Effec%ve use of IT and decision support in guideline implementa%on Blackford Middleton, MD, MPH, MSc Director, Clinical Informatics Research and Development Partners HealthCare System Harvard Medical School
Readings Bates, D.W. et al. Ten commandments for effective clinical decision support: making the practice of evidence- based medicine a reality. JAMIA 10, 523 530 (2003). Sittig, D.F., et al. Grand challenges in clinical decision support. J Biomed Inform 41, 387 392 (2008). Sittig, D.F., et al. The state of the art in clinical knowledge management: an inventory of tools and techniques. Int J Med Inform 79, 44 57 (2010). Wright, A., et al. Clinical decision support capabilities of commercially- available clinical information systems. JAMIA 16, 637 644 (2009). Horsky, J., et al. Interface design principles for usable decision support: A targeted review of best practices for clinical prescribing interventions. J Biomed Inform 45, 1202 1216 (2012). Phansalkar, S., et al. Drug- drug interactions that should be non- interruptive in order to reduce alert fatigue in electronic health records. JAMIA (2012).doi:10.1136/amiajnl- 2012-001089 Phansalkar, S., et al. High- priority drug- drug interactions for use in electronic health records. JAMIA 19, 735 743 (2012). Boxwala, A.A., et al. A multi- layered framework for disseminating knowledge for computer- based decision support. JAMIA (2011).doi:10.1136/amiajnl- 2011-000334. Hongsermeier, T., et al. A legal framework to enable sharing of Clinical Decision Support knowledge and services across institutional boundaries. AMIA Annual Symposium proceedings 2011, 925 933 (2011). Dixon, B.E., et al. A Pilot Study of Distributed Knowledge Management and Clinical Decision Support in the Cloud. In press. See also AHRQ CDS Consortium: www.partners.org/cird/cdsc (HHSA290200810010) See also ONC Advancing CDS: www.rand.org/health/projects/clinical- decision- support.html (HHSP23320095649WC)
Overview Motivation for CDS Approaches to Knowledge Sharing Experiences with the CDS Consortium
Clinical Informa%on Exceeding Human Cogni%ve Capacity
Clinical Care Failures ADA Guideline Compliance Measure blood pressure at every routine diabetes visit. 64.22% On average, Patients receive 54.9% of recommended care Test for lipid disorders at least annually and more often if needed to achieve goals. 57.86% Visual foot exam at every routine visit, comprehensive foot examination annually 44.92% Test for microalbuminuria in all type 2 diabetic patients at least annually and during pregnancy. 23.62% Dilated and comprehensive eye exam at diagnosis of Type 2 and annually. 14.21% McGlynn EA, NEJM 2003; 348:2635.
Herbert A. Simon, Nobel Laureate Economics, 1978 What information consumes is rather obvious: it consumes the attention of its recipients. Hence a wealth of information creates a poverty of attention, and a need to allocate that attention efdiciently among the overabundance of information sources that might consume it. Changing clinician roles: From Omniscient Oracle to Knowledge Broker.
CITL HIT Value Assessments Net US could save $150B with HIT adoption, or approximately 7.5% or US Healthcare Expenditure The Value of Ambulatory Computerized Order Entry (ACPOE) $44B US nationally; $29K per provider, per year The Value of HealthCare Information Exchange and Interoperability (HIEI) $78B/yr The Value of IT- enabled Chronic Diabetes Management (ITDM) $8.3B Disease Registries; Advanced EHR $17B The Value of Physician- Physician Tele- healthcare >$20B* The Value of Personal Health Records Approx. $20B
A perfect storm for CDS? Lots of clinical data going online Increasing standardization, health information exchange Lots of genetic data coming Lots of personal/social data coming Lots of geospatial data coming Inexorable rise of Healthcare costs Healthcare Reform
CDS Grand Challenges Summarize patient- level information Prioritize recommendations to users Combine recommendations for patients with co- morbidities Improve the human- computer interface Use free text information in clinical decision support Manage large clinical knowledge databases Create a internet- accessible, clinical decision support repository Prioritize CDS content development and implementation Disseminate best practices Create an architecture for sharing executable CDS modules Mine large clinical databases to create new CDS Sittig et al., J Bio Inf 2008
Inference Methods Used in Expert Systems Algorithmic Statistical Pattern Matching Rule- based (Heuristic) Fuzzy sets Neural nets Bayesian TBD Inference Engine Copy Knowledge Base Inference Engine
Knowledge Transla%on and Specifica%on Evidence Experience Guideline(s) K Repres n Shareable K Executable Decision Tables GEM Arden GEODE-CM ONCOCIN EON(T-Helper) GLIF2 MBTA Asbru EON2 GLIF3 Oxford System of Medicine DILEMMA PRODIGY PROforma PRESTIGE PRODIGY3 1980 P. L. Elkin, M. Peleg, R. Lacson, E. Bernstam, 1990 S. Tu, A. Boxwala, R. Greenes, & E. H. 2000 Shortliffe. Toward Standardization of Electronic Guidelines. MD Computing 17(6):39-44, 2000
