying inflammation that is characteristic of asthma (NHLBI/NAEPP 2007; British Thoracic Society [BTS]/Scottish Intercollegiate Guidelines Network [SIGN] 2014; Singapore Ministry of Health [SMOH] 2008; GINA 2014; Sveum et al. 2012; Joint Task Force on Practice Parameters 2005). However, sensitivity (and, consequently, clinical response) to ICSs can vary among patients (NHLBI/NAEPP 2007). Leukotriene receptor antagonists or chromones are alternative medications, although they have lower efficacy than ICSs (NHLBI/NAEPP 2007; GINA 2014; Joint Task Force on Practice Parameters 2005). Reliever (rescue) medications Reliever medications are provided to all patients for as-needed relief of symptoms during worsening asthma and exacerbations. Reducing (and, ideally, eliminating) the need for reliever treatment is an important goal in asthma management and a measure of success of asthma treatment (GINA 2014). Short-acting β-agonists (SABA) are bronchodilators that relax smooth muscle and are the preferred therapy for quick relief of asthma symptoms (NHLBI/NAEPP 2007; BTS/SIGN 2014; SMOH 2008; GINA 2014; Sveum et al. 2012; Joint Task Force on Practice Parameters 2005). Anticholinergics can be used as an alternative in patients who do not tolerate SABAs (NHLBI/NAEPP 2007; BTS/SIGN 2014; SMOH 2008). Combination therapy Patients with moderate to severe asthma who have persistent symptoms or exacerbations despite optimized treatment with high-dose controller medications can have other medications added to their primary medication therapy (NHLBI/NAEPP 2007; BTS/SIGN 2014; SMOH 2008; GINA 2014; Sveum 2012; Joint Task Force on Practice Parameters 2005). LABAs, leukotriene modifiers, and theophylline can be added to ICSs (NHLBI/NAEPP 2007; GINA 2014; BTS/SIGN 2014; Sveum 2012); immunomodulators can be added for patients 12 years or older who have allergies and severe persistent asthma (NHLBI/NAEPP 2007). For patients with moderate to severe exacerbations, anticholinergics and oral systemic corticosteroids can be added to SABA treatment to provide added benefit (NHLBI/NAEPP 2007).

109 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Gaps in Care Recent data from the HEDIS Health Plan measures continue to show room for improvement. Medicaid plans consistently perform at lower rates than commercial plans and there is a significant difference in performance rates between the 10th and 90th percentiles for both measures. Performance on the HEDIS asthma measures is summarized below: Medication Management for People With Asthma Asthma Medication Ratio Commercial plan performance on rate 1 ( remained on controller medication for 50 percent of treatment period ) was 67.3 percent from ; Medicaid plans averaged 52.5 percent. Both commercial and Medicaid plans had consistently lower performance on rate 2 ( remained on controller medication for 75 percent of treatment period ) with 43.9 and 30.1 percent, respectively, from Commercial plan performance from was 77.8 percent; Medicaid was 61.4 percent. Health Care Disparities One study highlights disparities in the delivery of care when considering socioeconomic status and race/ ethnicity. Data were collected using the Medical Expenditure Panel Survey (MEPS) ( ), surveying 982 children with asthma younger than 18 years. Results showed that non-hispanic African-American children utilized urgent care services more frequently than preventive care services (Kim et al. 2009). Additionally, children from low-income families were less likely to have prescriptions filled and receive annual primary health examinations (Kim et al. 2009). The study also examined insurance coverage, showing that children with insurance coverage utilized primary health care services for asthma more often (Kim et al. 2009). References Akinbami, L.J., J.E. Moorman, P.L. Garbe, E.J. Sondik Status of Childhood Asthma in the United States Pediatrics 123;S (July 8, 2014) doi: /peds C. Akinbami, L.J., J.E. Moorman, C. Bailey, H.S. Zahran, M. King, C.A. Johnson, X. Liu Trends in Asthma Prevalence, Health Care Use, and Mortality in the United States, NCHS Data Brief, no. 94 (May). (July 9, 2014) American Lung Association. 2012a. Trends in Asthma Morbidity and Mortality. (July 8, 2014) American Lung Association. 2012b. Asthma & Children Fact Sheet, October (July 8, 2014) Bahadori, K., M.M. Doyle-Waters, C. Marra, L. Lynd, K. Alasaly, J. Swiston, J.M. FitzGerald Economic Burden of Asthma: A Systematic Review. BMC Pulmonary Medicine 9: 24. (July 8, 2014) doi: / Berger, S Best Practice Asthma Program Saves the US Healthcare System More Than $4500 A Year Per Child. Columbia University Mailman School of Public Health, May 13. (July 8, 2014) British Thoracic Society (BTS)/Scottish Intercollegiate Guidelines Network (SIGN) British Guideline on the Management of Asthma: A National Clinical Guideline. Edinburgh (Scotland): Scottish Intercollegiate Guidelines Network (SIGN) (October). (February 11, 2015) Centers for Disease Control and Prevention (CDC). Asthma: A Presentation of Asthma Management and Prevention, September (July 8, 2014)

110 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Centers for Disease Control and Prevention (CDC) Asthma Attacks Among Persons with Current Asthma United States, Morbidity and Mortality Weekly Report 62(03); (November), (July 8, 2014) Centers for Disease Control and Prevention (CDC) FastStats: Asthma. Last modified February 25. (July 8, 2014) Centers for Disease Control and Prevention (CDC). Vital Signs: Asthma in the US, May (July 8, 2014) Global Initiative for Asthma (GINA) Global Strategy for Asthma Management and Prevention. (July 10, 2014) Joint Task Force on Practice Parameters Attaining Optimal Asthma Control: A Practice Parameter. Journal of Allergy and Clinical Immunology 116(5): S3-11. Kim, H., G.M. Kieckhefer, A.A. Greek, J.M. Joesch, N. Baydar Health Care Utilization by Children with Asthma. Preventing Chronic Disease 6(1): A12. Melani, A.S Management of Asthma in the Elderly Patient. Clinical Interventions in Aging 8: National Heart Lung and Blood Institute/National Asthma Education and Prevention Program Expert Panel Report 3: Guidelines for the Diagnosis and Management of Asthma. Washington (DC): National Heart Lung and Blood Institute (NHLBI), NIH Publication No (July 8, 2014) Oraka, E., H.J. Kim, M.E. King, D.B. Callahan Asthma Prevalence Among US Elderly by Age Groups: Age Still Matters. Journal of Asthma 49(6): Reed, C.E Asthma in the Elderly: Diagnosis and Management. Journal of Allergy and Clinical Immunology 126(4): Singapore Ministry of Health (SMOH) Clinical Practice Guidelines: Management of Asthma. Singapore: Singapore Ministry of Health. (July 8, 2014) sthma.html. Stillman, L Living with Asthma in New England: Results from the 2006 BRFSS and Call-back Survey. A report by the Asthma Regional Council of New England (February). (July 8, 2014) Sveum, R., J. Bergstrom, G. Brottman, et al Institute for Clinical Systems Improvement: Diagnosis and Management of Asthma (July). (July 8, 2014) World Health Organization Global Surveillance, Prevention and Control of Chronic Respiratory Diseases: A Comprehensive Approach. Switzerland: World Health Organization. (July 8, 2014)

111 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Table 1: Guidelines Organization, Guideline, Date British Thoracic Society/ Scottish Intercollegiate Guidelines Network (SIGN) (2014) Recommendation & Grade STEPWISE MANAGEMENT OF ASTHMA Step 1: Mild intermittent asthma Prescribe an inhaled short-acting β2 agonist as short term reliever therapy for all patients with symptomatic asthma. Age Group >12 years 5-12 years <5 years Grade A B D Step 2: Introduction of regular preventer therapy Inhaled steroids are the recommended preventer drug for adults and children for achieving overall treatment goals. Age Group >12 years 5-12 years <5 years Grade A A A Step 3: Initial add-on therapy The first choice as add-on therapy to inhaled steroids in adults and children (5-12 years) is an inhaled long-acting β2 agonist, which should be considered before going above a dose of 400 micrograms BDP or equivalent per day and certainly before going above 800 micrograms BDP. Age Group >12 years 5-12 years <5 years Grade A B NR Step 4: Poor control on moderate dose of inhaled steroid + additional therapy: addition of fourth drug If control remains inadequate on 800 micrograms BDP daily (adults) and 400 micrograms daily (children) of an inhaled steroid plus a long-acting β2 agonist, consider the following interventions: Increasing inhaled steroids to 2000 micrograms BDP/day (adults) or 800 micrograms BDP/day (children 5-12 years) * Leukotriene receptor antagonists Theophyllines Slow release β2 agonist tablets, though caution needs to be used in patients already on long-acting β2 agonists.

