1 / 32

Active Clinical Decision Support in a RHIO

Active Clinical Decision Support in a RHIO. Introduction. The Clinical Document Architecture has emerged as a means of making systems interoperable. Clinical data can cross barriers. We are approaching an era of Computable Semantic Interoperability.

Download Presentation

Active Clinical Decision Support in a RHIO

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.


Presentation Transcript

  1. Active Clinical Decision Support in a RHIO

  2. Introduction The Clinical Document Architecture has emerged as a means of making systems interoperable. Clinical data can cross barriers. We are approaching an era of Computable Semantic Interoperability. Decision Support works when complex information is available. It works for a community when that data is available across systems so that patients can be treated with quality wherever they are seen. Active decision support provides guidance and recommendations at the point of care. In a RHIO, decision support will support evidence based care using standards to obtain and manipulate information.

  3. What is Active Decision Support? • Decision support: any means of helping get the right thing done • Passive Decision Support: documents or systems that can be searched or browsed to get answers • Active Decision Support: can reason with its knowledge and likely can communicate or act on its recommendations

  4. CDA Enables Active Decision Support • Active Decision Support depends on access to clinical and administrative data about patients. • CDA documents carry information that can be understood across systems. • So, Decision Support is enabled by CDA.Knowledge systems can operate in RHIOs, and as services in cooperation with clinical data systems. Computable Semantic Interoperability - Charlie Mead, ‘06

  5. “A collaborative project to develop a universal framework for sharing health knowledge in the form of computable clinical practice guidelines”

  6. An R&D consortium to develop the technology infrastructure to enable computable clinical guidelines, that will be shareable and interoperable across multiple clinical information system platforms. The SAGE Project • A 5-year, industry-academic research collaboration • Led by IDX Systems, now GE Healthcare • In partnership with: • Apelon, Inc. • Intermountain Healthcare • Mayo Clinic • Stanford Medical Informatics • University of Nebraska Medical Center

  7. Guidelines are Active What If . . . . • Guideline content became active, offering targeted, relevant guidance at the point of care? • Patients were evaluated against proven guidelines -- automatically? • Key data, care rationale and guidance were presented at critical decision points -- automatically?

  8. The SAGE Project Vision • A technology infrastructure that supports sharable, computable clinical practice guidelines -- augmenting clinical knowledge processing at the point of care. • With SAGE • Health experts can author and encode evidence-based clinical guidelines in a standard computable format. • Organizations throughout the world can easily deploy those guidelines using any conforming clinical information system.

  9. SAGE in a RHIO Environment Guideline File(s) EMR EMR EMR EMR EMR • Guidelines could execute at the “RHIO” level. • Guideline execution could obtain patient EMR data from a central or distributed repository. • Real-time, patient-specific recommendations could be provided via functions of the local CIS. • CDA documents are the basis for exchanging information. SAGE Guideline Engine Central Repository Model

  10. Active Guideline Environment good idea! Think of SAGE as guidelines tapping busyclinicians on the shoulder at just the right time.

  11. Gentle SAGE Recommendations • Guideline recommendations integrated into a nurse care flowsheet • View suggested orders • Process suggested orders Real time access to reference information Shows history, and future plans

  12. Order Sets as SAGE Recommendations SAGE adds patient specific comments to orders,and chooses preferred orders. Open Here HL7 Standard Order Set

  13. How Classic SAGE Technology Works Guideline File(s) • SAGE reads and executes an encoded guideline using standard terminology. • It communicates with CIS via a Virtual Medical Record standard interfaces. • SAGE detects events in the clinical workflow. • Queries patient data from the electronic medical record. • SAGE executes guideline logic based on patient specific data. • Real-time, patient-specific recommendations are expressed by the local CIS. Your Clinical Information System Events PAS Queries EMR Data SAGE Guideline Engine Orders Actions Labs More…

  14. RHIOs: Where is the patient record? EMR EMR EMR EMR EMR EMR EMR EMR EMR EMR Distributed EMR (no central EMR) (cf. Santa Barbara) Central Repository (“copy” of local EMRs) (cf. INPC, UK Spine) Central EMR (with distributed access) (cf. PeaceHealth)

  15. How CDA SAGE Technology Works Guideline File(s) • SAGE reads and executes an encoded guidelines using standard terminology. • It communicates with users and stores via CDA interfaces. • SAGE responds to events or requests from clinical systems. • Queries patient data from the RHIO or requester. The data are returned in CDA documents. • SAGE executes guideline logic based on patient specific data. • Patient-specific recommendations are expressed through CDA. CDA SAGE Guideline Engine CDA CDA CDA RHIO store

  16. Clinical Document Architecture • CDA defines a document structure that contains medical information. • CDA docs (XML) cross the wire. • In CDA v2, coded admin and clinical data may be present for use by decision support systems.

  17. XML Body: two types of content • Human readable (usually HTML)- required • Machine readable which can drive automated processes including decision support- optional except when doing decision support

  18. Major Components Header Narrative Block D O C U M E N T E N T R I E S S E C T I O N S B O D Y External References <ClinicalDocument> ... <structuredBody> <section> <text>...</text> <observation>...</observation> <substanceAdministration> <supply>...</supply> </substanceAdministration> <observation> <externalObservation> ... </externalObservation> </observation> </section> <section> <section>...</section> </section> </structuredBody> </ClinicalDocument> Thanks to Bob Dolin, MD Kaiser Permanente Major Comp-onents of a CDA Document

