Rate Review Process Table of Contents Summary of Findings...2 Background...2 Federal Health Insurance Reform Requirements in Puerto Rico...2 ACA Applicability to U.S. Territories...2 Rate Review Requirements...3 Current Puerto Rico Rating Requirements... 3 Carrier Rate Filing Requirements... 3 Rating Rules...3 Current Carrier Underwriting Policies...4 From the On-site Interviews... 4 From the Materials Provided... 4 Medical Card Systems... 4 First Medical... 5 Triple-S Salud... 6 Recommendations...7 Rate Filing Process... 7 Rate Review Process...7 Is the filing complete?...8 Are the rates actuarially supported?...8 Rates are considered discriminatory if they charge different amount for policies having essentially the same risks and expense elements. An actuarial value model can be used to determine the relative risks. If the premiums between policies with the same actuarial value differ by more than the difference in actuarial value +/- 10%, an actuarial determination must be made if the policies are discriminatory. It is possible that other factors are driving the discrepancy such as the relative health of the policy holders, but many of these factors will not be allowed starting in 2014. In Puerto Rico the information in the benefits map can also be used to determine the actuarial value. The actuarial value will be the paid claims divided by the allowed cost. Analysis of Claims Costs...10 A.Reasonableness of Base Period Experience...10 Base Period Selected...10 Unpaid Claim Liability Estimates...10 Large Claim Impact...11 1
B. Reasonableness of Projected Claims Trends...11 Documentation of Claims Trend Development... 11 Historic Trend Analysis...11 Provider Contract Detail...12 C.Additional Adjustments...12 Rate Factor Relativity Studies... 12 Enrollment Related Changes...12 Details of All Benefit and Formulary Changes, including Projected Cost of Changes...12 Analysis of Non-Claims Costs...12 A.Administrative Expense Trends... 13 Administration and Marketing Cost Trend Tests...14 Detailed Actual Historic Expenses...14 Detailed Budget Projections... 14 Budget Allocations...14 B. Solvency Level Tests and Gain/Loss Margin...14 Requirements for Margins... 14 Solvency levels... 15 Risk-based capital... 15 Contract and Policy Reserve Estimates Including Deficiency Reserves...15 Does the filing meet regulatory requirements?...15 Appendix A ACA Requirements...16 Benefit Requirements...16 Coverage Requirements...16 Rating Restrictions...17 Rate Review - Applicability...17 Rate Review Standards for Review... 17 Rate Review Preliminary Justification...18 Rate Review Required Reporting by States...18 Rate Review Procedures when an Increase is Determined to be Unreasonable...18 Appendix B Excel Instructions for Trend Analysis... 20 How to Create a Graph and Trendline in Excel... 21 Create a chart... 21 1. Worksheet data...21 3. Chart created from worksheet data...22 2
4. Modifying a basic chart to meet your needs...23 How to Add a Trendline and R-squared in Excel... 23 5. Add a Trendline...24 6. Display the R-squared value for a trendline... 26 3
Summary of Findings We examined the process currently used by the Puerto Rico Office of the Commissioner of Insurance (OCI) to review premium rate filings for health coverage submitted by carriers. We have identified some opportunities for improvement. The most significant of these are in the areas of: Standardizing the form and content of the data submitted by carriers to support premium rate filings to allow more consistent analysis and standardized decision rules concerning rate adequacy, excessiveness and discrimination that would make the process less subjective1; Standardizing the time period for the rate filing to be 60 days prior to effective date excluding any days needed to reply to an objection letter.; Providing clearer filing instructions to improve the quality and consistency of filings; Increasing communications during the rate review process; Documenting the review process and decisions made by the OCI concerning the rate filing2; and Providing automated support for rate reviews to make the analysis process more efficient. Our work contracts include: Standard Excel data forms as a required element in rate filings; Requiring carriers to complete and submit a standardized checklist with each filing; Improved filing instructions for carriers; Standard methodology for calculating the rate increase percentage3 A rate review manual for use by the OCI; and A rate filing database designed to support the review process. Background In order to understand our recommendations it is important to first understand the federal health care reform law, Patient Protection and Affordable Care Act (ACA), and how it applies to Puerto Rico as well as the requirements of the Puerto Rican health insurance law. We will provide a brief summary of both here with more detailed information on ACA in the Appendix. 1 We believe that standardizing the information provided in rate filings would also allow for more consistent information to be made available to the public, and thus make it easier for the public to understand and comment on rate filings. 2 The paragraph submitted to HIOS is sufficient for documentation. 3 See Puerto Rico Rate Filing Instruction Manual for a description of the rate filing forms]. 4
Federal Health Insurance Reform Requirements in Puerto Rico ACA Applicability to U.S. Territories The applicability of the various provisions of the ACA to U.S. territories is not always clear. However, based on the law and interpretations made by the Secretary of Health and Human Services: Based on HHS s interpretation, all sections of the ACA that amend the Public Health Services Act apply to the territories. This includes benefit requirements, rate review, MLR, guaranteed issue (beginning in 2014), and modified community rating.4 Provisions relating to exchanges apply to the territories. The individual mandate (beginning in 2014) does NOT apply to the territories. Low-income premium subsidies (beginning in 2014) do NOT apply to the territories. It is not clear whether reduced cost-sharing for low-income members (beginning in 2014) applies to the territories. Rate Review Requirements These requirements apply to states and territories, but for simplicity, the term state is used. The ACA requires the Secretary of Health and Human Services, in conjunction with States, to establish a process for reviewing unreasonable rate increases. That process requires issuers to justify unreasonable rate increases prior to implementation. The Secretary adopted a rule to establish the process. Because it is impossible to determine whether a rate increase is unreasonable without reviewing the basis for the increase, it was necessary to determine which filings would be subject to review. The rule establishes a threshold, above which increases are considered potentially unreasonable and therefore subject to review. For the 12 months beginning September 1, 2011, the rule established the threshold as 10%. Therefore if the weighted average5 rate increase for a contract, combined with any other increases that were implemented within the previous 12 months exceeds 10%, then it is subject to review. For future 12-month periods, state-specific thresholds are to be used. Guidance was recently issued regarding state-specific thresholds for the 12 months beginning September 1, 2012.6 For that period, a State may propose an increase or decrease in its threshold. Proposals must be submitted by May 1, 2012. CMS will review the proposals and provide States with its determinations by June 1, 2012. If no change is proposed or if a proposal is not accepted, the threshold remains at 10%. For states determined to have an effective rate review mechanism, the rule provides that a rate increase in excess of the threshold is unreasonable if the State determines it is excessive, unjustified, unfairly discriminatory, or otherwise unreasonable as provided under applicable State 4 See Section 8.050 of the Health Insurance Code of Puerto Rico effective February 2012. 5 The rate increases are weighted by the amount of premium for each category that has a rate increase. This is done my multiplying the rate increase by the total premium for each category and divide the sum of these products by the total premiums for all categories. 6 CMS will post a list of all pending State-specific threshold proposals on the Center for Consumer Information and Insurance Oversight website at http://cciio.cms.gov/. 5
law. CMS will adopt the State s determinations if, within five days following a State s final determination, the State submits its final determination and a brief explanation of the analysis that led to that determination to HHS via HIOS. A more detailed summary of these provisions as well as ACA provisions relating to benefits, coverage requirements, and rating restrictions can be found in Appendix A. Current Puerto Rico Rating Requirements Carrier Rate Filing Requirements A health insurance issuer must submit every rate manual,, as well as any other information concerning the application and computation of rates made and used by it, and every modification of any of the foregoing which it contemplates using 60 days prior to implementation for approval by the Office of the Commissioner. The Commissioner can then extend the period for an additional 60 days if needed. If at the end of this period the Commissioner has not disapproved the rate increase, it is marked as approved and can be used. The Commissioner has a right to withdraw a previous approval after a hearing. An issuer cannot use any rate increase that has not been approved by the Office of the Commissioner. Rate filings must include the maximum portions for commissions that the carrier must pay, as well as the portions that shall be designated to profits and other expenses incurred in the underwriting of the insurance. If the Commissioner considers these portions excessive, the filing may be disapproved. Every filing must have its proposed effective date and clearly indicate the type of coverage being provided. When a filing does not have sufficient information for the Commissioner to make a determination, he shall require the carrier to provide the information. The filing and any supporting information will be made public after the filing is made effective. No carrier shall charge any rate which deviates from the rates filed and subsequently approved. The Commissioner may audit carriers to ensure their compliance. Rating Rules Rates shall be made with the following provisions: 1. Basic classification, manual, minimum, class rates, rating schedules or rating plans shall be made and adopted 2. Rates shall not be excessive, inadequate, or unfairly discriminatory 3. No rate shall discriminate unfairly between risk involving essentially the same hazards and expense elements or between risks in the application of like charges and credits Current Carrier Underwriting Policies From the On-site Interviews Currently some carriers consider the medical experience and health of even small groups when quoting initial and renewal rates. Rates can vary significantly based on the anticipated relative health of a group. In Puerto Rico, rating based on the health of the insured is not allowed. MAPFRE currently has two plans (Excel 6000 and Choice 6000) each with various riders. To rate the plans they use the credibility of each group and the claims experience of the group. To 6