112 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Organization, Guideline, Date continued Age Group >12 years 5-12 years <5 years Recommendation & Grade Grade D. D NR Step 5: Continuous or frequent use of oral steroids For the small number of patients not controlled at step 4, use daily steroid tablets in the lowest dose providing adequate control. ACUTE ASTHMA Use high-dose inhaled β2 agonists as first line agents in acute asthma and administer as early as possible. Reserve intravenous β2 agonists for those patients in whom inhaled therapy cannot be used reliably (Grade: A) Add nebulized ipratropium bromide (0.5 mg 4-6 hourly) to β2 agonist treatment for patients with acute severe or life threatening asthma or those with a poor initial response to β2 agonist therapy (Grade: B) Consider giving a single dose of IV magnesium sulphate for patients with acute severe asthma who have not had a good initial response to inhaled bronchodilator therapy or life threatening or near fatal asthma (Grade: B) Global Initiative for Asthma (2014) TREATMENT STEPS FOR ACHIEVING CONTROL Step 1: As-Needed Reliever Medication Reserved for untreated patients with occasional daytime symptoms of short duration comparable with controlled asthma. When symptoms are more frequent, and/or worsen periodically, patients require regular controller treatment (see Steps 2 or higher) in addition to as-needed reliever medication (Evidence B). For the majority of patients, a rapid-acting inhaled β2-agonist is the recommended reliever treatment (Evidence A). An inhaled anticholinergic, short-acting oral β2-agonist, or short-acting theophylline may be considered as alternatives, although they have a slower onset of action and higher risk of side effects (Evidence A). Step 2: Reliever Medication Plus a Single Controller At Step 2, a low-dose inhaled corticosteroid is recommended as the initial controller treatment for asthma patients of all ages (Evidence A). Alternative controller medications include leukotriene modifiers (Evidence A), appropriate particularly for patients who are unable or unwilling to use inhaled corticosteroids, or who experience intolerable side effects such as persistent hoarseness from inhaled corticosteroid treatment and those with concomitant allergic rhinitis (Evidence C).

113 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Organization, Guideline, Date continued Recommendation & Grade Other options are available but not recommended for routine use as initial or first-line controllers in Step 2. Sustainedrelease theophylline has only weak anti-inflammatory and controller efficacy (Evidence B) and is commonly associated with side effects that range from trivial to intolerable. Cromones (nedocromil sodium and sodium cromoglycate) have comparatively low efficacy, though a favorable safety profile (Evidence A). Step 3: Reliever Medication Plus One or Two Controllers For children, adolescents and adults recommendation is to combine a low-dose of inhaled corticosteroid with an inhaled long-acting β2-agonist, either in a combination inhaler device or as separate components (Evidence A). Because of the additive effect of this combination, the low-dose of corticosteroid is usually sufficient, and need only be increased if control is not achieved within 3 or 4 months with this regimen (Evidence A). The long-acting β2-agonist formoterol, which has a rapid onset of action whether given alone or in combination inhaler with budesonide, has been shown to be as effective as short-acting β2-agonist in acute asthma exacerbation. However its use as monotherapy as a reliever medication is strongly discouraged since it must always be used in association with an inhaled corticosteroid. For all children but particularly those 5 years and younger, combination therapy has been less well studied and the addition of a long-acting β2-agonist may not be as effective as increasing the dose of inhaled corticosteroids in reducing exacerbations. If a combination inhaler containing formoterol and budesonide is selected, it may be used for both rescue and maintenance. This approach has been shown to result in reductions in exacerbations and improvements in asthma control in adults and adolescents at relatively low doses of treatment (Evidence A). Whether this approach can be employed with other combinations of controller and reliever requires further study. Another option for both adults and children, but the one recommended for children, is to increase to a medium-dose of inhaled corticosteroids (Evidence A). For patients of all ages on medium- or high-dose of inhaled corticosteroid delivered by a pressurized metered-dose inhaler (MDI), use of a spacer device is recommended to improve delivery to the airways, reduce oropharyngeal side effects, and reduce systemic absorption (Evidence A). Another option at Step 3 is to combine a low-dose inhaled corticosteroid with leukotriene modifiers (Evidence A). Alternatively, the use of sustained-release theophylline given at low-dose may be considered (Evidence B). These options have not been fully studied in children 5 years and younger.

114 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Organization, Guideline, Date continued Recommendation & Grade Step 4: Reliever Medication Plus Two or More Controllers The selection of treatment at Step 4 depends on prior selections at Steps 2 and 3. However, the order in which additional medications should be added is based upon evidence of their relative efficacy in clinical trials. The preferred treatment at Step 4 is to combine a medium- or high-dose of inhaled corticosteroid with a long-acting inhaled β2-agonist. However, in most patients, the increase from a medium- to a high-dose of inhaled corticosteroid provides relatively little additional benefit (Evidence A), and the high-dose is recommended only on a trial basis for 3 to 6 months when control cannot be achieved with medium-dose inhaled corticosteroid combined with a long-acting β2- agonist and/or a third controller (e.g., leukotriene modifiers or sustained-release theophylline) (Evidence B). Prolonged use of high-dose inhaled corticosteroid is also associated with increased potential for adverse effects. At medium- and high-doses, twice-daily dosing is necessary for most but not all inhaled corticosteroid (Evidence A). With budesonide, efficacy may be improved with more frequent dosing (four times daily) (Evidence B). Leukotriene modifiers as add-on treatment to medium-to high-dose inhaled corticosteroids have been shown to provide benefit (Evidence A), but usually less than that achieved with the addition of a long-acting β2-agonist (Evidence A). The addition of a low-dose of sustained-release theophylline to medium- or high-dose inhaled corticosteroid and long-acting β2-agonist may also provide benefit (Evidence B). Step 5: Reliever Medication Plus Additional Controller Options Addition of oral corticosteroid to other controller medications may be effective (Evidence D) but is associated with severe side effects (Evidence A) and should only be considered if the patient's asthma remains severely uncontrolled on Step 4 medications with daily limitation of activities and frequent exacerbations. Patients should be counseled about potential side effects and all other alternative treatments must be considered. Addition of anti-immunoglobulin E (anti-ige) treatment to other controller medications has been shown to improve control of allergic asthma when control has not been achieved on combinations of other controllers including high-doses of inhaled or oral corticosteroids (Evidence B).

115 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Organization, Guideline, Date Institute for Clinical Systems Improvement (ICSI) (2012) Recommendation & Grade OVERVIEW Clinicians should follow the stepwise approach in asthma management therapy. Clinicians should use inhaled corticosteroids as the preferred treatment over leukotriene receptor antagonists in mild persistent asthma in adults and children. Based on data comparing leukotriene receptor antagonists (LTRAs) to inhaled corticosteroids, inhaled corticosteroids are the preferred treatment option for mild persistent asthma in adults and children. LTRAs are an alternative, although not preferred, treatment. (High Quality Evidence) MANAGEMENT APPROACH FOR ASTHMA IN CHILDREN 5-11 YEARS OF AGE (High Quality Evidence) Step 1: Short-acting Beta2-Agonist prn Step 2: Low-Dose ICS Alternative: Leukotriene Modifier Step 3: Medium-Dose ICS Alternative: Medium-Dose ICS plus (add one) LABA or Leukotriene Modifier Step 4: Medium-Dose ICS plus (add one) LABA or Leukotriene Modifier Step 5: High-dose ICS plus one or more LABA Alternative: High-dose ICS plus leukotriene modifier Alternative: High-dose ICS plus LABA plus oral systemic corticosteroid Alternative: High-dose ICS plus leukotriene modifier plus oral systemic corticosteroid Step 6: High-dose ICS plus LABA plus oral systemic corticosteroid Alternative: High-dose ICS plus leukotriene modifier plus oral systemic corticosteroid MANAGEMENT APPROACH FOR ASTHMA 12 YEARS OF AGE AND OLDER (High Quality Evidence) Step 1: Short-acting Beta2-Agonist as needed Step 2: Low-Dose ICS Alternative: Leukotriene Modifier Step 3: Medium- Dose ICS Alternative: Low-Dose ICS plus LABA Alternative: Low-Dose ICS plus Leukotriene Modifier

116 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Organization, Guideline, Date continued Step 4: Medium-Dose ICS plus LABA Recommendation & Grade Alternative: Medium-Dose ICS plus Leukotriene Modifier Step 5: High-dose ICS plus LABA Alternative: High-dose ICS plus LABA plus one or more leukotriene modifier or Anti-IgE if applicable Step 6: High-Dose ICS plus LABA plus oral corticosteroid Alternative: High-Dose ICS plus LABA plus oral corticosteroid plus one or more leukotriene modifier or Anti-IgE if applicable Joint Task Force on Practice Parameters (American Academy of Allergy, Asthma & Immunology [AAAAI], American College of Allergy, Asthma & Immunology [ACAAI], and the Joint Council of Allergy, Asthma & Immunology [JCAAI]) (2005) National Heart Lung and Blood Institute/National Asthma and Education Prevention Program (NHLBI/NAEPP) (2007) GUIDELINES FOR THE PHARMACOTHERAPY OF ASTHMA The step care of asthma should be based on asthma control. (A) Step 1: Short-acting β-agonist as needed (indicated for all patients) Step 2: Low-dose ICSs, leukotriene modifiers, theophylline, cromolyn, or nedocromil Step 3: Low-dose/medium-dose ICSs plus inhaled LABA or medium-dose ICSs; low-dose/medium-dose ICSs plus either leukotriene modifier or theophylline Step 4: High-dose ICSs and LABA plus systemic corticosteroids if needed (consider monoclonal anti-ige) LONG-TERM CONTROL MEDICATIONS The Expert Panel recommends that long-term control medications (including ICSs, inhaled long-acting bronchodilators, leukotriene modifiers, cromolyn, theophylline, and immunomodulators) be taken daily on a longterm basis to achieve and maintain control of persistent asthma. The most effective long-term-control medications are those that attenuate the underlying inflammation that is characteristic of asthma (Evidence A). Inhaled Corticosteroids (for children and adults with mild persistent asthma) The Expert Panel concludes that ICSs are the most potent and consistently effective long-term control medication for asthma (Evidence A). The Expert Panel concludes that sensitivity and consequently clinical response to ICS can vary among patients (Evidence B). The Expert Panel concludes that studies demonstrate that ICSs improve asthma control more effectively in both children and adults than LTRAs or any other single long-term control medication (Evidence A). Inhaled Long-Acting Beta2-Agonists LABAs are not recommended for use as monotherapy for long-term control of persistent asthma (Evidence A). Use of LABA is not currently recommended to treat acute symptoms or exacerbations of asthma (Evidence D).

