6 Data Governance Support Services PoliciesData StandardsData QualityData Architecture and InfrastructureData Governance Support Services
7 Terminology Vision Mayo Clinic Production Terminology Support To incrementally standardize Mayo Clinic terminologyTo access and use standardized terminology in applications, interfaces, warehouses, and processes across Mayo ClinicTo reduce cost and effort of developing and maintaining redundant and disparate terminologies and mappings for operations, analytics, and decision supportTo support business needs and strategies that require or benefit from standardized terminology (e.g., ICD-10 conversion, Meaningful Use, Health Information Exchange)
8 PrinciplesLeverage external standardsSupport Mayo needs that are not yet part of an external standard and promote them to the industryAlign with the Enterprise Data Model and other types of Mayo Clinic Data StandardsMaintain a balance between short-term and strategic needs and solutions
10 Solution VisionTools and Services: Load, Author, Map, Store, Search, Browse, Publish, AccessCode SystemsMayo Clinic Backbone Terminology, ICD-9 CM, ICD-10 CM, SNOMED CT, RxNorm, LOINC, etc.MappingsClinical Problems to ICD-9 CM, ICD-10 CM, and SNOMED CT, Lab Orders/Results to LOINCValue SetsClinical Problem, Body Structure, Body Side, Administrative Gender, Unit of MeasurePicklistsAdmitting Diagnosis, Lab Unit of MeasureEnterpriseTerminology
11 Value Set and Mappings Example Code SystemsClinical ProblemsHypertensionHyperlipidemiaDepressionHypothyroidismObesityOsteoporosisCoronary artery diseaseAnemiaDiabetes mellitusObstructive sleep apneaClinical FacingSNOMED CTMapsHealthInformationExchangesAdministrative BillingICD-9ICD-10Payers
12 Problem List Terminology Example Develop and approveProvide in an accessible formatUse in clinical and admin processesDistribute and LoadEMR’s, Scheduling, Orders, etc.
13 Terminology Management Offerings Methodologies and Best PracticesTerminology Standardization FacilitationExternal Standards ExpertiseTerminology Mapping (and 3rd party recommendations)External Terminology ManagementMayo Clinic Terminology Standards Management
14 Tools and Terminology Services Health Language Inc.Central authoring tool (LExScape)Mapping assistance tool (LEAP)Web-based browser (LExPlorer)Importer/Exporter and Web ServicesMayo Developed & SupportedExportsWeb Services (as needed)
15 Terminology Resources Data Governance Support Services Terminology ManagementTechnical support: IT Reference Repositories Operational SystemsData Governance Committee
16 Approved Data Standards and Policy Approved Data Standards on WebMayo Clinic Data StandardsEnterprise Data ModelData Standards PolicyData Governance Policy for Data Standards
17 Data Granularity Dissonance Practice vs. Research vs. Public Health Clinical data is impressionisticNot protocol driven – quaint rambling text…Detail and consistency is a happy coincidenceHowever, can be holistic; intuit the unanticipatedResearch data is rigid though rigorousDetailed, structured, complete (ideally), coherentMay miss the forest for the chlorophyllPublic Health comprehends broader interestsMetrics of life-style, high risk behavior, fitnessNon-clinical circumstances, settings, environmentCan there be reconciliation among use-cases?