be considered fully credible a group most have over 250 members and have twelve months of experience. All of their small groups are non-grandfathered. Medical Card System s (MCS) main small group contract is MCS Global, which accounts for 99% of their book of business. They have two older small group contracts, which are cost sharing contracts. MCS also has an HMO contract, which they are currently not selling. MCS adjusts the rates for the entire block of business at one time, and does not adjust for each small group. They do not know how the 10% rate increase should be calculated. MCS increases their small group rates on a frequently throughout the year. MCS also offers five individual contracts. The individual rates are changed on a quarterly basis Humana has three series of plans for small group. They also, have three old contracts that they are currently not selling. When they increase rates they take the small group experience into consideration as a whole not on a particular group. Currently, Humana adjusts the rates on a monthly basis and the trends on a quarterly basis. To determine if the rate increase is over 10%, they review the pooled block in its entirety and not looking at the plans on an individual basis. Humana is currently not selling any new individual contracts, but does have 700 members covered in a contract that they are no longer selling. Triple-S currently has two small group contracts, each that has variations on co-pay and coinsurance. They also have a new contract in the development stages as of February 22, 2012. The majority of Triple-S plans are grandfathered. They also sell three individual contracts, one of which is intended for younger individuals. First Medical offers three small group contracts that are offered mostly to the government of Puerto Rico, government unions and municipalities. They also offer individual contracts and have not had a rate increase for their individual contracts for four years. Currently, First Medical reviews contracts as a group to determine if the increase is over the 10% threshold. From the Materials Provided Medical Card Systems As of April 2012, MCS continues to employ an Excel-based rating tool developed in-house to produce rate quotes for all new (and renewal) business for its small group block, defined as groups from 2-50 employees. Following is a brief description of the mechanisms employed in the tool to produce rates. Base Rates Base rates are updated on a semi-annual basis in order to keep pace with claims trend, targeting a medical loss ratio (MLR) of approximately 82%. Rating Factors Base rates are adjusted via rating factors for contract tier type (i.e. Individual, Couple and Family) and for contract design (e.g. copay and coverage variations) in medical, prescription drug, dental and vision care coverages. Rating Bands All groups are classified into one of nine (9) risk bands (from lowest to highest): Low 1 (L1), Low 2 (L2), Low 3 (L3), Medium 1 (M1), Medium 2 (M2), Medium 3 (M3), High 1 (H1), High 2 (H2), and High 3 (H3). Classification is based exclusively on the average demographic profile of participating employees, regardless of contract tier type and regardless of gender. Rate Quote Development 1. Employee census data is entered into the Pool Rating Tool. 2. Each employee is assigned an Employee Age Factor according to the applicable age interval. 7
3. A Group Risk Factor is calculated using the average Age Factor of all enrolled employees. 4. The Group Risk Factor determines which Rating Band applies for the Group (L1 to H3). 5. Contract selection in the Rating Tool determines the applicable factors for adjusting rates for the selected benefit design, producing final rates for Individual, Couple and Family contracts. First Medical First Medical indicates that its rates offered to all new and renewal groups, including groups of small and medium businesses (SMBs), use rating methods and practices based on actuarial assumptions that are widely accepted and in accordance with reasonable actuarial principles. The rates include a deviation of up to 15%. All premiums are payable on a monthly basis, guaranteed for 12 months unless the following changes: Membership of the employer of SMEs; Family composition of the eligible employee, or Health Plan benefits requested by the employer of SMEs. The selection process is to assess risk and classify the level of risk, assessing the factors or characteristics of the group including: Distribution by age and sex, family composition and % share; Location of business or industry and inflation; and Plan design and deductible levels. The renewal of each group focuses on the claims experience of the group as a whole and within the framework of the other groups assigned under the same conditions. Within this process of reevaluating the pricing of the group, First Medical will renew all eligible employees and their dependents, except in the following cases: (1) (2) (3) (4) (5) (6) (7) For nonpayment of premium, considering the period of grace; When the subscriber performs an act that constitutes fraud; When the subscriber has made an intentional misrepresentation of a material fact and material under the terms of the health plan; For failure to meet the minimum participation requirements set by the plan; For failure to comply with the requirements of employer contribution; When prescribed by the Commissioner of Insurance of Puerto Rico; and In cases where so provided in the Insurance Code of Puerto Rico. Triple-S Salud Rate Quote Development Triple-S Salud uses a StepWise rating tool to compute premiums. The process is as follows: All new groups require the following: 1. Census 2. Medical Questionnaire 3. Current Rates 4. Experience Report 5. Description of Current Benefits Evaluators validate all above documents are complete. They use a software package (StepWise) to determine premiums. StepWise uses group demographic information, and claims experience 8
data where applicable. The system generates the proposal which is reviewed by evaluator before sent to sales and marketing to begin the process of sale. 90 days before the renewal date of a group, assessors use StepWise to calculate new premiums. Experience data tables for StepWise are updated automatically based on files in the Data Warehouse. Renewal adjustment is based on demographics or group, experience, inflation, pharmacy coverage, medical and dental care. 60 days before StepWise renewal adjustments are reported to sales and marketing who prepare documents for renewal negotiations. Risk Assessment 1. Companies with 600 or more employees (Groups of Experience) a. Census if required for new groups and experience with previous carrier for a quote based on the group experience, and renewal based on experience. 2. Companies with 51 to 599 employees (Groups of Merit) a. Census is required for new groups and experience with previous carrier based on group experience b. Renewal fees are determined based on an average, credibility adjusted, among the group s experience and segment worthwhile overall experience. 3. Companies with 2-50 employees (Small Groups) a. All eligible groups are negotiated with effective community rates and adjusted for age, sex7, geographic area, size group plus any necessary adjustment for evaluation of individual health questionnaires Adjustments to Premium The premium charged per year cannot be adjusted more than once per-year-per-contract unless: 1. Changes in family composition 2. Changes in benefits requested by employer 3. Changes in membership of employer Recommendations Rate Filing Process Proposed rates should be filed at least 60 days prior to the proposed effective date. Rates should be filed with sufficient detail and with sufficient time to allow the Office of the Commissioner of Insurance to review the rates prior to the proposed effective date. A suspension will be effective when an objection letter is sent until the reply is received by the OCI. The rate filing should be prepared using assumptions and methodologies that conform to Actuarial Standards of Practice (ASOPs) published by the Actuarial Standards Board and actuarial practice notes published by the Academy of Actuaries. Rate Review Process Currently the OCI is only receiving information on rate increases in excess of 10%, except for HMOs, which file all rate changes. Since the OCI is not receiving information on rate increases under 10% for non-hmos, there is less information on the current medical trends in Puerto Rico. 7 The OCI may want to check that the use of gender has been stopped. 9