117 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Organization, Guideline, Date continued Oral Systemic Corticosteroids Recommendation & Grade The Expert Panel recommends that, because the magnitude of adverse effects is often related to the dose, frequency of administration, and the duration of corticosteroid use (Evidence A), every consideration should be given to minimize systemic corticosteroid doses and maximize other modes of therapy (Evidence D). It is necessary, therefore, to monitor for the development and progression of adverse effects and to take appropriate steps to minimize the risk and impact of adverse corticosteroid effects (Evidence D). Cromolyn Sodium and Nedocromil for mild persistent asthma Cromolyn and nedocromil are alternative, not preferred, medications for the treatment of mild persistent asthma (Evidence A). Immunomodulators for persistent severe asthma The Expert Panel recommends that omalizumab may be considered as adjunctive therapy in step 5 or 6 care for patients who have allergies and severe persistent asthma that is inadequately controlled with the combination of high-dose ICS and LABA (Evidence B). Leukotriene Modifiers for mild persistent asthma The Expert Panel recommends that LTRAs are an alternative, not preferred, treatment option for mild persistent asthma (Step 2 care) (Evidence A). Combination Therapy (for children and adults with moderate to severe persistent asthma) The Expert Panel recommends that when patients 12 years of age require more than low-dose ICS alone to control asthma (i.e., step 3 care or higher), a therapeutic option is to add LABA to ICS (Evidence A). Alternative, but not preferred adjunctive therapies include LTRA (Evidence B), theophylline (Evidence B), or, in adults, zileuton (Evidence D). For children 0 11 years of age, LABA, LTRA, and, in children 5 11 years of age, theophylline may be considered as adjunctive therapies in combination with ICS (Evidence B). QUICK RELIEF MEDICATIONS Inhaled Short-Acting Beta2-Agonists The Expert Panel recommends that SABAs are the drug of choice for treating acute asthma symptoms and exacerbations and for preventing EIB (Evidence A). The Expert Panel recommends the use of SABA as the most effective medication for relieving acute bronchoconstriction; SABAs have few negative cardiovascular effects (Evidence A). The Expert Panel does not recommend regularly scheduled, daily, long-term use of SABA (Evidence A).

118 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Organization, Guideline, Date Singapore Ministry of Health (SMOH) (2008) Systemic Corticosteroids Recommendation & Grade The Expert Panel recommends the use of oral systemic corticosteroids in moderate or severe exacerbations (Evidence A). Combination Therapy: AnticholinergicsThe Expert Panel concludes that ipratropium bromide, administered in multiple doses along with SABA in moderate or severe asthma exacerbations in the ED, provides additive benefit (Evidence B). OBJECTIVES OF ASTHMA MANAGEMENT Inhaled corticosteroids are best used at low to moderate doses (Grade A, Level 1+) Long acting β2-agonists including salmeterol and formoterol should never be used as monotherapy in asthma (Grade A, Level 1+) The strategy of add on therapy with long acting β2-agonists is recommended when a low to medium-dose of inhaled corticosteroids alone fails to achieve control of asthma (Grade A, Level 1++) Formoterol is a long acting β2-agonist which has a rapid onset of action comparable to that of a rapid acting β2-agonist drug. If a combination inhaler containing formoterol and budesonide is considered, it may be used for both rescue and maintenance. This has been shown to reduce exacerbations and improve asthma control in adults and adolescents at relatively low doses of treatment (Grade A, Level 1+) Theophylline has a bronchodilator action and also modest anti-inflammatory properties. It cannot however be used as a controller drug. It may be useful as an add-on drug in patients who do not achieve good control on inhaled corticosteroids alone Leukotreine modifiers such as montelukast have a small and variable bronchodilator effect, reducing symptoms including cough, improving lung function and reducing exacerbations and airway inflammation. It can either be used as an alternative to low dose inhaled corticosteroids in patients with mild persistent asthma, or as an add-on drug when low dose inhaled corticosteroids or when the combination of inhaled corticosteroids with long acting β2-agonist have not given the desired effect (Grade A, Level 1+) The combination of inhaled ipratropium and inhaled β2-agonist may be used in the treatment of acute severe asthma exacerbation (Grade A, Level 1+) Short-term burst oral corticosteroids may be given at the dose of mg/day for 5-10 days as treatment of severe acute exacerbation of asthma and in worsening asthma (Grade A, Level 1+) Regular low doses of oral steroids cause severe and intolerable long-term side effects and should not be used in primary care

119 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Organization, Guideline, Date continued MANAGEMENT OF ASTHMA IN CHILDREN: Recommendation & Grade Rapid-acting inhaled β2-agonists are the medications of choice for relief of bronchoconstriction and for the pre-treatment of exercise induced asthma. β2-agonist metered-dose-inhaler (MDI) delivered by the holding chamber/spacer has been shown to be at least as effective as the nebulizer. Hence routine use of nebulizers is not recommended. During asthma exacerbations, as many as 4-8 puffs of salbutamol inhaler or puffs/kg (max 10 puffs) may be used (Grade A, Level 1++) Long acting inhaled β2-agonists may be used as add-on therapy for children with symptoms which are not controlled with low dose inhaled steroids. These should not be used without concomitant inhaled corticosteroids (Grade A, Level 1+) Only formoterol may be used as a reliever medicine in view of its rapid onset of action (Grade A, Level 1+) For the younger children with nocturnal symptoms, oral long acting β2-agonists may be useful. Sustained release theophylline can be useful for a short duration. It is important to monitor for side effects such as agitation, muscle tremors, palpitations and headache (Grade B, Level 2+) In older children above 5 years, leukotriene modifiers may be used as they provide clinical benefit at all levels of asthma severity. However, clinical benefits are generally less than those with inhaled corticosteroids (Grade A, Level 1++) Leukotriene modifiers may be used as an add-on therapy in children on low to moderate doses of inhaled steroids. In children with poor asthma control, adding a leukotriene modifier may provide additional benefit, including reducing the number of exacerbations (Grade A, Level 1+) A long acting β-agonist or a leukotriene modifier should be added rather than increasing the dose of inhaled steroids if children with mild persistent asthma do not show clinical improvement with inhaled steroids alone (Grade A, Level 1+) Combination agents containing long acting β2-agonists and inhaled steroids may be used in children above 5 years of age whose control is not optimum with low dose inhaled steroids

120 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, GRADING SYSTEM KEY: British Thoracic Society/SIGN Recommendation Grade A: At least one meta-analysis, systematic review, or RCT rated as 1++, and directly applicable to the target population; or A body of evidence consisting principally of studies rated as 1+, directly applicable to the target population, and demonstrating overall consistency of results B: A body of evidence including studies rated as 2++, directly applicable to the target population, and demonstrating overall consistency of results; or Extrapolated evidence from studies rated as 1++ or 1 C: A body of evidence including studies rated as 2+, directly applicable to the target population and demonstrating overall consistency of results; or Extrapolated evidence from studies rated as 2++ D: Evidence level 3 or 4; or Extrapolated evidence from studies rated as 2+ Evidence Level 1++ High quality meta-analyses, systematic reviews of RCTs, or RCTs with a very low risk of bias 1+ Well conducted meta-analyses, systematic reviews, or RCTs with a low risk of bias 1 - Meta-analyses, systematic reviews, or RCTs with a high risk of bias 2++ High quality systematic reviews of case control or cohort studies; High quality case control or cohort studies with a very low risk of confounding or bias and a high probability that the relationship is causal 2+ Well conducted case control or cohort studies with a low risk of confounding or bias and a moderate probability that the relationship is causal 2 - Case control or cohort studies with a high risk of confounding or bias and a significant risk that the relationship is not causal 3 Non-analytic studies, eg case reports, case series 4 Expert opinion Global Initiative for Asthma Evidence Levels: A : Randomized controlled trials (RCT). Rich body of data. Evidence is from endpoints of well-designed RCTs that provide a consistent pattern of findings in the population for which the recommendation is made. Category A requires substantial numbers of studies involving substantial numbers of participants. B: RCTs. Limited body of data. Evidence is from endpoints of intervention studies that include only a limited number of patients, post hoc or subgroup analysis of RCTs, or meta-analysis of RCTs. In general, Category B pertains when few randomized trials exist, they are small in size, they were undertaken in a population that differs from the target population of the recommendation, or the results are somewhat inconsistent. C: Nonrandomized trials. Observational studies. Evidence is from outcomes of uncontrolled or nonrandomized trials or from observational studies. D: Panel consensus judgment. This category is used only in cases where the provision of some guidance was deemed valuable but the clinical literature addressing the subject was insufficient to justify placement in one of the other categories. The Panel Consensus is based on clinical experience or knowledge that does not meet the above-listed criteria. ICSI High Quality Evidence = Further research is very unlikely to change our confidence in the estimate of effect. Moderate Quality Evidence = Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low Quality Evidence = Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate or any estimate of effect is very uncertain.