18 electronic MEdical Records and GEnomics NHGRI eMERGE (U01) Goals GWAS – Genome Wide Association Study600k Affy chipHigh-throughput PhenotypingDisease algorithm scans across EMRs“catch up” with high throughput genomicsGeneralize Phenotypes across the ConsortiumMeasure reproducibility of algorithms among membersVanderbilt, Northwestern, Marshfield, Group Health Seattle, Mayo11/16/10
19 SHARP: Area 4: Secondary Use of EHR Data A $15M National Consortium 16 academic and industry partnersDevelop tools and resources that influence and extend secondary uses of clinical dataCross-integrated suite of project and productsClinical Data NormalizationNatural Language Processing (NLP)Phenotyping (cohorts and eligibility)Common pipeline tooling (UIMA) and scalingData Quality (metrics, missing value management)Evaluation Framework (population networks)
20 SHARP Area 4: Secondary Use of EHR Data Agilex TechnologiesCDISC (Clinical Data Interchange Standards Consortium)Centerphase SolutionsDeloitteGroup Health, SeattleIBM Watson Research LabsUniversity of UtahUniversity of PittsburghHarvard Univ.Intermountain HealthcareMayo ClinicMirth Corporation, Inc.MITMITRE Corp.Regenstrief Institute, Inc.SUNYUniversity of Colorado
22 Clinical Data Normalization Clinical data comes in all different forms even for the same kind of information.Comparable and consistent data is foundational to secondary useClinical Data Models – Clinical Element Models (CEMs)Basis for retaining computable meaning when data is exchanged between heterogeneous computer systems.Basis for shared computable meaning when clinical data is referenced in decision support logic.
23 A diagram of a simple clinical model Clinical Element Model for Systolic Blood PressureSystolicBPObsdata138 mmHgqualsBodyLocationdataRight ArmPatientPositiondataSitting
24 In order to represent detailed clinical data models, we have designed The Clinical Element Model (CEM). When we state “The Clinical Element Model” we are referring to the global modeling effort as a whole, or in other words, our approach to representing detailed clinical data models and the instances of data which conform to these models.How are Clinical Models used?Data entry screens, flow sheets, reports, ad hoc queriesBasis for application access to clinical dataComputer-to-Computer InterfacesCreation of maps from departmental/foreign system models to the standard database modelCore data storage servicesValidation of data as it is stored in the databaseDecision logicBasis for referencing data in decision support logicDoes NOT dictate physical storage strategy
25 Data Element Harmonization Stan Huff – CIMIClinical Information Model InitiativeNHS Clinical StatementCEN TC251/OpenEHR ArchetypesHL7 TemplatesISO TC215 Detailed Clinical ModelsCDISC Common Clinical ElementsIntermountain/GE CEMs
26 Normalization Pipelines Input heterogeneous clinical dataHL7, CDA/CCD, structured feedsOutput Normalized CEMsCreate logical structures within UIMA CASSerialize to a persistence layerSQL, RDF, “PCAST like”, XMLRobust Prototypes Q1 2012Early version production Q3 2010
27 This slide is obvious Physical Quantity 2. Coded Values 3. Text field Determine payloadIs laboratory data
28 Natural Language Processing Information extraction (IE): transformation of unstructured text into structured representations and merging clinical data extracted from free text with structured dataEntity and Event discoveryRelation discoveryNormalization template: Clinical Element ModelImproving the functionality, interoperability, and usability of a clinical NLP system(s), namely the Clinical Text Analysis and Knowledge Extraction System (cTAKES)
29 High-Throughput Phenotyping Phenotype - identifying a set of characteristics of about a patient, such as:A diagnosisDemographicsA set of lab resultsPhenotyping – overload of termsOriginally for research cohorts from EMRsObvious extension to clinical trial eligibilityNote relevance to quality metricsNumerator and denominator constitute phenotypesClinical decision supportTrigger criteria
30 EMR Phenotype Algorithms Typical componentsBilling and diagnoses codes; Procedure codesLabs; MedicationsPhenotype-specific co-variates (e.g., Demographics, Vitals, Smoking Status, CASI scores)Pathology; Imaging?Organized into inclusion and exclusion criteriaExperience from eMERGE Electronic Medical Records and Genomics Network (http://www.gwas.net)
32 Where is This Going?Standards and interoperability are crucial to the operation and success of academic medical centersThe boundaries among clinical and research standards are erodingInformation standards, especially vocabularies, are the foundation for scientific synergiesData Governance, at an enterprise (and arguably a national) level is critical