Because of this, it may be awhile until Puerto Rico specific trend statistics can be developed by the OCI. Therefore other sources of trend data will be needed to use in the review of rate filings. The rate review process is designed to efficiently determine if health insurance premiums are not actuarially supported or if premium rate increases that are unjustified or inadequate. The process should be as objective as possible avoiding subjective decision making. According to current Puerto Rican regulations rates shall not be excessive, inadequate, or unfairly discriminatory. The rate review process is intended to determine if the filed rates meet these requirements. Premium rates will be considered unreasonable if: The filing is incomplete; The assumptions and methods used are not actuarially supported (in the context of current Puerto Rico regulations premium rates are determined to be actuarially supported if the filing demonstrates, using reasonable and appropriate assumptions and methods, that they meet the minimum loss ratio requirements); The projected premium does not satisfy the federal rebate loss ratio threshold (MLR); The projected premium is unfairly discriminatory; or The projected premium is inadequate or undermines the future solvency of the health plan. The proposed rate review process consists of three distinct steps: 1. Review the rate filing for completeness, including supporting documentation that supports key elements of the filing; 2. Determine if the proposed rates are actuarially supported: 3. When the review is complete, finalize findings and prepare documentation of the review. A letter is written to inform the carrier of the results of the review and web information for public review is prepared and uploaded. For all rate increases 10% or more, the OCI will inform the Department of Health and Human Services (HHS) of the final determination along with the rationale for the final determination in the next five (5) days. Is the filing complete? The first step in the rate review process is to verify that the filing is complete. Any filing that is incomplete should be returned with a letter identifying the deficiencies in the filing and the SERFF status should indicate that it was rejected. The rate filing will be reviewed against the standardized filing checklist to verify that all of the required elements are included in the rate filing and that they have been completed in full. Attachments, including standardized and non-standardized spreadsheets, as well as descriptive information, should be reviewed for completeness. The standardized filing checklist to be filed by carriers will indicate 1) whether the specific elements are included, and 2) the attachment and location where the filing element can be found. Since standardized spreadsheet templates are required for the submission of some data and 10
assumptions, only the name of the spreadsheet template will be indicated on the standardized filing checklist. Supplemental information such as the description of the rating methodology and the reason for the rate increase request will be included in descriptive information and a reference should be provided. One of the completeness checks should be to determine if the rate increase percentage was calculated according to the instructions. The rate increase percentage should be calculated according to the standardized methodology. The rate increase percentage should not include changes in rating categories due to changes in ages of participants. However, it will include the effects of changes in benefits and changes in the relative health of the population, and the impact of these elements will be clearly identified by the attribution analysis discussed later. Another completeness check should be to verify the completeness and consistency of the federal filing on CMS s Health Insurance Oversight System (HIOS) website8. The consistency of the HIOS data and the rate filing received by the state should be verified. If inaccuracies are found in the HIOS data, the carriers and CMS should be notified. If the rate filing is found to be missing required elements, the carrier should be notified of the missing items and the rate filing should be deemed incomplete. Are the rates actuarially supported? Before rates can be tested for compliance with the regulations, they must be deemed actuarially supported. If they are actuarially supported they can be considered adequate, not excessive and not discriminatory. Insurance premium rates can be considered actuarially supported if they are developed using assumptions and methodologies that conform to ASOPs published by the Actuarial Standards Board and actuarial practice notes published by the Academy of Actuaries. For example, if a premium rate was developed using a claims trend factor that is not actuarially supported, it may be determined to be excessive when a more appropriate trend factor is used. The OCI rate review process will be able to use historic information to determine the most appropriate trend factors once there is sufficient data collected in the rate filing data base (IRIS). Before that, each carrier s historic trends can be used to determine the appropriateness of the trends in the filing (see Historic Trend Analysis). To determine if the rates are actuarially supported the documentation supplied with the rate filing will be reviewed. All assumptions will be reviewed to determine if they are reasonable based on recent local experience that is reported in other rate filings and collected by the OCI. Specific attention should be given to the assumptions that have the greatest impact on the rate increase. The review will begin with the identification of the main drivers for the increase in health care costs projected in the filing and used for the development of rates. To facilitate this, carriers will be required to provide an attribution analysis using a standardized exhibit that breaks down the proposed rate increase by source. This exhibit will be included in the standardized rate filing template. 8 The rate filing instructions indicate that carriers should send this information directly to the OCI, but if they do not, the information can be found on HIOS. 11
The percentage change for each source of increase will be compared to historical and current normative data for similar contracts9. The sources of increase included in the standard attribution analysis are: Difference between actual and expected benefit costs in the current rate year; Change in utilization of services; Change in unit cost of services; Legally required changes in benefits; Other changes in benefits; Changes in administrative costs; Changes in sales & marketing expenses; Changes in taxes, fees and assessments; Other changes in non-benefit expenses (margin or underwriting gain/loss). As rate filings are received they will be processed into a database. Eventually that database can be used to calculate statistical information on the data elements listed above as well as other data and information. Data will be reported on these drivers for other comparable company rate filings and contracts in Puerto Rico and for this set of contracts historically. Until the database has sufficient data, other potential benchmark information may include carrier surveys performed by the OCI. In addition to the standardized filing metrics, the OCI will review the actuarial memorandum, which describes the actuarial methodology used in developing the premium rates. The methodology used in the rate filing will be reviewed to determine if it was appropriate considering the contract design. The methodology includes, among other issues, the base period data used for claims projections, the method used to project the ultimate claim experience in the base period, the treatment of large claims, and any appropriate adjustments for deductible leveraging due to high levels of fixed cost sharing. In addition to reviewing claims projections, it is important to review non-benefit expenses including administrative costs and profit margins. If a loss ratio approach is used to determine non-benefit expenses, the resulting rates may be inadequate due to Puerto Rico s low health care costs. Therefore, it will be important to review information on current and projected administrative costs. If the information provided is not sufficient to determine if the rates are actuarially supported the OCI should request additional information. This additional information will typically be in the form of more detail support of assumptions that are driving the rate increase and that do not appear to be appropriate from the information provided in the initial filing. Additional reviews may result in multiple data requests and various levels of review depending on the particular area of concern. Potential areas for further review include: Base period experience; Reasonableness of medical trends and Rx trends that are used for projections (see B. Reasonableness of Projected Claims Trends); Enrollment related changes; Details of all benefit and formulary changes found in the contracts, including projected cost of changes; 9 National data can be found in reports by S&P Healthcare Economic Indices, which may not apply to PR http://www.standardandpoors.com/indices/sp-healthcare-economic-indices/en/us/?indexid=sp-healthcareeconomic-indices 12
Development of projected non-benefit expenses; Support for the gain/loss margin requirements provided in the rate filing (if the review extends beyond MLR compliance10). For each source of increase that cannot be verified as reasonable based on the information included in the filing and appropriate historical and current benchmarks, the reviewer should request additional supporting information. The additional information needed falls into two major categories: (1) Analysis of Claims Costs and (2) Analysis of Non-Benefit Expenses. If rates are not actuarially supported they may be excessive, inadequate or discriminatory. Rates are considered excessive if they are higher than the amount needed to pay for claims, administrative costs and a profit margin. Currently many regulators consider this level to equate to the ACA rebate loss ratio of 80% for individual and small group and 85% for large group. Rates are considered inadequate if they cannot pay for claims, administrative costs and a profit margin. A projected loss ratio of over 100%, is clearly an indication of rates being inadequate. If the projected claims plus projected administrative costs are more than the premium the premium is also considered inadequate. Rates are considered discriminatory if they charge different amount for policies having essentially the same risks and expense elements. An actuarial value model can be used to determine the relative risks. If the premiums between policies with the same actuarial value differ by more than the difference in actuarial value +/- 10%, an actuarial determination must be made if the policies are discriminatory. It is possible that other factors are driving the discrepancy such as the relative health of the policy holders, but many of these factors will not be allowed starting in 2014. In Puerto Rico the information in the benefits map can also be used to determine the actuarial value. The actuarial value will be the paid claims divided by the allowed cost. Analysis of Claims Costs Benefit costs start with credible base period experience. This base period experience is then trended to the projection period. Once the projection period claims expense is determined adjustments may be made for a number of conditions including: Changes in rating factors; Changes in enrollment; and Changes in benefits. A. Reasonableness of Base Period Experience The reasonableness of the base period includes the selection of the actual period to use and the adjustments to the base period. Depending on the specific situation issues may include: Base period selected; Unpaid claim liability estimates; and Large claim impact. 10 If carriers depend on the MLR rebate loss ratio, they will not provide information on non-benefit expenses or margins. In that case the OCI will not be able to review these. 13