121 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Joint Task Force on Practice Parameters (AAAAI, ACAAI, JCAAI) Strength of Recommendation A: Directly based on category I evidence B: Directly based on category II evidence or extrapolated recommendation from category I evidence C: Directly based on category III evidence or extrapolated recommendation from category I or II evidence D: Directly based on category IV evidence or extrapolated recommendation from category I, II, or III evidence Category of Evidence Ia: Evidence from meta-analysis of randomized controlled trials Ib: Evidence from at least one randomized controlled trial IIa: Evidence from at least one controlled study without randomization IIb: Evidence from at least one other type of quasiexperimental study III: Evidence from non-experimental descriptive studies, such as comparative studies, correlation studies, and case-control studies IV: Evidence from expert committee reports, opinions or clinical experiences of respected authorities, or both (NHLBI/NAEPP Level of Evidence: Category A: Randomized controlled trials (RCTs), rich body of data. Evidence is from end points of welldesigned RCTs that provide a consistent pattern of findings in the population for which the recommendation is made. Category A requires substantial numbers of studies involving substantial numbers of participants. Category B: RCTs, limited body of data. Evidence is from end points of intervention studies that include only a limited number of patients, post hoc or subgroup analysis of RCTs, or meta-analysis of RCTs. In general, category B pertains when few randomized trials exist; they are small in size, they were undertaken in a population that differs from the target population of the recommendation or the results are somewhat inconsistent. Category C: Nonrandomized trials and observational studies. Evidence is from outcomes of uncontrolled or nonrandomized trials or from observational studies. Category D: Panel consensus judgment. This category is used only in cases where the provision of some guidance was deemed valuable, but the clinical literature addressing the subject was insufficient to justify placement in one of the other categories. The Panel consensus is based on clinical experience or knowledge that does not meet the criteria for categories A through C.

122 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, SMOH Recommendation Grate A: At least one meta-analysis, systematic review of RCTs, or RCT rated as 1+ + and directly applicable to the target population; or A body of evidence consisting principally of studies rated as 1+, directly applicable to the target population, and demonstrating overall consistency of results B: A body of evidence including studies rated as 2++, directly applicable to the target population, and demonstrating overall consistency of results; or Extrapolated evidence from studies rated as 1+ + or 1+ C: A body of evidence including studies rated as 2+, directly applicable to the target population and demonstrating overall consistency of results; or Extrapolated evidence from studies rated as 2+ + D: Evidence level 3 or 4; or Extrapolated evidence from studies rated as 2+ GPP: Recommended best practice based on the clinical experience of the guideline development group. Evidence Levels Category 1++: High quality meta-analyses, systematic reviews of randomized controlled trials (RCTs), or RCTs with a very low risk of bias. Category 1+: Well conducted meta-analyses, systematic reviews of RCTs, or RCTs with a low risk of bias. Category 1-: Meta-analyses, systematic reviews of RCTs, or RCTs with a high risk of bias. Category 2++: High quality systematic reviews of case control or cohort studies. High quality case control or cohort studies with a very low risk of confounding or bias and a high probability that the relationship is causal Category 2+: Well conducted case control or cohort studies with a low risk of confounding or bias and a moderate probability that the relationship is causal Category 2-: Case control or cohort studies with a high risk of confounding or bias and a significant risk that the relationship is not causal Category 3: Non-analytic studies, e.g. case reports, case series Category 4: Expert opinion Guideline References British Thoracic Society (BTS)/Scottish Intercollegiate Guidelines Network (SIGN) British Guideline on the Management of Asthma: A National Clinical Guideline. Edinburgh (Scotland): Scottish Intercollegiate Guidelines Network (SIGN) (October). (February 11, 2015) Global Initiative for Asthma (GINA) Global Strategy for Asthma Management and Prevention. (July 10, 2014) Joint Task Force on Practice Parameters Attaining Optimal Asthma Control: A Practice Parameter. Journal of Allergy and Clinical Immunology 116(5): S3-11. National Heart Lung and Blood Institute/National Asthma Education and Prevention Program Expert Panel Report 3: Guidelines for the Diagnosis and Management of Asthma. Washington (DC): National Heart Lung and Blood Institute (NHLBI), NIH Publication No (July 8, 2014) Singapore Ministry of Health (SMOH) Clinical Practice Guidelines: Management of Asthma. Singapore: Singapore Ministry of Health. (July 8, 2014) sthma.html. Sveum, R., J. Bergstrom, G. Brottman, et al Institute for Clinical Systems Improvement: Diagnosis and Management of Asthma (July). (July 8, 2014)

123 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Proposed Changes to Existing Measure for HEDIS : Medication Reconciliation Post-Discharge (MRP) NCQA seeks comments on proposed modifications to Medication Reconciliation Post-Discharge, which assesses the percentage of discharges from acute or nonacute inpatient facilities for members 66 years of age and older, for whom medications were reconciled within 30 days of discharge. We propose the following changes to this measure: Eligible population: Add Medicare as a product line and expand the age range to include Medicare beneficiaries 18 years and older. Hybrid Specification: Add examples of medication reconciliation. This measure is currently reported only by Medicare Advantage Special Needs Plans (SNP), which serve individuals with chronic conditions, individuals with dual eligibility for Medicare and Medicaid and individuals who reside in institutional care settings. Prescription medication use is common among adults of all ages, particularly adults with chronic conditions, who make up the majority of Medicare Advantage beneficiaries. More than two-thirds of Medicare beneficiaries have two or more chronic conditions; 14 percent have six or more (CMS 2012). Expanding the measure to include all Medicare Advantage beneficiaries provides an opportunity to measure the quality of care coordination post-discharge, as well as patient safety. The majority of beneficiaries served by SNPs and Medicare Advantage Plans are 65 and older; however, both can (and do) serve individuals who are under 65; for example, those with significant disability (e.g., dualeligible). In 2013, 30 percent of hospital discharges in SNPs and 16 percent of hospital discharges in MA plans were for adults 18 64, who are as likely to benefit from medication reconciliation as older adults. Expanding the age range would encourage medication reconciliation among all adults who are discharged from an inpatient facility. Adding medication reconciliation examples to the hybrid specification will provide clarity about the type of documentation that counts as evidence of medication reconciliation. Although these examples do not identify the ideal process of medication reconciliation, they identify the type of documentation likely to be available in a medical record to demonstrate medication reconciliation. Supporting documents for the proposed measure include the draft measure specifications, evidence work-up and performance data. NCQA acknowledges the contributions of the Geriatric Measurement Advisory Panel and the Technical Measurement Advisory Panel. Centers for Medicare & Medicaid Services (CMS). Chronic Conditions among Medicare Beneficiaries, Chartbook, 2012 Edition. Baltimore, MD HEDIS is a registered trademark of the National Committee for Quality Assurance (NCQA).

124 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Medication Reconciliation Post-Discharge (MRP) SUMMARY OF CHANGES TO HEDIS 2016 Added Medicare as a product line. Expanded age range to include Medicare beneficiaries 18 years and older. Added examples of medication reconciliation. Description The percentage of discharges from January 1 December 1 of the measurement year for members 6618 years of age and older for whom medications were reconciled on or within 30 days of discharge. Definition Medication reconciliation A type of review in which the discharge medications are reconciled with the most recent medication list in the outpatient medical record. Eligible Population Product line Ages Continuous enrollment Allowable gap Anchor date Benefit Event/ diagnosis Medicare SNP, Medicare years and older as of December 31 of the measurement year. Date of discharge through 30 days after discharge. None. Date of discharge. Medical. An acute or nonacute inpatient discharge on or between January 1 and December 1 of the measurement year. The denominator for this measure is based on discharges, not members. If members have more than one discharge, include all discharges on or between January 1 and December 1 of the measurement year. Readmission or direct transfer If the discharge is followed by a readmission or direct transfer to an acute or nonacute facility within the 30-day follow-up period, count only the readmission discharge or the discharge from the facility to which the member was transferred. Exclude both the initial discharge and the readmission/direct transfer discharge if the readmission/direct transfer discharge occurs after December 1 of the measurement year. Note: If a member remains in an acute or nonacute facility through December 1 of the measurement year, a discharge is not included in the measure for this member. However, the organization must have a method for identifying the member s status for the remainder of the measurement year, and may not assume the member remained in the facility based only on the absence of a discharge before December 1.