  19. CDA, Release Two Example CDA v2 for an Allergy List <section> <code code="101155-0" codeSystem="2.16.840.1.113883.6.1" codeSystemName="LOINC"/> <title>Allergies and Adverse Reactions</title> <text> <list> <item><content ID="A1">Penicillin - Hives</content></item> <item>Aspirin - Wheezing</item> <item>Codeine - Itching and nausea</item> </list> </text> <entry> <observation classCode="OBS" moodCode="EVN"> <code code="247472004" codeSystem="2.16.840.1.113883.6.96" codeSystemName="SNOMED CT" displayName="Hives"> <originalText><reference value="#A1"/></originalText> </code> <entryRelationship typeCode="MFST"> <observation classCode="OBS" moodCode="EVN"> <code code="91936005" codeSystem="2.16.840.1.113883.6.96" codeSystemName="SNOMED CT" displayName="Allergy to penicillin"/> </observation> </entryRelationship> </observation> </entry> </section> Thanks to Bob Dolin, MD Kaiser Permanente

  20. 1st Scenario: Consultation on Request Office • Small medical office, part of RHIO network • Patient goes to the office. Patient has some previous history of hypertension. • MD requests a consultation on medications for hypertension management. VERY SIMPLE!

  21. Small Practice in a RHIO 40 y.o. seen in clinic: BP checked, Hypertensionalready known

  22. SAGE Installed Centrally Office CDA document containsVital Signs, Hx, Labs, Allergies, Current Meds, … MD requests SAGE advice SAGE Guideline Engine Central EMR EMR SAGE Online CDA EMR

  23. SAGE reasons Sources Sources SAGE analyzes data SAGE Guideline Engine • Is ACE Inhibitor at max dose? • Any contraindications to Ca Channel Blocker? • Is Average Systolic BP over last 5 measurements> 150 mmHg • etc. CDA Central EMR CDA CDA Central DB containsseveral visit historiesfrom multiple institutions Office

  24. Recommendations Delivered Recommends • BP Not under Control. • Consider one of the following: 1) increase dose of lisinopril to 20mg qd 2) add felodipine 3) add thiazide diuretic, monitor K+ 4) add atenolol SAGE replies:

  25. Why is SAGE not just an Expert System? Recommends • Context (a series of events) • State (enrollment) • Sharable (across institutions) • Active (influence on care)

  26. Where are the data? • SAGE uses • CDA data submitted with consultation request, AND • Stored centrally in RHIO for this patient • Central store is CDA XML documents • Past history, Problem list, Allergies, Current Meds, History of Meds, Vital Signs, Laboratory work, etc.

  27. For Example, Current Medications include <section> <text>Digoxin 0.125mg, 1 PO qDay, #30, 5 refills.</text> <entry> <substanceAdministration classCode="SBADM" moodCode="RQO"> <effectiveTime xsi:type="PIVL_TS"> <period value="24" unit="h"/> </effectiveTime> <routeCode code="PO" codeSystem="2.16.840.1.113883.5.112"/> <doseQuantity value="1"/> <consumable> <manufacturedProduct><manufacturedLabeledDrug> <code code="317896006" codeSystem="2.16.840.1.113883.6.96" codeSystemName="SNOMED CT" displayName="Digoxin 125micrograms tablet"/> </manufacturedLabeledDrug></manufacturedProduct> </consumable> <entryRelationship typeCode="COMP"> <supply classCode="SPLY" moodCode="RQO"> <repeatNumber> <low value="0"/><high value="5"/> </repeatNumber> <independentInd value="false"/> <quantity value="30"/> </supply> </entryRelationship> </substanceAdministration> </entry> </section> This SNOMED-CT Coded entry is thus available for for SAGE to use in reasoning to make recommendations. <manufacturedProduct><manufacturedLabeledDrug> <code code="317896006" codeSystem="2.16.840.1.113883.6.96" codeSystemName="SNOMED CT" displayName="Digoxin 125micrograms tablet"/> </manufacturedLabeledDrug></manufacturedProduct>

  28. 2nd Scenario: SAGE helps out at the Cancer Clinic CDA CDA CDA CDA CDA CDA 05:30 Patient list sent to SAGE • RHIO/SAGE is sent a patient list before the clinic opens on Friday morning. • List is actually 47 CDA docs with up to date labs and findings for those patients who will appear in clinic that day. SAGE Guideline Engine Central EMR

  29. SAGE assesses each patient against protocol 0531-0546SAGE executes decision logic • Decision logic possibly requires additional historical, or missing patient data • SAGE can look for trends, specific alert levels, recent posted warnings, … SAGE Guideline Engine Other Systems Other Systems Central RHIO EMR

  30. SAGE Guideline Provides Advice for Each Patient SAGE publishes recommendations • When each patient is first seen by a clinician, that person is notified of any recommendations. • The clinician may act on some of those recommendations directly. • Each recommendation comeswith literature and logic references. John Abelson, #76789576F5, DOB: 03-04-1945 has a very low platelet count (5240). Other Labs: Na 134 K 14 WBC 10,100 SAGE Guideline Engine • Recommend • D/C Heparin • Reduce Met to 3.4g qd • Repeat count in 24 hr

  31. Recommendations may be comprehensive Recent BP: 06-07-05 152/86 Enter Today’s BP: MRN: 60946T3 NAME: Frederick T. Withers • SAGE may suggest • new diagnoses • new interventions • changes in schedules • adjustments in current treatments • additional studies • all with explanations and literature references. Goal: SBP < 135, DBP < 95 [diabetes, chf, renal insuffiency] Stage 1 Hypertension.BP out of control. Systolic trend upwards. Recommendations: Patient may benefit from 1 or more additional medication(s). Consider one of: ACE lisinopril 10mg qd ARB spironolactone 25mg qd BB atenolol 25mg qd

  32. www.sageproject.com

More Related