Base Period Selected The selection of the base period for rate development is always a compromise between having more recent data and reducing the percentage of the incurred claims that are estimated and therefore subject to error. If the base period is too long before the projection period, it may not be relevant to recent cost and utilization and therefore not appropriate for use in estimating the claims cost in the projection period. On the other hand, if the base period is too recent then the paid claims may not be complete and an estimate will have to be done of the claims that have been incurred and not paid. This estimate may distort the actual incurred claims. Unpaid Claim Liability Estimates Unpaid claim liabilities may be needed and will impact the calculation of the base year data. When the base period is significantly prior to the point in time when the rates are being developed, there may not be a need to estimate the unpaid claims. When the base period is closer to the point in time when the rates are being developed, there may not be sufficient run out11 and the actuary will have to estimate the amount of incurred claims in the base period that have not been paid. If the base year claim costs are not correct, the projected claim costs and projected loss ratio are not actuarially supported. If, during the preliminary review, the completion factors12 reported for the base period experience appear inconsistent in current and local payment patterns13, the reviewer must request additional information on the unpaid claim liability (UCL). The unpaid claim liability is the largest estimate affecting the base claim cost estimate. To review the UCL, some or all of the following should be requested, depending on the specifics of the rate filing: Claims triangles14 and membership; Documentation of the UCL calculation performed by the company; Past UCL estimates and the retrospective calculation of the past UCLs; and/or Any information that was used in development of the base, such as new claims processing system implementation, slow down or speed up of claims submissions, changes in provider reimbursement methods during the base period, etc. The review of the UCL may consist of a recalculation of the UCL using the claim triangles and standard actuarial methodologies or a review of the UCL documentation. If it is determined that the UCL is not correct, the base year data and the resulting rate increase will need to be adjusted. Past UCL estimates will indicate the level of accuracy typically found in the carrier s UCL estimates. Overall historic UCL accuracy by market can be reviewed by looking at the carrier s 11 Run out refers to the claims that are paid after the base period that are for health care services received in the base period. 12 Completion factors are the factors used to estimate the amount of total incurred claims based on the paid claims. These factors are developed based on historic payment pattern from the date of service until claims are paid by the carrier. The average completion factors are provided in the written documentation filed by carriers. 13 Current completion factors can be found in other rate filings or by the use of carrier surveys. 14 Claims triangles contain the data used to calculate completion factors. They consist of paid claims for each month separated by month of service. 14
financial statements. If margins vary significantly it may be an indication of a poor estimation methodology or a conscious decision to change margins. Either way, it will add to any concerns in the estimate of the UCL impacting the base year data. There are other potential changes to the base period data including adjustments for seasonality and credibility. Seasonality adjustments may be needed if the base period is not full years. The adjustment is needed because the amount that a carrier is responsible for varies during the year. Early in the year the deductible reduces the carrier responsibility and later in the year when deductibles have been paid, the carrier is responsible for paying a higher percentage of the covered health care costs. For any base period that does not consist of full calendar years, the actuarial documentation should explain how this has been adjusted for. If that documentation is not provided, the OCI should request it. For blocks that are small and not fully credible, the base period may be either blended with other credible data or it may be based on other data in total. The other data used should be from a comparable population and set of benefits. If the appropriateness of the data used when the contract experience is not credible cannot be established based on the documentation provided saying that the data was for a similar population and similar benefits, the reviewer should request additional documentation of the data used. Large Claim Impact Unusually high or low levels of large claims in different time periods can affect historical trend experience and thus trend assumptions. If high levels of large claims fall in the base period experience for the filing, the trend may look significantly higher than the underlying trend and actually is distorting projected costs. Some carriers smooth the impact of statistical fluctuation in large claims by purchasing reinsurance or through internal pooling. Internal pooling eliminates the large claims from the base period experience and replaces them with an average large claim amount or pooling charge that is added to the base period experience. Carriers that do not use either of these techniques may make explicit adjustments for large claims, if they believe the base year has an unusually high or low number of large claims. Explicit adjustments are made by reducing base year incurred claims in years of unusually high number of large claims or by increasing incurred claims in years of unusually low number of large claims. If the rate filing shows an unusually high or low level of large claims compared to prior year rate filings, documentation of the treatment of large claims should be reviewed. If adjustments for large claims does not appear to have been made, request that the carrier to provide an explanation. B. Reasonableness of Projected Claims Trends If the trend appears to be above or below other trends seen in the market15, the reviewer should request a quantitative development of the trend including historic information used and methodology for projecting past experience. 15 Until a sufficient database is developed of trends, a carrier survey may be needed to understand recent trends in Puerto Rico 15
Documentation of Claims Trend Development If the increase in premiums attributable to the change in unit cost of services or to the change in utilization of services appears, on initial review, to require additional support, the reviewing actuary should request: Documentation for the development of the trend assumptions16, which should include a description of any circumstances that would result in future trends deviating from past trends and how these factors were developed; Monthly allowed and paid claim utilization and cost trends, for each service category that appears unusual compared to what would be expected, for the last three years; and Utilization rates and unit costs, by major service category (medical, prescription drug, dental etc.), for the last three years. If the drivers are due to the utilization of services, review changes in utilization in other PR rate filings to determine if the increases are reasonable. PR rate filings include current utilization increases and cost of service increases filed by other carriers. If the drivers are utilization of services, review any special explanation provided in the rate filing for unusual utilization increases. If the increases in utilization for this filing are more than 10% higher than other rate filings in PR, ask the carrier to explain the high utilization increases. If the drivers are the cost of services, review research materials and the database metrics to determine if the increases are reasonable. Research materials include rate filings in other states and health department data. Database metrics include current cost increases identified by other carriers. Also, review any special explanation provided in the rate filing for unusual cost increases, such as changes to provider contracts. If the projected cost trends are more than 10% higher than other rate filings, ask the carrier to explain the high utilization increases Historic Trend Analysis Typically trend assumptions are based on a combination of historic trends and assumptions of how the future trends will differ from past trends. Health care cost trends vary significantly from year-to-year, but past trends might give an indication of the potential direction and magnitude of future trends. If any of the trends that are driving the rate increase appear to be more than 10% higher or lower than other rate filings, request the historic trend information from the carrier for the last 5 years. Once the historic information is received17: Use the Microsoft Excel trend line function (or another equivalent software function) to develop a trend on the data to determine a historic trend; Review the R-squared value18 associated with the trend line to decide how good of a fit the regression is to the data; and Review the documentation to determine any factors that would result in future trends deviating from past trends. 16 See Provider Contracting for how to consider provider contracts and potential additional carrier requests 17 See Appendix B for detailed instructions on how to perform these Excel functions. 18 An R-squared value close to 1 (over.5) shows that the trend line is very close to the historic data and that the data shows a clear pattern. 16