125 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Administrative Specification Denominator Numerator The eligible population. Medication reconciliation (Medication Reconciliation Value Set) conducted by a prescribing practitioner, clinical pharmacist or registered nurse on or within 30 days of discharge. Hybrid Specification Denominator A systematic sample drawn from the eligible population. Organizations may reduce the sample size using the current year s administrative rate or the prior year s audited, product line-specific rate. Refer to the Guidelines for Calculations and Sampling for information on reducing the sample size. The denominator is based on episodes, not on members. Members may appear more than once in the sample. Numerator Administrative Medical record Medication reconciliation conducted by a prescribing practitioner, clinical pharmacist or registered nurse, as documented through either administrative data or medical record review on or within 30 days of discharge. Refer to Administrative Specification to identify positive numerator hits from administrative data. Documentation in the medical record must include evidence of medication reconciliation and the date when it was performed. Any of the following evidence meets criteria: Documentation of the current medications, with a notation that references the discharge medications (e.g., no changes in meds since discharge, same meds at discharge, discontinue all discharge meds ). Documentation of the member s current medications, with a notation that the discharge medications were reviewed. Documentation that the provider reconciled the current and discharge meds. Documentation of a current medication list, a discharge medication list and notation that the appropriate practitioner type reviewed both lists on the same date of service. Notation that no medications were prescribed or ordered upon discharge. Notation that the medications prescribed or ordered upon discharge were reconciled with the current medications (in the outpatient record) by the appropriate practitioner type. A medication list in a discharge summary that is present in the outpatient chart and evidence of a reconciliation with the current medications conducted by an appropriate practitioner type (the organization must be able to distinguish between the member s discharge medications and the member s current medications). Notation that no medications were prescribed or ordered upon discharge. Only documentation in the outpatient chart meets the intent of the measure, but an outpatient visit is not required.

126 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Note The denominator is based on the discharge date found in administrative/claims data, but organizations may use other systems (including data found during medical record review) to identify data errors and make corrections. Data Elements for Reporting Organizations that submit HEDIS data to NCQA must provide the following data elements. Table MRP-3: Data Elements for Medication Reconciliation Post-Discharge Administrative Hybrid Measurement year Data collection methodology (Administrative or Hybrid) Eligible population Number of numerator events by administrative data in eligible population (before exclusions) Current year s administrative rate (before exclusions) Minimum required sample size (MRSS) or other sample size Oversampling rate Final sample size (FSS) Number of numerator events by administrative data in FSS Administrative rate on FSS Number of original sample records excluded because of valid data errors Number of employee/dependent medical records excluded Records added from the oversample list Denominator Numerator events by administrative data Numerator events by medical records Reported rate Lower 95% confidence interval Upper 95% confidence interval

127 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Medication Reconciliation Post-Discharge (MRP) Measure Work-Up Measure Description The percentage of discharges from January 1 December 1 of the measurement year for members 18 years of age and older, for whom medications were reconciled on or within 30 days of discharge. Topic Overview Importance and Prevalence Medication reconciliation is a critical piece of care coordination post-discharge for all individuals who use prescription medications. On average, 82 percent of all adults in the U.S. take at least one medication (prescription or nonprescription, vitamin/mineral, herbal/natural supplement); 29 percent take five or more. Older adults are the biggest consumers of medications: 17 percent 19 percent of people 65 and older take at least 10 medications in a given week (BU 2006). In addition, 62 percent of adults 65 and older have multiple chronic conditions. The more chronic conditions they experience, the more providers are involved in their care. As the number of providers increases, the less likely patients are to understand, remember and reconcile their instructions (Vogeli 2007). Younger adults with disability are also likely to have multiple chronic conditions. 43 percent of Medicare- Medicaid beneficiaries under 65 have more than one physical condition; 23 percent have more than one mental or cognitive condition (Kasper 2010). Many conditions require complex medication regimens that have the potential for significant adverse effects if not appropriately monitored across sites of care, such as antipsychotic medication for individuals with schizophrenia. Therefore, ensuring that proper medication reconciliation is conducted for both older adults and younger adults is an important element of patient safety. The high prevalence of prescription medication use can result in potentially negative consequences for patients if not used and monitored appropriately. It is estimated that approximately 1.5 million preventable adverse drug events occur in the United States each year (IOM 2007). Many result from medication errors, drug interaction or inappropriate use of medications. Hospitalization puts patients at a high risk for medication errors because hospital medication records are often incomplete. A comparison of medication histories for admitted patients with community pharmacy records revealed that hospital records omitted 25 percent of the medications in use. Patients are discharged without being continued on chronic medications (Lau 2000). Significant changes can occur to a patient s medications during hospitalization. A study by Beers et al. found that 45 percent of all discharge medications are initiated during hospitalization (1989). Provider errors and patient misunderstanding of discharge medications is also common. One observational study found that 81.4 percent of patients experienced a provider error or had no understanding of at least one intended medication change upon discharge. These patients were more likely to misunderstand medication changes that were unrelated to the primary diagnosis, suggesting the importance of proper communication to the patient prior to and following discharge (Ziaeian 2012). Implementing routine medication reconciliation after discharge from an inpatient facility is an important step to ensuring that medication errors are addressed and that patients understand their new medications. Resolving discrepancies in a patient s medication list reduces the risk of adverse drug interactions and helps physicians minimize duplication and complexity of a medication regimen (Wenger 2004). This supports patient adherence to the regimen and has the potential to reduce hospital readmission rates (Burniske 2007).

128 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Opportunity for improvement Communication between the inpatient facility and the patient s primary caregiver can be delayed or incomplete, resulting in duplication of medications or administration of medications with potentially harmful interactions (Williams 1990). Hospital admissions are also associated with unintentional discontinuation of medication for chronic conditions (Bell 2011). The medication list provided by the hospital may not by complete at the time of discharge, resulting in patients inadvertently stopping medication for ongoing chronic conditions or starting new medications that may adversely affect existing conditions or other medications (Stafford 2011, IOM 2006). Numerous evaluations have established that medication reconciliation postdischarge is an effective tool to reduce preventable adverse drug events that can be associated with injury or death (Pronovost 2003, IHI 2011). Evidence Supporting Medication Reconciliation Post-Discharge In the case of medication reconciliation after discharge from the hospital, the patient s discharge medication list is compared with the list of medications the patient was taking prior to hospitalization. This can avoid medication errors such as omissions, duplications, dosing errors or drug interactions, and should be done at every transition of care in which new medications are ordered or existing orders are rewritten. The Joint Commission identifies five steps in the medication reconciliation process: 1. Develop a list of current medications. 2. Develop a list of medications to be prescribed. 3. Compare the medications on the two lists. 4. Make clinical decisions based on the comparison. 5. Communicate the new list to appropriate caregivers and to the patient (The Joint Commission 2006). Medication reconciliation post-discharge can catch potentially harmful omissions or changes in prescribed medications, particularly for elderly and disabled patients, who are prescribed a greater quantity and variety of medications (Leape 1991). Although the magnitude of the effect of medication reconciliation alone on patient outcomes is not well studied in isolation, there is agreement among experts that potential benefits outweigh the harm (Coleman 2003, Pronovost 2003, IOM 2002, IOM 2006). To our knowledge, there are no systematic reviews of the effect of medication reconciliation in the outpatient setting alone on health outcomes for adults. However, individual studies have shown a decrease in medication errors when medication reconciliation, and other transition interventions, are implemented (Bayoumi 2009, Coleman 2003, Gillespie 2009, Nassaralla 2007, Geurts 2012, Midlov 2012). Medication reconciliation is a critical component of several widely disseminated care transitions models, including the Transitional Care Model, Care Transitions Program, Project RED, and Project BOOST. A systematic review to identify the most effective methods for medication reconciliation in the hospital setting found evidence to support medication reconciliation interventions that use pharmacy staff and focus on patients at high risk for adverse events (Mueller 2012). A study by Gillespie et al utilized a randomized pharmacist-led medication review process of hospitalized patients and demonstrated a subsequent 16 percent reduction in all visits to the hospital and a 47 percent reduction in visits to the emergency department (Gillespie 2009). This intervention may also ease the financial burden that medication errors place on the medical system. A study utilizing a pharmacist-led medication review concluded that there was a $230 decrease in cost per patient (Gillespie 2009). Other successful interventions have used multidisciplinary groups to facilitate medication reconciliation (Greenwald 2010). An academic medical center found that by instituting a pharmacy-driven multidisciplinary admission history and medication reconciliation process, medication errors were reduced from 90 percent to 47 percent on the surgical unit and from 57 percent to 33 percent on the medicine unit (Murphy 2009). Another care setting used consumer-based kiosk technology to check in