The Unified Theory for CDS Clinical Guidelines GEM CDSC L2 Structured Knowledge Evidence/ Experience Implementable/ Executable Knowledge CDSC L3 EHR EHR EHR OpenCDS/HL7 CDSC L4 Service CDSC CCD+ VMR
CDS Consortium: Goal and Significance Goal: To assess, define, demonstrate, and evaluate best practices for knowledge management and clinical decision support in healthcare information technology at scale across multiple ambulatory care settings and EHR technology platforms. Significance: The CDS Consortium will carry out a variety of activities to improve knowledge about decision support, with the ultimate goal of supporting and enabling widespread sharing and adoption of clinical decision support. 2. Knowledge Specification 1. Knowledge Management Life Cycle 3. Knowledge Portal and Repository 4. CDS Public Services and Content 5. Evaluation Process for each CDS Assessment and Research Area 6. Dissemination Process for each Assessment and Research Area
Knowledge is Like a Cake-Stack! Enterprise or Standard App Rules If Braden Score < 11: Low Air Loss Bed, etc If Abn Vasc Exam: Vascular Consult! Enterprise or Standard App Templates, " Flowsheets, Forms, Order Sets, etc! Collections of Concepts Braden Assessment Full Nursing Assessment Collections of Orders Order Sets! Enterprise Order Catalogues and Classes! Med Orders, Special Beds, Topicals Consults -Neurology or Vascular!! Enterprise Meds" (Dictionaries, Classes," Contraindications" Indications" Adverse Effects" Allergies"! Intermediate Concept Classes"! Enterprise Problem Lists! Dorsalis Pedis Pulse: Present or Absent Posterior Tibial Pulse: Present or Absent Color Pink, Pale, or Rubor on Dependency Ankle Brachial Index range 0.7à1.0 Taxonomies of Problems such as CAD, Diabetes, Peripheral Vascular DZ! Enterprise Terminologies Svs! Taxonomies of Terms such as Skin Exam, Decub Ulcer, Pulse, Skin Turgor 5
Building Blocks For Rule Modularity and Reusability Assessment Rules Screening Rules Management Rules Indication State Rules Goal State Rules Contraindication State Rules Disease State Rules Observation State rules Problem Class State Rules Drug Class State Rules Order Class State Rules Classes of Observations LOINC or SNOMED SNOMED Problem Subsets NDF- RT Drug Classes Other Order Classes Observation Dictionaries LOINC or SNOMED Problem List Dictionaries SNOMED Drug Dictionaries RX Norm Order Catalogues
Three Models to Accelerate Knowledge -> Practice Current paper-based approach" " Guideline " EMR Knowledge artifact import into EMR" Computer Interpretable Guideline " " Cloud-based clinical decision support services " Web Services " CCD/VMR Patient Data Object Decision Support
Knowledge Translation and Specification: Four-Layer Model derived from Level 1 Unstructured Format :. jpeg,. html,. doc,. xl derived from Level 2 Semi - structured Format : xml Level 3 Structured Format : xml derived from Initial evaluation results: Structured recommendation (L3) was considered more implementable than the semi- structured recommendation (L2). Level 4 Machine Execution Format : any + metadata + metadata + metadata + metadata Boxwala, A.A., Rocha, B.H., Maviglia, S., Kashyap, V., Meltzer, S., Kim, J., Tsurikova, R., Wright, A., Paterno, M.D., Fairbanks, A. & Middleton, B. A multi- layered framework for disseminating knowledge for computer- based decision support. JAMIA 18 Suppl 1, i132 9 (2011).
An external repository of clinical content with web-based viewer Search Criteria Content Type Specialty http://cdsportal.partners.org
Enterprise CDS Framework CDS Consumers Input (CCD) External Consumers PHS internal applications SMArt Apps PHSese SMArt Metadata Query ECRS Controller ECRS Rule Execution Recc Server Vendor Products Rule DB Rule Authoring VMR External Consumers OpenEHR... Open EHR CCD Factory Pt Data Access Supporting Services Translations / Normalization Services Normalization Translation Services Patient Data Reference Data
CDS Consor%um Demonstra%ons Toward a National Knowledge Sharing Service Mid- Valley IPA (NextGen) Salem, Oregon Kaiser Roseville UC Davis Kaiser Sacramento Kaiser San Rafael Kaiser San Francisco California Children s Hospital Colorado Wishard Hospital Indianapolis, IN Cincinnati Children s Nationwide Children s Ohio NYP NY PHS UMDNJ (GE) Newark, NJ CDS Consortium PECARN TBI CDS
Partners LMR
Screenshot of CDSC Reminders, in Regenstrief G3 EMR: Preven%on/ Recommenda%ons
NextGen EMR
Epic EMR: TBI Risk Es%mate
PHS Immuniza%ons SMArt App
v And knowledge!
Acknowledgements AHRQ: HHSA290200810010 Principal Investigator: Blackford Middleton, MD, MPH, MSc CDSC Team Leads: Research Management Team: Lana Tsurikova, MSc, MA KMLA/Recommendations Teams: Dean F. Sittig, PhD Knowledge Translation and Specification Team: Aziz Boxwala, PhD KM Portal Team: Tonya Hongsermeier, MD, MBA CDS Services Team: Howard Goldberg, MD CDS Demonstrations Team: Adam Wright, PhD CDS Dashboards Team: Jonathan Einbinder, MD CDS Evaluation Team: David Bates, MD, MSc Content Governance Committee: Saverio Maviglia, MD, MSc
Where are we? Thank you! Blackford Middleton, MD bmiddleton1@partners.org www.partners.org/cird