Past trends may not be a good indicator of future trends, but they can provide useful information. If past trends are fairly stable they may be better predictors of future trends. If there are changes in the market such as changing benefit designs, changing provider contracts or changes in how providers provide services, future trends will differ from past trends. In that case the actuary must estimate how much these changes will impact future trends. Use Excel s trend line function (or another equivalent software function) to calculate trends for each set of data points. Review the R2 determination feature of the software to identify the R2 for each set of data19. Note that the R2 only measures how close the trend line matches the data; it does not measure how well past history can predict future events. The predictive value20 of past history is dependent on how similar all pertinent elements (such as provider contracts and patient behavior) will be in the future compared to the past. Provider Contract Detail One of the causes of future trends being different from historic trends is unusual changes in provider contracting. If medical cost trends appear significantly higher than those seen from other carriers or significantly higher than the carrier s own historic trends, ask what unusual circumstances may be driving the unusual trends (e.g. changes in provider contracting). Request the carrier provide an explanation of any changes in provider contracts that would result in unusual trends. Determine if the information provided by the carrier explains the unusual trends because provider fees are increasing or provider discounts or rebates are being reduced.. C. Additional Adjustments There are a number of other situations that may impact the projected trends from the base period to the rating period. Rate Factor Relativity Studies Some rate filings may include changes to rating factors such as age and geography. If the resulting rate factors are significantly different than those of other carriers with similar contracts, request information on the development of the rate factors including: Date of last modification to rate factors; Definition of the historic data used for rate factor development; and Demonstration of the rate factor development. Enrollment Related Changes If the health status of the population is changing, the projected claim costs will be expected to change accordingly. If the carrier makes a separate adjustment for the change in enrollment of more than 3%, request that the actuary provides a detailed demonstration of the change and a certification that it is actuarially sound. For example, if changes in the market lead the carrier to expect its contracts to attract a less healthy population, it may make a special adjustment for the expected increase in claims cost. Care must be taken to ensure that any separate health status adjustment is appropriately coordinated with the trends factors to avoid double-counting. Double counting will occur if some of the population change has already started impacting the historic 19 For more details see Appendix B. 20 For more details see Appendix B. 17
trends. Ask the carrier s actuary to certify that the change has not been double counted and is not in historic trend data. Details of All Benefit and Formulary Changes, including Projected Cost of Changes If the rate filing contains significant changes due to benefit changes, request information describing those changes and documentation for the development of the cost impact of the changes. If the change is due to new mandated benefits, compare the carrier s estimate of rate impact to that of other carriers with similar new benefit requirements. Also review the information supplied by the carrier and determine if the estimated increase is appropriate. If not, determine a more appropriate increase and the impact on the total rate increase. Request that an actuary experienced in rate development review the documentation and determine if it is appropriate. If the change is due to other benefit changes, review the documentation and determine if the rate changes are appropriate compared to the benefit changes. Use the actuarial value model provided as part of the project to measure the difference in actuarial value of different benefits such as different deductible or cost sharing levels. The actuarial value model can be used to verify that rate changes due to benefit changes are within an expected range. If the rate increase is larger than the difference in actuarial value, ask the carrier to explain why. If the benefit changes are not in one of the categories covered by the actuarial value model, request that the carrier provide their detailed calculation of the increase and have a qualified actuary review the documentation. For prescription drug changes, request information on the impact of changes to the formulary which may result in an anticipated shift in the generic/brand mix or changes to the brand cost trend. Request that an actuary experienced in rate development review the documentation and determine if it is appropriate. Analysis of Non-Claims Costs Using a MLR standard alone as a measure of justified rate level, administrative costs and margins may increase at the same rate as in health care claims costs. This often results in excessive increases in the amount allowed for administrative costs and profits. In Puerto Rico the situation is very unique since the cost of health care is so low. In fact using a MLR standard may result in too little being available for administrative cost and profits. We recommend that the OCI monitor the appropriateness of the administrative cost and margins compared to the cost of living and other corporate return on capital requirements in Puerto Rico. The amount available from an 80% MLR for administrative cost may not be sufficient resulting in an inadequate premium rate. Depending on the specific regulatory goals, it may still be appropriate to ensure that non-benefit expense loads for administrative expenses and margins not only meet the federal MLR rebate requirements, but are consistent with the true cost to the carrier of administering the contracts and assuming risk for which the premiums will be charged, In other words, it is possible that projected premiums and health care claim costs result in loss ratios that are sufficient to not require a rebate, but are still not actuarially supported or may be inadequate. Especially in Puerto Rico, a loss ratio of 80% with the low cost of health care may result in premiums not being sufficient to pay administrative costs. The 20% of premiums or 25% of claims costs is less than it is in other states with high health care costs, but the administrative expenses in Puerto Rico are not that much lower than those in other states since computers, software etc. cost the same. 18
If trends in administrative costs are more than the trends in the cost of living in Puerto Rico, request carriers to explain. It is possible that required capital investments are driving the costs, but carriers should be able to explain higher trends. A. Administrative Expense Trends Compare administrative and marketing cost trends and PMPM costs to past trend estimates and PMPM costs for this carrier for the following categories (note: some of the detailed categories may or may not be available in the documentation): Quality Improvement Expenses Administrative Expenses Claims Adjudication Expenses Other Direct Administrative Expenses Indirect Expenses/Corporate Overhead Allocations Sales and Marketing Expenses Percentage of Premium Agent/Broker Commissions Other Agent/Broker Commissions Marketing Expenses Other Sales-Related Expenses State Taxes, Licenses and Fees State Premium Taxes State Assessments on Premiums for Charity Care State Corporate Business Tax Other State Taxes, Fees and Assessments Federal Taxes and Fees Federal Income Tax Other Federal Taxes, Fees and Assessments If non-claim trends are more than 2% percent above those of other carriers, more than 2% above the Puerto Rico cost of living, or the PMPMs are 5% percent above or below those of other carriers, review any special explanations. If there is not appropriate actuarial support, request further information and documentation. Administration and Marketing Cost Trend Tests Depending on the item from the list above with an unusually high trend, a number of potential data requests and analyses are possible. Potential additional reviews include: Detailed actual historic expenses for the item in question; and Budget allocations that may result in a higher amount of the item in question being allocated to the product. Detailed Actual Historic Expenses A detailed analysis of actual historic expenses facilitates determination of specific events or initiatives that impacted historic expenses but may or may not impact future expenses. Since carriers often allocate administrative expenses at a corporate level, documentation for administrative expenses should be requested at the corporate level (see Budget Allocation below). Elements necessary for this review include actual expense, actual premium revenue and corresponding membership by product and detailed administrative expense category (such as salaries, benefits, customer service, information systems, claims administration, commissions, and postage). The elements should also be summarized by high-level administrative expense 19