129 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, patients for appointments and, as part of the check-in process, required patients to review their current medication list (each medication was matched with a pill picture). The goal of this model was to compare the medication list gathered at the kiosk against the existing list in the health record. Instituting this process found an average of 4.59 medication discrepancies and an average of 1.61 potentially lethal discrepancies. This process also reduced nursing time dedicated to reconciliation activities by 50 percent (Lesselroth 2009). Medication reconciliation post-discharge is recommended by the Joint Commission patient safety goals (Kienle 2008), the American Geriatric Society (Coleman 2003), Society of Hospital Medicine (Kripalani 2007; Grennwald 2010), Assessing Care of Vulnerable Elders (ACOVE) (Knight 2001),and the Task Force on Medicines Partnership (2002). Additionally, measurement of medication reconciliation post-discharge has been cited by the National Quality Forum and the National Priorities Partnership as a measurement priority area (NQF 2010). Gaps in care The majority (82.1 percent) of SNP plans are able to report on this measure. Recent data from the HEDIS Health Plan measure set shows that while the rates for this measure are low, they have increased in the past year. Between 2013 and 2014, average performance for this measure increased 10.4 percent (2013: 26.2 percent; 2014: 36.6 percent). This is a change in trend from previous years that have shown a decrease in performance. From , average performance for this measure decreased 4.9 percent (2011: 31.1 percent; 2013: 26.2 percent). Part of the observed decline could be attributed to a change in the measure specification from In the prior specification (HEDIS 2012), NCQA permitted health plans to conclude that members were still in the inpatient (acute or nonacute) setting if there was no record that they had been discharged home. In the most recent specification (HEDIS 2013), a health plan must have documentation that a member is still in the inpatient setting. If there is no documentation, the member may not be excluded from the denominator. Because this documentation can be challenging for health plans to track down, it might have led to a decrease in performance from We looked at the relationship between SNP performance on this measure and the Care for Older Adults (COA) medication review measure. We hypothesized that performance on these measures should be highly correlated because plans that excel at medication management should perform well on both measure. We found a moderate and statistically significant correlation between the two measures for 2013 (r statistic=.4408; significance=<.0001) which suggests MRP is a valid measure of a health plan s quality of medication management. Health care disparities Although there is no evidence on disparities in receiving medication reconciliation, there is evidence that disparities in health outcomes may be explained by a patients inability to afford prescription medications. Studies have documented particularly low adherence rates among poor and ethnic minorities (Cobaugh 2008). Income inadequacy is a strong predictor for not filling prescription medications. One study that looked at racial disparities in the quality of medication use in older adults found that 28 percent of Blacks could not purchase their medication due to cost, compared with 12 percent of Whites (Roth 2010). Medication reconciliation post-discharge may be particularly important for patients with poor adherence to their medications, so the prescriber can evaluate what the patient is taking and reinforce which medications are most needed to improve health.

130 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, References Bayoumi, I., M. Howard, A.M. Holbrook, I. Schabort Interventions to improve medication reconciliation in primary care. Ann Parmacother 43: Beers, M.H., J. Dang, J. Hasegawa, I.Y. Tamai Influence of hospitalization on drug therapy in the elderly. J Am Geriatr Soc 37(8): Bell, C.M., S.S. Brener, N. Gunraj, et al Association of ICU or hospital admission with unintentional discontinuation of medication for chronic diseases. JAMA 306: Burniske G., J. O Donnell, V.K. Chetty, et al Impact of pharmacist follow-up telephone calls after intensive nurse-based patient education upon hospital discharge. Poster presented at the annual meeting of the American College of Clinical Pharmacy, Denver, Colorado, October. Cobaugh, D.J., E. Angner, C.I. Kiefe, et al Effect of racial differences on ability to afford prescription medications. Am J Health Syst Pharm. 65(22): doi: /ajhp Coleman, E.A., C.E. Boult, American Geriatrics Society Health Care Systems Committee Improving the Quality of Transitional Care for Persons with Complex Care Needs. J Am Geriatr Soc. 51(4): Geurts, M.M., J. Talsma, J.R. Brouwers, J.J. de Gier Medication Review and Reconciliation with Cooperation between Pharmacist and General Practitioner and the Benefit for the Patient: a Systematic Review. Br J Clin Pharmacol. Epub ahead of print. Jan 13. Gillespie, U., A. Alassaad, D. Henrohn, et al A Comprehensive Pharmacist Intervention to Reduce Morbidity in Patients 80 Years or Older. Arch Intern Med 169: Grennwald, J.L., L. Halasyamani, J. Greene, et al Making inpatient medication reconciliation patient centered, clinically relevant, and implementable: A consensus statement on key principles and necessary first steps. J Hosp Med 5: Institute for Healthcare Improvement (IHI) Reconcile Medications at All Transition Points: Reconcile Medications in Outpatient Settings. Available at: Accessed December Institute of Medicine (IOM) Preventing Medication Errors: Quality Chasm Series. Washington, DC: The National Academies Press. IOM Committee on Quality Health Care in America: To err is human: building a safer health system. Washington, DC: National Academy Press. IOM Preventing Medication Errors. National Academies Press, Washington D.C. Kasper, J., M. Watts, B. Lyons Chronic Disease and Co-Morbidity Among Dual Eligibles: implication for Patterns of Medicaid and Medicare Service Use and Spending. Kaiser Commission on Medicaid and Uninsured. Available at Kienle, P., J.P. Uselton Maintaining Compliance with Joint Commission Medication Management Standards. Patient Safety and Quality Healthcare. July/August. Kripalani, S., A.T. Jackson, J.L. Schnipper, E.A. Coleman Promoting effective transitions of care at hospital discharge: a review of key issues for hospitals. J Hosp Med 2: Knight, E.L., J. Avorn Quality Indicators for appropriate medication use in vulnerable elders. Ann Intern Med 135(8 Pt 2): Lau, H.S., C. Florax, A.J. Porsius, A. De Boer The completeness of medication histories in hospital medical records of patients admitted to general internal medicine wards. Br J Clin Pharmacol 49(6): Leape, L.L., T.A. Brennan, et al The Nature of Adverse Events in Hospitalized Patients. N Engl J Med 324(6): Lesselroth, B., S. Adams, R. Felder, et al Using consumer-based kiosk technology to improve and standardize medications reconciliation in a specialty care setting. Jt Comm J Qual Patient Saf 35(5): Midlov, P., L. Bahrani, M. Seyfali et al The effect of medication reconciliation in elderly patients at hospital discharge. Int J Clin Pharm 34(1): doi: /s Epub 2011 Dec 30. Mueller, S.K., K. Cunningham, S. Kripalani, J. Schnipper Hospital-Based Medication Reconciliation Practices. Arch Intern Med 172(14): doi: /archinernmed Murphy, E.M.,C.J. Oxencis, J.A. Klauck, et al Medication reconciliation at an academic medical center: Implementation of a comprehensive program from admission to discharge. Am J Health Syst Pharm 66(23): doi: /ajhp

131 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Nassaralla, C.L., J.M. Naessens, R. Chaudhry, et al Implementation of a medication reconciliation process in an ambulatory internal medicine clinic. Qual Saf Health Care 16(2):90-4. National Quality Forum (NQF) Preferred Practices and Performance Measures for Measuring and Reporting Care Coordination: A Consensus Report. Washington, DC. Boston University (BU) Patterns of medications use in the United States 2006: a report from the Slone Survey. Available at: Accessed: July 17, Pronovost, P., B. Weast, M. Schwarz, et al Medication Reconciliation: A Practical Tool to Reduce the Risk of Medication Errors. J Crit Care 18(4): Roth, M. T., D.A. Esserman, J.L. Ivey, M. Weinberger Racial disparities in the quality of medication use in older adults: baseline findings from a longitudinal study. 25(3): doi: /s Epub 2009 Dec 11. Stafford, L., A. Stafford, J. Hughes, et al Drug-related problems identified in post-discharge medication reviews for patients taking warfarin. Int J Clin Pharm 33(4): doi: /s Epub 2011 May 19. Task Force on Medicines Partnership The National Collaborative medicines Management Services Programme: Room for Review, A Guide to Medication Review. London. Accessed via: Reviewed September The Joint Commission Medication reconciliation, sentinel event alert. Vogeli, C., A.E. Shields, T.A. Lee, et al Multiple Chronic Conditions: Prevalence, Health Consequences, and Implications for Quality, Care Management, and Costs. J Gen Intern Med 22(suppl 3): Wenger, N.S. and R. Young Working paper: Quality Indicators of Continuity and Coordination of Care for Vulnerable Elder Persons. Rand. Williams, E.I. and F. Filton General practitioner response to elderly patients discharged from hospital. BMJ 300(6718): Ziaeian, B., K.L.B. Araujo, P. Van Ness, L. Horwitz Medication Reconciliation Accuracy and Patient Understanding of Intended Medication Changes on Hospital Discharge. J Gen Intern Med 27(11): doi: /s Epub 2012 Jul 14.

132 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, HEDIS Health Plan Performance Rates: Medication Reconciliation Post-Discharge (MRP) Table 1. HEDIS MRP Measure Performance Medicare HMO Plans Year Total Number of Plans Plans Able to Report (%) Average Standard Deviation 10th Percentile 25th Percentile 50th Percentile 75th Percentile (83.2) (79.5) (80.3) * (82.1) *For 2014 the average number of discharges was 655.9, with a standard deviation of 1, th Percentile Note: Due to a specification change, trending between should be interpreted with caution.