category (sales and marketing expenses, other administrative expenses, taxes, expenses and fees). Ideally, other administrative expenses would be separated into direct and indirect expenses. Information supplied by the carrier should include a numeric illustration and a narrative of the specific allocation methodology between market segments (large group, small group, individual, Medicare, etc.) contracts, geographic area, etc. for indirect expenses (such as overhead expense). Per-member-per-month expenses and expenses as a percent of revenue should be calculated and compared to industry normative data. Normative data includes other carrier data and administrative benchmark information such as that provided by the Sherlock Company21. The administrative benchmarks provide the typical health plan administrative expense spending, which can be compared to the projected spending in the rate filing. Additional support should be requested for administrative expense categories that compare unfavorably with normative data. Detailed Budget Projections Many carriers base administrative expense projections on budgeted administrative expenses as opposed to historic actual expenses. Carriers should be able to explain why budgeted expenses may differ significantly from actual. As with historic expense data, elements necessary for this review include budgeted expense, budgeted premium revenue and corresponding membership by product and detailed administrative expense category (such as salaries, benefits, customer service, information systems, claims administration, and postage). Projected per-member-permonth expenses and expenses as a percent of revenue should be calculated and compared to industry normative data. Normative data includes other carrier data and administrative benchmark information, such as that provided by the Sherlock Company22. A detailed review of budget projections is most useful in conjunction with a detailed review of actual historic expenses in order to enable determination of specific events or initiatives that impacted historic expenses but may not be anticipated to continue. Adjustments should be supported by the documentation provided. In addition, any trend factors used to adjust historic administrative expenses should be compared to publicly available information such as the CPI23. Additional support should be requested for administrative expense categories that compare unfavorably with normative data. Budget Allocations The allocation methodology between market segments, contract, and geographic region should be reviewed to make sure the methodology is reasonable, that expenses are being appropriately allocated, and that the allocation is not discriminatory. For example, corporate overhead expense should be allocated on a proportional basis across all contracts and market segments as opposed to a subset. The unit used to allocate expenses should be one that drives the cost. For example, allocation of the claim paying area should be allocated by claims count or alternatively by claims 21 http://www.sherlockco.com/ 22 http://www.sherlockco.com/ 23 Recent information on the Consumer Price Index can be found at http://www.bls.gov/cpi/ 20
volume, which drives the need for claims processors rather than by member months, since different products have different claims per member per month. Expenses may be allocated in several ways: percent of premium, number of claims, number of members, etc. The most appropriate allocation methodology will depend on the type of expense. For example, claim adjudication expense is most commonly allocated based on number of claims whereas corporate overhead may be allocated based on premium or membership. Detailed numeric support and a descriptive narrative of the allocation methodology should be requested of the carrier. Review the support for appropriateness and reasonableness. Errors or inappropriate allocations should be identified and an assessment should be made of their impact. B. Solvency Level Tests and Gain/Loss Margin Requirements for Margins Gain/loss margins are needed to: Provide for solvency protection; Provide a return on investment; and Build capital for capital expenditures. To review gain/loss margins, we recommend an analysis of other carriers in the market and the solvency level of the company based on their risk-based capital percentage. When doing this analysis consideration could be made of the for-profit or not-for-profit status of the carrier24. Solvency levels Gains are often needed to increase capital for solvency protection. Carriers may request higher rate increases due to reported low surplus levels using the logic that they need to increase surplus to protect solvency. If the carrier is requesting a higher rate increase due to solvency protection, then perform the following analysis. Risk-based capital Risk-based capital (RBC) is a measure of the capital needed for solvency protection. RBC is defined by a formula developed by the NAIC25. The NAIC formula result is the Authorized Control Level RBC (ACL), which is then compared to the capital level of the carrier. The resulting RBC percentage can be impacted by either a change in the ACL or by the capital level reported in the carrier s financial statement. The ACL is primarily driven by claim levels, which increase over time due to health care trends. The capital level reported in the financial statement is the result of accounting treatment of all of the financial statement amounts including company liabilities. The reason that this is important is that many company liabilities are estimates and can be subject to distortion. If the RBC level26 is low (under 350%27), the margin percentage may be higher than other carriers margins in order to build up surplus levels. If the RBC level is high (over 900%), the margin percentage may be lower than other carriers to adjust for past margins. Therefore, the surplus level should be considered when reviewing margin levels. 24 Usually this information can be found on a carrier web site if they are not-for-profit. The information is also available in the carrier s articles of incorporation. 25 The formula and software to perform the calculation can be purchased from the NAIC 21
Contract and Policy Reserve Estimates Including Deficiency Reserves Contract and policy reserves are estimated liabilities that reduce the capital levels of the company. If these reserves are set too high, then the surplus levels of the company will be artificially low. Therefore, any rationale provided that margins are needed to protect solvency due to low surplus levels may be unsubstantiated. In health insurance companies, deficiency reserves, which are reserves that are held to cover future projected losses, are the largest estimate that can cause inappropriate reductions in surplus. If a carrier is justifying part of a rate increase by indicating solvency issues, review the financial statement for a deficiency reserve28. If there is a deficiency reserve, request documentation of the reserve estimate and verification that the reserve was not considered in calculating the base claims experience. Review the deficiency reserve documentation to verify that projections of costs are actuarially appropriate and reflect likely expected losses. The NAIC Health Reserve Guidance Manual can be purchased from the NAIC and contains information on the estimation of a deficiency reserve and the construction of appropriate assumptions. If the deficiency reserve is not based on the instructions provided in the NAIC Health Reserve Guidance Manual (whether insufficient or excessive), ask the carrier s actuary to explain the discrepancies. Does the filing meet regulatory requirements? Once rates are deemed to be actuarially supported, they can be reviewed against the regulatory requirements. Unfortunately, it is possible for rates to be actuarially sound and not meet regulatory requirements. To determine if rates are excessive or inadequate, the development of the proposed rate should be actuarially sound. If the proposed rate is based on accepted actuarial assumptions and methodology, it can be considered to not be inadequate or excessive. In Puerto Rico if the administrative costs are developed according to actuarial methods, the loss ratio may be less than the federal rebate formula. If the federal rebate loss ratio29 is used to develop the rates, they may be inadequate to cover the expected claims and the administrative expenses. 26 The RBC percentage for the current and past years can be calculated from the information on page 28 of the Health or Orange blank or page 22 of the Life or Blue blank (Five-Year Historical Data) by dividing the Total adjusted capital on line 14 (line 30 of the Life blank) by the Authorized control level risk-based capital on line 15 (31 of the Life blank). 27 There is no agreement on the percentages of RBC that are considered low or high, but we have provided some percentages to use as a guideline until the OCI determines its own benchmarks. 28 Deficiency reserves are reported on page 13 line 2 of the health statutory annual statement and page 38 (Schedule H) of the life statutory statement. 29 The current ACA rebate loss ratio id 80% for individual and small group policies and 85% for large groups. 22