133 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Proposed Retirement for HEDIS : Use of Appropriate Medications for People With Asthma (ASM) Proposed Changes to Existing Measures for HEDIS 2016: Medication Management for People With Asthma (MMA) Asthma Medication Ratio (AMR) NCQA seeks comments on proposed changes to the HEDIS Asthma measure set: Retire Use of Appropriate Medications for People With Asthma. Expand the upper age limit for Medication Management for People With Asthma and Asthma Medication Ratio to include health plan members up to 85 years of age; add Medicare as a reporting product line. The proposed retirement of Use of Appropriate Medications for People With Asthma measure results from consistently high HEDIS performance rates and little variation in plan performance for both commercial and Medicaid plans. Additionally, Medication Management for People With Asthma is a more effective way of assessing asthma medication management. The Medication Management for People With Asthma and Asthma Medication Ratio measures evaluate the effectiveness of asthma management in members 5 64 years of age. Stakeholders expressed interest in expanding the measures to include older adults. NCQA tested the feasibility of expanding the eligible population to ages 65 and older and whether it would provide meaningful and valid HEDIS performance information: specifically, because the frequency of clinical measure exclusions increases with age, would health plans have a sufficient denominator? Based on test findings, a sufficient number of adults 65 and older remain after applying exclusions to the older commercial and Medicare populations, which supports including these adults in the HEDIS Asthma measure set. Supporting documents for these measures include the current measure specifications and measure work-up. NCQA acknowledges the contributions of the Respiratory Measurement Advisory Panel and the Geriatric Measurement Advisory Panel. 1 HEDIS is a registered trademark of the National Committee for Quality Assurance (NCQA).

134 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Use of Appropriate Medications for People With Asthma (ASM) PROPOSED RETIREMENT FOR HEDIS 2016 Description The percentage of members 5-64 years of age during the measurement year who were identified as having persistent asthma and who were appropriately prescribed medication during the measurement year. Definitions Oral medication dispensing event One prescription of an amount lasting 30 days or less. To calculate dispensing events for prescriptions longer than 30 days, divide the day s supply by 30 and round down to convert. For example, a 100-day prescription is equal to three dispensing events (100/30 = 3.33, rounded down to 3). The organization should allocate the dispensing events to the appropriate year based on the date when the prescription is filled. Multiple prescriptions for different medications dispensed on the same day are counted as separate dispensing events. If multiple prescriptions for the same medication are dispensed on the same day, sum the day s supply and divide by 30. Use the Drug ID to determine if the prescriptions are the same or different. Two prescriptions for different medications dispensed on the same day, each with a 60-day supply, equals four dispensing events (two prescriptions with two dispensing events each). Two prescriptions for different medications dispensed on the same day, each with a 15-day supply, equals two dispensing events (two prescriptions with one dispensing event each). Two prescriptions for the same medication dispensed on the same day, each with a 15-day supply, equals one dispensing event (sum the day s supply for a total of 30 days). Two prescriptions for the same medication dispensed on the same day, each with a 60-day supply, equals four dispensing events (sum the day s supply for a total of 120 days). Inhaler dispensing event Injection dispensing event All inhalers (i.e., canisters) dispensed on the same calendar day count as one dispensing event, regardless if they are the same medication or a different medication. For example, two inhalers dispensed on the same day count as one dispensing event. Two inhalers dispensed on different dates of service count as two dispensing events. Allocate the dispensing events to the appropriate year based on the date when the prescription was filled. Injections count as one dispensing event. Multiple dispensing events of the same medication or a different medication count as separate dispensing events. Allocate the dispensing events to the appropriate year based on the date when the prescription was filled.

135 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Eligible Population Product lines Ages Commercial, Medicaid (report each product line separately) years by December 31 of the measurement year. There are four age stratifications and a total rate: 5 11 years years years years. Total. The total is the sum of the age stratifications. Continuous enrollment Allowable gap Anchor date Benefits Event/ diagnosis Step 1 The measurement year and the year prior to the measurement year. No more than one gap in enrollment of up to 45 days during each year of continuous enrollment. To determine continuous enrollment for a Medicaid beneficiary for whom enrollment is verified monthly, the member may not have more than a 1-month gap in coverage during each year of continuous enrollment year. December 31 of the measurement year. Medical. Pharmacy during the measurement year. Follow the steps below to identify the eligible population for the measure. Identify members as having persistent asthma who met at least one of the following criteria during both the measurement year and the year prior to the measurement year. Criteria need not be the same across both years. At least one ED visit (ED Value Set), with a principal diagnosis of asthma (Asthma Value Set). At least one acute inpatient encounter (Acute Inpatient Value Set), with a principal diagnosis of asthma (Asthma Value Set). At least four outpatient visits (Outpatient Value Set) or observation visits (Observation Value Set) on different dates of service, with any diagnosis of Asthma (Asthma Value Set) and at least two asthma medication dispensing events (Table ASM-C). Visit type need not be the same for the four visits. At least four asthma medication dispensing events (Table ASM-C).

136 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Table ASM-C: Asthma Medications Description Prescriptions Antiasthmatic combinations Dyphylline-guaifenesin Guaifenesin-theophylline Antibody inhibitor Omalizumab Inhaled steroid combinations Budesonide-formoterol Fluticasone-salmeterol Mometasone-formoterol Inhaled corticosteroids Beclomethasone Budesonide Ciclesonide Flunisolide Fluticasone CFC free Mometasone Triamcinolone Leukotriene modifiers Montelukast Zafirlukast Zileuton Long-acting, inhaled beta-2 agonists Mast cell stabilizers Methylxanthines Short-acting, inhaled beta-2 agonists Arformoterol Salmeterol Cromolyn Aminophylline Dyphylline Albuterol Levalbuterol Formoterol Theophylline Metaproterenol Note: NCQA will post a comprehensive list of medications and NDC codes to by November 3, Step 2 Step 3: Required exclusions A member identified as having persistent asthma because of at least four asthma medication dispensing events, where leukotriene modifiers or antibody inhibitors were the sole asthma medication dispensed in that year, must also have at least one diagnosis of asthma (Asthma Value Set) during the same year as the leukotriene modifier or antibody inhibitor (i.e., the measurement year or the year prior to the measurement year). Exclude members who had any diagnosis from any of the following value sets, any time during the member s history through December 31 of the measurement year: Emphysema Value Set. Other Emphysema Value Set. COPD Value Set. Obstructive Chronic Bronchitis Value Set. Chronic Respiratory Conditions Due To Fumes/Vapors Value Set. Cystic Fibrosis Value Set. Acute Respiratory Failure Value Set.

137 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Administrative Specification Denominator Numerator The eligible population. Dispensed at least one prescription for an asthma controller medication during the measurement year (Table ASM-D). Table ASM-D: Asthma Controller Medications Description Prescriptions Antiasthmatic combinations Dyphylline-guaifenesin Guaifenesin-theophylline Antibody inhibitor Omalizumab Inhaled steroid combinations Budesonide-formoterol Fluticasone-salmeterol Mometasone-formoterol Inhaled corticosteroids Beclomethasone Budesonide Ciclesonide Flunisolide Fluticasone CFC free Mometasone Triamcinolone Leukotriene modifiers Montelukast Zafirlukast Zileuton Mast cell stabilizers Methylxanthines Cromolyn Aminophylline Dyphylline Theophylline Note: NCQA will post a comprehensive list of medications and NDC codes to by November 3, Data Elements for Reporting Organizations that submit HEDIS data to NCQA must provide the following data elements. Table ASM-1/2: Data Elements for Use of Appropriate Medications for People With Asthma Measurement year Data collection methodology (Administrative) Eligible population Number of required exclusions Numerator events by administrative data Reported rate Lower 95% confidence interval Upper 95% confidence interval Administrative For each age stratification and total For each age stratification and total For each age stratification and total For each age stratification and total For each age stratification and total For each age stratification and total

138 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Proposed Changes for HEDIS : General Guideline 28: Members Who Switch Products NCQA requests comments on proposed changes to the Volume 2 Technical Specifications General Guideline 28: Members Who Switch Products. NCQA proposes to count members as continuously enrolled if they switch among HMO, POS and PPO products. These members would be reported in the product in which they are enrolled at the end of the continuous enrollment period; enrollment in a PPO would no longer be considered a gap. Background General Guideline 28: Members Who Switch Products In 2004, NCQA began to phase in the requirement for PPOs to report HEDIS. Full implementation (including use of the Hybrid Method and reporting of the full measure set) was in HEDIS As reporting requirements progressed, open access choices grew, and many organizations offer a PPO in addition to the HMO and POS products. These organization are often seamless across product lines and use the same systems to track enrollment and pay claims. Additionally, where separate data systems are used, technology has progressed and it is easier to combine data from large, disparate systems. For these reasons, NCQA recommends that the barrier to treat PPO enrollment segments as a gap be removed from General Guideline 28: Members Who Switch Products. Supporting document for the proposed General Guideline changes: draft guideline for General Guideline 28: Members Who Switch Products. 1 HEDIS is a registered trademark of the National Committee for Quality Assurance (NCQA).

139 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Proposed Changes for HEDIS : General Guideline 28: Members Who Switch Products Membership Changes 28. Members Who Switch Products Measures with a continuous enrollment requirement If the organization reports separately by product, members who switch among HMO, POS and PPO products, or from the commercial HMO product to the commercial POS or PPO product (or vice versa), in the time specified for continuous enrollment for a measure are continuously enrolled and are included in the product-specific HEDIS report in which they were enrolled as of the end of the continuous enrollment period. For HMO or POS HEDIS reporting, count enrollment in a PPO as a gap in continuous enrollment. For PPO HEDIS reporting, count enrollment in an HMO or POS product as a gap in continuous enrollment. For NCQA-approved combined HMO/POS/PPO reporting, consider members continuously enrolled. The organization must use claims data from all products, even when there is a gap in enrollment. Enrollment in a Medicare Private Fee-for-Service (PFFS) plan is considered a gap in HMO/POS and PPO enrollment. Measures without a continuous enrollment requirement If the organization reports commercial HEDIS separately by product (i.e., HMO, POS, PPO), members who switch among products during the measurement year are reported in the product in which they were enrolled on the date of service (outpatient services) or the date of discharge (inpatient services). 1 1 HEDIS is a registered trademark of the National Committee for Quality Assurance (NCQA).

140 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Proposed Changes to Existing Measures for HEDIS : Relative Resource Use (RRU) NCQA seeks comments on proposed modifications to the suite of Relative Resource Use measures to improve the usability and timeliness of the measures for health plans, public reporting programs and consumers. Background NCQA s Relative Resource Use measures address consumer and purchaser value in health care. To support their efforts with timely access to valid and actionable data, we plan to modify our calculation of expected performance as described below for HEDIS Current Method Health plans currently report observed annual resource use by members with one of five chronic conditions: asthma, cardiovascular disease, COPD, diabetes and hypertension. Plans provide aggregated standardized spending or utilization data for cohorts of members based on combinations of age, gender and 13 levels of patient risk (based on comorbid conditions). NCQA calculates and reports each plan s expected resource use from the percentage of member months in each cohort and the average resource use reported by plans in those cohorts and by peer groups of: Reporting type: HMOs and PPOs. Geography: National and regional. Alternative Method Combine HMO and PPO. For several years, all the measures used in health plan accreditation have held plans to a common performance standard based on combined HMO and PPO data. Combining HMO and PPO plans into a common benchmark will bring these measures into alignment with the rest of HEDIS. Report national-only expected performance. NCQA already reports national-only performance for RRU (and for all HEDIS and CAHPS performance); these values have been used in Health Plan Rankings and will be used in the Marketplace Quality Rating System. A common scale means we can compare Florida and Minnesota plans directly, as we do for all other HEDIS measures. Regional results indicate relative performance in that group, but users can infer these differences from the national results. Eliminate the indexed O/E ratio. Currently, NCQA reports both the observed-to-expected (O/E) ratio (i.e., how many resources did the plan use, compared to what we would expect, given its own case-mix) and an indexed O/E ratio. Indexing means calculating the ratio of the plan s O/E to the average O/E. Consider a plan with an O/E value of 1.02 (using 2 percent more resources than expected given the plan s case mix), when the average of plans in its peer group is On the index scale, the plan appears as Its O/E is 2 percent better than the average plan s O/E, but: 1. We already know it is better than the average plan (and all other plans above 1.02) through inspection of the pre-indexed O/E value. 2. Indexing leads to confusion: 0.98 on the index scale means the plan s O/E is 2 percent better than the average plan, not that its resource use is 2 percent better than expected. We know that its resource use is actually 2 percent worse than expected for a plan with its case-mix. Use prior year expected performance (as opposed to current year). NCQA will calculate expected resource use from the prior year s average performance within each cohort and will continue to use the health plan s case mix from the measurement year. 1 1 HEDIS is a registered trademark of the National Committee for Quality Assurance (NCQA).

141 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Opens the black box on expected performance. Plans will know their average resource use, a year in advance, by cohort and overall, to use as performance targets. They will know their expected and O/E ratio performance results when they submit data to NCQA. 2. Speeds availability of performance results to plans and reporting programs. Prior-year average performance locks the reference standard for plans. When we use current-year data to define expected performance (as we have to date), every plan s performance changes any time a single plan resubmits data, because the average utilization changes (if only slightly). This slows the availability of the measures for report cards and open enrollment information that relies on the most recent HEDIS data. Analysis We modeled performance results of these changes using two time intervals (HEDIS , ) for all three product lines (commercial, Medicare, Medicaid) and all five RRU conditions. NCQA already reports national RRU results and raw O/E results, so we did not test removing those changes. We found: Combined HMO and PPO results are similar to separated HMO and PPO results. Plans performance was strongly correlated, although there were exceptions. Some PPOs will see their O/E rise and some HMOs will see theirs fall. To the extent that PPO spending exceeds HMO spending, on average, the cohort benchmarks will rise for PPOs and fall for HMOs, which accounts for observed deviations. Results based on prior-year expected performance and current-year performance are consistent. The O/E ratios for Total Medical and Total Pharmacy resource use were correlated above 0.9 for almost all measures, all product lines and both years.

142 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Announcement: Retirement for HEDIS : General Guidelines Measure Rotation NCQA would like to notify plans of the retirement of General Guidelines 12 16: Measure Rotation. We are not requesting comments on this decision, which was made by NCQA s Leadership Team and is based on the information below. Background NCQA has allowed scheduled rotation of hybrid HEDIS measures for more than 15 years (hybrid measures allow medical record review), but the increase in a plan s measure rate from chart review (hybrid lift) has declined. The hybrid lift in 2014 was far less than in Improved electronic and claims systems and better billing practices have resulted in more complete and accurate data. Most plans use supplemental data sources in areas where they previously relied on medical records. Additionally, over the last 10 years, measurement has become a larger part of the health care-quality conversation and many regulators (i.e., CMS for Medicare measures, Marketplace/QRS reporting, 19 state Departments of Insurance for commercial and Medicaid measures) do not allow rotation. Recently, several states asked NCQA s Public Sector Advisory Council to consider eliminating measure rotation, arguing that NCQA s measure rotation policy supports gaming by plans, violates transparency to data users and undermines quality improvement activities. The pool of potential measures that may be rotated includes only the 15 hybrid measures. Between 2010 and 2014, 39 percent of commercial plans and 11 percent of Medicaid plans chose rotation for at least one measure. Although rotation was originally meant to reduce the burden of data collection, NCQA s experience is that the policy is not applied by plans as intended, resulting in unplanned consequences: Rotation results in unequal burden and unequal opportunity to choose the best rate for organizations that must measure and organizations that choose to rotate. Organizations sometimes report a rotated rate to NCQA and a current rate to state regulators, confusing customers and regulators who use the measures. Reported results are a mix of performance data from the current and previous years, depending on how many plans choose to rotate, making interpretation of results difficult and complex. Rotation affects benchmarks and skews measure trending, which affects all organizations. 1 HEDIS is a registered trademark of the National Committee for Quality Assurance (NCQA).

143 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Retirement for HEDIS : General Guidelines Measure Rotation Measure Rotation 12. How Rotation Works To reduce the overall HEDIS reporting burden and allow allocation of resources to improvement activities, organizations may rotate select commercial and Medicaid hybrid measures and surveys on a biennial basis. Measure rotation allows an organization to use the audited and reportable Hybrid Method rate or survey from the prior year s data collection in lieu of collecting the measure for the measurement year. Each year, NCQA specifies a list of measures eligible for rotation. Measures may not be rotated in a year when they are not eligible for rotation. 13. Criteria for Rotation Eligibility The following criteria must be satisfied in order for the organization to rotate a measure: The measure is on the list of measures eligible for rotation in The organization has an audited and reportable rate from the prior year, produced using the Hybrid Method. The organization s reporting entity has remained constant since the preceding year. The organization had a small denominator audit result (NA for HEDIS 2014) that still applies (NA for HEDIS 2016). Note: The HEDIS Compliance Audit may include source code review for a selected core set of measures. Even if an organization chooses the rotation option, the certified auditor selects an appropriate core set (excluding rotated measures) and conducts the audit. 14. Measures Eligible for Rotation Measure rotation applies to the commercial and Medicaid product lines only. Organizations should defer to state regulatory agencies about individual state decisions regarding rotation. The following measures are eligible for rotation for HEDIS 2016: Controlling High Blood Pressure. Frequency of Ongoing Prenatal Care. Prenatal and Postpartum Care. Weeks of Pregnancy at Time of Enrollment. 15. Rotation and HEDIS Scoring for Accreditation A number of measures eligible for rotation are used for accreditation scoring. NCQA holds thresholds constant for rotated measures. Organizations may rotate measures and retain HEDIS results and accreditation scores or may collect and report rotated measures and update HEDIS results to increase their accreditation score. 1 HEDIS is a registered trademark of the National Committee for Quality Assurance (NCQA).

144 Draft Document for HEDIS 2016 Public Comment Obsolete After March 18, Measure Rotation and Data Submission Organizations must use the Interactive Data Submission System (IDSS) to indicate rotated measures and must submit data to NCQA by the HEDIS reporting deadline. Refer to the IDSS Users Guide for instructions on completing the IDSS.