Appendix A ACA Requirements Premium Development30 To determine rates for a specific insurance contract, the issuer must estimate future claims costs in connection with that contract and then the revenue needed to pay anticipated claims and nonclaims expense, such as administrative expenses including profits. Because costs and revenue are not known, these rates must be based on actuarial estimates of these costs and of non-claims expenses. These estimates, as well as the methodology used to determine them, are the subject of actuarial review conducted by States that have authority to review premium or rate increases. Premiums are then determined by various factors (age, geography, etc.) as well as based on choices made by the insured (deductible level, co-pays, etc.). The criteria that may be used and the differences in premium that may be charged are determined by State law. Benefit Requirements A group health plan and a health insurance issuer offering non-grandfathered group or individual health insurance coverage may not establish lifetime limits on the dollar value of benefits or unreasonable annual limits on the dollar value of benefits.31 A group health plan and a health insurance issuer offering non-grandfathered group or individual health insurance coverage shall, at a minimum provide coverage for and shall not impose any cost sharing requirements for: 1. Evidence-based items or services that have in effect a rating of A or B in the current recommendations of the United States Preventative Services Task Force; 2. Immunizations that have in effect a recommendation from the Advisory Committee on Immunization Practices of the Centers for Disease Control and Prevention with respect to the individual involved; 3. With respect to infants, children, and adolescents, evidence-informed preventative care and screenings provided for in the comprehensive guidelines supported by the Health Resources and Services Administration; 4. With respect to women, such additional preventative care and screenings not described in paragraph (1) as provided for in comprehensive guidelines supported by the Health Resources and Services Administration for purposes of this paragraph; and 5. For the purposes of this Act, and for the purposes of any other provision of law, the current recommendations of the United States Preventative Service Task Force regarding breast cancer screening, mammography, and prevention shall be considered the most current other than those issued in or around November 2009.32 30 Notice of Proposed Rulemaking - Rate Increase Disclosure and Review, Federal Register Vol. 75, No. 246, Sect. II-A. 12-23-2010. Note: Although this is from the Notice of Proposed Rulemaking and not the final rule, it provides background for the rule and is equally applicable to the final rule. 31 Public Health Service Act (PHSA), Title XXVII, Part A, Section 2711, enacted by Sections 1001 of the ACA. 32 Public Health Service Act (PHSA), Title XXVII, Part A, Section 2713, enacted by Sections 1001 of the ACA. 23
A health insurance issuer that offers health insurance coverage in the individual or small group market shall ensure that such coverage includes essential benefits.33 Coverage Requirements A group health plan and a health insurance issuer offering group or individual health insurance coverage that provides dependent coverage of children shall continue to make such coverage available for an adult child (who is not married) until the child turns 26 years of age.34 A group health plan or health insurance issuer offering group or individual health insurance coverage may not impose any preexisting condition exclusion with respect to such plan or coverage for dependents to age 19 and as of 2014 for all plans.35 Each health insurance issuer that offers health insurance coverage in the individual or small group market in a State must accept every employer and individual in the State that applies for such coverage and may only restrict enrollment to open or special enrollment periods.36 A group health plan or health insurance issuer may not establish rules for eligibility based on the following factors: 1. Health status; 2. Medical condition (including both physical and mental illness); 3. Claims experience; 4. Receipt of health care; 5. Medical history; 6. Genetic information; 7. Evidence of insurability (including conditions arising out of acts of domestic violence); 8. Disability; and 9. Any other health status-related factor determined appropriate by the Secretary.37 Health Insurance plans are guaranteed renewable.38 Rating Restrictions Premium rates shall vary with respect to the particular plan or coverage involved only by: 1. Whether such plan or coverage covers an individual or family; 2. Rating Area; 3. Age, except that such rate shall not vary by more than 3 to 1 for adults; and 33 Public Health Service Act (PHSA), Title XXVII, Part A, Section 2707, enacted by Sections 1201 of the ACA. 34 Public Health Service Act (PHSA), Title XXVII, Part A, Section 2714, enacted by Sections 1001 of the ACA. 35 Public Health Service Act (PHSA), Title XXVII, Part A, Section 2704, enacted by Sections 1201 of the ACA. 36 Public Health Service Act (PHSA), Title XXVII, Part A, Section 2702, enacted by Sections 1201 of the ACA. 37 Public Health Service Act (PHSA), Title XXVII, Part A, Section 2705, enacted by Sections 1201 of the ACA. 38 Public Health Service Act (PHSA), Title XXVII, Part A, Section 2703, enacted by Sections 1201 of the ACA. 24
4. Tobacco use, except that such rate shall not vary by more than 1.5 to 1.39 Rate Review - Applicability Section 1003 of the ACA adds a new Section 2794 to the Public Health Service Act (PHSA) directing the Secretary, in conjunction with the States, to establish a process for the annual review of unreasonable rate increases in. States are defined by the PHSA to include each of the several States, the District of Columbia, Puerto Rico, the Virgin Islands, Guam, American Samoa, and the Northern Mariana Islands. The Secretary adopted a rule40 to establish the process. Under the rule, rate increases filed on or after September 1, 2011, or effective on or after September 1, 2011 in a State that does not require a rate increase to be filed, are subject to review if the weighted average rate increase, alone or in combination with prior increases in the preceding 12 month period, exceeds a threshold. The weighted average is calculated as follows: For each subcategory of enrollees subject to the same rate increase, multiply the number of enrollees by the respective rate increase. The contracts are then summed over all subcategories. The sum is then divided by the total number of enrollees to arrive at the weighted average rate increase. The level of aggregation that applies when determining the weighted average rate increase is at the contract level. The rule defines contract as a package of health insurance coverage benefits with a discrete set of rating and pricing methodologies that a health insurance issuer offers in a state. For the first 12 months, the rule establishes a threshold of 10%. For future 12-month periods, state-specific thresholds are to be used. Guidance was recently issued regarding state-specific thresholds for the 12 months beginning September 1, 2012.41 For that period, a State may propose an increase or decrease in its threshold. Proposals must be submitted by May 1, 2012. CMS will review the proposals and provide States with its determinations by June 1, 2012. If no change is proposed or if a proposal is not accepted, the threshold remains at 10%. The cut-off between small and large groups is 50 employees, but states have the option of using 100 employees. Beginning in 2016, the cut-off will be 100 employees in all states. HHS would defer to the definitions used under applicable State rate filing laws when determining whether a rate filing relates to health insurance coverage offered in the individual market, small group market, or large group market. For example, the rule does not require that a State with an Effective Rate Review Program review proposed rate increases for short-term limited duration coverage if the State s rate filing law does not consider short-term limited duration coverage to be individual market coverage. 39 Public Health Service Act (PHSA), Title XXVII, Part A, Section 2701, enacted by Sections 1201 of the ACA. 40 45 CFR Part 154. See http://cciio.cms.gov/resources/files/rate_increase_final_rule.pdf. 41 State-Specific Threshold Proposals Guidance for States. See http://cciio.cms.gov/resources/files/rrjssptguidance.pdf. 25
Under the regulation, rates in the large group market, grandfathered plans, and excepted benefit plans are not subject to the rate review process.42 Excepted Benefit plans are described in the Public Health Service Act subsection (c) paragraphs (1) through (4). Rate Review Standards for Review For states determined to have an effective rate review mechanism, the rule provides that a rate increase in excess of the threshold is unreasonable if the State determines it is excessive, unjustified, unfairly discriminatory, or otherwise unreasonable as provided under applicable State law. For rate filings reviewed by CMS, the rule provides that a rate increase in excess of the threshold is unreasonable if it is excessive, unjustified, or unfairly discriminatory as defined below: A rate increase reviewed by CMS is excessive if (1) the rate increase results in a projected medical loss ratio that is below the applicable Federal MLR standard, (2) one or more of the assumptions on which the rate increase is based are not supported by substantial evidence, or (3) the choice of assumptions or combination of assumptions on which the rate increased is based is unreasonable. A rate increase reviewed by CMS is unjustified if the issuer provides incomplete, inadequate, or otherwise does not provide a basis upon which the reasonableness of an increase may be determined. A rate increase reviewed by CMS is unfairly discriminatory if it results in premium differences for a particular contract between insureds within similar risk categories that are not permissible under State law or, if no State law applies, do not reasonably correspond to differences in expected costs. Rate Review Preliminary Justification For increases that are subject to review, the issuer must submit a preliminary justification to the State in which the increase is proposed to be implemented, as long as a State accepts such submissions, and to HHS. The preliminary justification consists of three parts. Health insurance issuers are required to complete part I and II regardless of whether the rate increase is being reviewed by the State or by HHS. The information in parts I and II is intended to provide consumers with a description of the rate increase and the factors contributing to the increase, including both a descriptive and a quantitative analysis. For filings reviewed by HHS, Part III is also required. It includes additional information modeled on the actuarial memorandum guidelines included in the NAIC Model regulation 134-1. Part I is a rate increase summary and requires the following: 1. Historical and projected claims experience; 2. Trend projections related to utilization, and service or unit cost; 3. Any claims assumptions related to benefit changes; 4. Allocation of the overall rate increase to claims and non-claims costs; 5. Per enrollee per month allocation of current and projected premium; 6. Three year history of rate increases for the contract associated with the rate increase; and 42 45 CFR Part 154 (Health Insurance Issuer Rate Increases: Disclosure and Review Requirements), Section 154.103. 26
Part II of the preliminary justification is a written description justifying the rate increase and must include a simple and brief narrative describing the data and assumptions that were used to develop the rate increase and must include: 1. An explanation of the most significant factors causing the rate increase, including a brief description of the relevant claims and non-claims expense increases reported in the rate increase summary; and 2. A brief description of the overall experience of the contract, including historical and projected expenses, and loss ratios. Part III of the preliminary justification is rate filing documentation and, when required, must be sufficient for CMS to conduct its review as specified in the rule. Rate Review Required Reporting by States CMS will adopt the State s determinations regarding whether rate increases are unreasonable only if, within five days following a State s final determination, the State submits to HHS, on a form and in a manner prescribed by HHS, its final determination and a brief explanation of the analysis that led to that determination. Rate Review Procedures when an Increase is Determined to be Unreasonable If a State determines that a rate increase is unreasonable and the health insurance issuer is legally permitted to implement the unreasonable rate increase under applicable State law, CMS will provide the State s final determination and brief explanation to the health insurance issuer within five business days of when it is received by CMS. In Puerto Rico this will not be applicable since if the rate increase is determined to be unreasonable it cannot be used. If HHS reviews a rate increase and determines it to be unreasonable, HHS will provide its final determination to the health insurance issuer. If the issuer chooses not to implement the unreasonable rate increase, or to implement a lower increase than the threshold, the issuer is required to submit a final notification to this effect to HHS. However if this lower increase is still greater than the threshold, the lower increase would be subject to review and the issuer would be required to submit a new preliminary justification. If an issuer intends to implement a rate increase determined by CMS or a State to be unreasonable, the issuer is required, by the later of 10 business days after (i) the implementation of such increase or (ii) the health insurance issuer s receipt of HHS s final determination that a rate increase is an unreasonable rate increase, to: (1) Submit to CMS a Final Justification in response to CMS s or the State s final determination, as applicable. The information in the Final Justification must be consistent with the information submitted in the Preliminary Justification supporting the rate increase; and (2) Prominently post on its website, for at least three years, the following information on a form and in the manner prescribed by the Secretary: (i) The information made available to the public by CMS; (ii) CMS s or the State s final determination and brief explanation; and (iii) The health insurance issuer s Final Justification for implementing an increase that has been determined to be unreasonable. In addition, CMS will post all Final Justifications on the CMS Web site for three years. 27
Appendix B Excel Instructions for Trend Analysis 28
How to Create a Graph and Trendline in Excel You can search www.office.microsoft.com for more information Create a chart 1. Worksheet data To create a chart in Excel, you start by entering the numeric data for the chart on a worksheet. Then you can plot that data into a chart by selecting the chart type that you want to use on the Office Fluent Ribbon (Insert tab, Charts group). On the worksheet, arrange the data that you want to plot in a chart. The data can be arranged in rows or columns Excel automatically determines the best way to plot the data in the chart. 12-Month Incurred Period Ending 5/1/2008 6/1/2008 7/1/2008 8/1/2008 9/1/2008 Hospital Inpatient Allowed Claims PMPM 116.15 119.54 117.36 117.09 118.52 Hospital Outpatient Allowed Claims PMPM 115.14 117.09 116.54 115.84 115.03 29
2. Elements of a Chart A chart has many elements. Some of these elements are displayed by default, others can be added as needed. You can change the display of the chart elements by moving them to other locations in the chart, resizing them, or by changing the format. You can also remove chart elements that you do not want to display. The chart area (chart area: The entire chart and all its elements.) of the chart. The plot area (plot area: In a 2-D chart, the area bounded by the axes, including all data series. In a 3-D chart, the area bounded by the axes, including the data series, category names, tick-mark labels, and axis titles.) of the chart. The data points (data points: Individual values plotted in a chart and represented by bars, columns, lines, pie or doughnut slices, dots, and various other shapes called data markers. Data markers of the same color constitute a data series.) of the data series (data series: Related data points that are plotted in a chart. Each data series in a chart has a unique color or pattern and is represented in the chart legend. You can plot one or more data series in a chart. Pie charts have only one data series.) that are plotted in the chart. The horizontal (category) and vertical (value) axis (axis: A line bordering the chart plot area used as a frame of reference for measurement. The y axis is usually the vertical axis and contains data. The x-axis is usually the horizontal axis and contains categories.) along which the data is plotted in the chart. The legend (legend: A box that identifies the patterns or colors that are assigned to the data series or categories in a chart.) of the chart. A chart and axis title (titles in charts: Descriptive text that is automatically aligned to an axis or centered at the top of a chart.) that you can use in the chart. 30
A data label (data label: A label that provides additional information about a data marker, which represents a single data point or value that originates from a datasheet cell.) that you can use to identify the details of a data point in a data series. 3. Chart created from worksheet data Excel supports many types of charts to help you display data. When you create a chart or change an existing chart, you can select from a variety of chart types (such as a column chart or a pie chart) and their subtypes (such as a stacked column chart or a pie in 3-D chart). You can also create a combination chart by using more than one chart type in your chart. For the purposes of OCI we would recommend the use of line graphs. 3A) Highlight the data that you want to graph 3B) On the menu bar under Insert click on Line 3C) A new graph will appear on the workbook 31
3D) If you want to have multiple lines on one graph: highlight all the data and click on Line 4. Modifying a basic chart to meet your needs After you create a chart, you can modify any one of its elements. For example, you might want to change the way that axes are displayed, add a chart title, move or hide the legend, or display additional chart elements. To modify a chart, you can: 32
For more information on step by step instruction for these modifications please see excel help, or visit http://support.microsoft.com/. There are many videos you can find available by using your favorite search engine. Change the display of chart axes: You can specify the scale of axes and adjust the interval between the values or categories that are displayed. To make your chart easier to read, you can also add tick marks (tick marks and tick-mark labels: Tick marks are small lines of measurement, similar to divisions on a ruler, that intersect an axis. Tick-mark labels identify the categories, values, or series in the chart.) to an axis, and specify the interval at which they will appear. Add titles and data labels to a chart: To help clarify the information that appears in your chart, you can add a chart title, axis titles, and data labels. Add a legend or data table: You can show or hide a legend, change its location, or modify the legend entries. In some charts, you can also show a data table (data table: A range of cells that shows the results of substituting different values in one or more formulas. There are two types of data tables: one-input tables and two-input tables.) that displays the legend keys (legend keys: Symbols in legends that show the patterns and colors assigned to the data series (or categories) in a chart. Legend keys appear to the left of legend entries. Formatting a legend key also formats the data marker that's associated with it.) and the values that are presented in the chart. Apply special options for each chart type: Special lines (such as high-low lines and trendlines (trendline: A graphic representation of trends in data series, such as a line sloping upward to represent increased sales over a period of months. Trendlines are used for the study of problems of prediction, also called regression analysis.)), bars (such as up-down bars and error bars), data markers (data marker: A bar, area, dot, slice, or other symbol in a chart that represents a single data point or value that originates from a worksheet cell. Related data markers in a chart constitute a data series.), and other options are available for different chart types. How to Add a Trendline and R-squared in Excel There are many options to choose from when creating a Trendline. These are the most used and basic options. For more detail you can search office.microsoft.com for more information. 33
5. Add a Trendline 5A) On a chart sheet click the data series to which you want to add a Trendline 5B) On the Chart menu, click Add Trendline 34
5C) On Format Tendline tab, click the type of trendline that you want. We recommend that you use the linear option. 35
6. Display the R-squared value for a trendline 6A) On a chart sheet click the trendline for which you want to display the R-sqaured value 6B) On the options tab, select Display R-squared value on the chart 36