Patients. Government Payers. Providers. Quality. Measure Developers. Implementers

Transcription

1 Patients Government Payers Quality Providers Measure Developers Implementers

2 Capturing EHR Enabled Quality Improvement- Decision Support, Measurement, and Reporting REQUIRES an understanding of context February 23, 2014 John Chuo, MD, MS Children s Hospital of Philadelphia Neonatal Quality Officer Medical director, Telemedicine DISCLAIMER: The v iews and opinions expressed in this presentation are those of the author and do not necessarily represent of f icial policy or position of HIMSS.

3 CQM measures how good or bad healthcare delivery is but Recognizing a bad care delivery system does not mean you know how to improve it.

4 Poor understanding of system = Poor implementation of best practices Managing Clinical Knowledge for Health care Improvement, EA Balas, 2000

5 Using clinical quality measures to drive improvement in healthcare delivery requires understanding of context.

6 CDS as a means for improving CQM Evidence, expert, experience Can it work? (Efficacy) Principles of best practice for CDS implementation Use of clinical decision support system Does it work In MY context? (Effectivenes s) Implementation Outcome Compliance with Using CDS intervention CQM Outcome (Prophylactic antibiotic received within 1 hour prior to surgery) John Chuo, manuscript in progress

7 Successful Change Management FOR CDS IMPLEMENTATION Is Does Why the would user the desired have user the use the patient skills CDS, to why use not outcome the override? CDS? clear? (training) Do you have resources to react to feedback What is the Promptly action plan for improving the CDS? Knoster, T., Villa R., & Thousand, J. (2000) A framework for thinking about systems change.

8 EXAMPLE: Implementing a paper reminder system using a CDS backend on patient rounds in the NICU

9 Medical Team Rounding tool workflow Conducts patient rounds QI Student intern 4. Student ask reminders during rounds and record answers Admin 1. Runs the rounding tool system each am 3. Delivers the paper QI tool with the reminders to each of the rounding teams 5. Updates data repository Computer system 2. Print out QI reminders depending on patient conditions (by searching thru previous day notes)

10 Rounding tool

11 % of time assigned questions were applicable to patient Results % of times assigned question needed prompting by intern % times assigned question had Response unfavorable to patient (Importance factor) % of time assigned question not asked, Unfavorable and an Action Was Prompted (Impact factor) AVG. 93% 11% 10% 6% (600 actions) Top 5 questions that needed prompting by intern, response unfavorable to patient, prompted Action 80.00% 60.00% 40.00% 20.00% 0.00% "Third Trimester HIV Results" (7/11) "Social History" (3/5) "Family History" (1/2) "O2 Sat. Limits" (1/7) "Immunizations" (1/8) John Chuo, manuscript in progress

12 Contact Info John Chuo 1

13 Conflict of Interest Disclosure John Chuo, MD, MS No conflicts of interest to disclose 2014 HIMSS

14 Capturing EHR Enabled Quality Improvement- Decision Support, Measurement, and Reporting February 23, 2014 Floyd Eisenberg, MD, MPH, FACP iparsimony, LLC The presentation includes content from the American Hospital Association. A Study of the Impact of Meaningful Use Clinical Quality Measures. July American Hospital Association DISCLAIMER: The v iews and opinions expressed in this presentation are those of the author and do not necessarily represent of f icial policy or position of HIMSS.

15 ecqm implementation process Eligible Hospitals Gap analysis Data capture and workflow redesign Data extraction and ecqm calculation Validation Downstream uses of ecqm results Common implementation steps observed at all sites Iterative, non-linear process American Hospital Association. A Study of the Impact of Meaningful Use Clinical Quality Measures. July Available at:

16 ecqm implementation experience Eligible Hospitals Largely prescribed by EHR/eCQM reporting tool Workarounds and successive iterations 80% of effort entailed changes to hospital workflow solely to accommodate ecqm data capture Gap analysis Data capture and workflow redesign Data extraction and ecqm calculation Innacurate ecqm results Validation Sensitivity issues causing under reporting: Data not in prescribed location (as expected by ecqm tool) Internal systems with needed data not interoperable with EHR Usability affecting data quality Downstream uses of ecqm results No trust or reliance in ecqm results American Hospital Association. A Study of the Impact of Meaningful Use Clinical Quality Measures. July Available at:

17 ecqm Impact Study: Challenges Program Challenges 1. ecqms were not tested for validity, accuracy and feasibility 2. ecqms were hard to find, lengthy, and often contained errors 3. MU ecqms require understanding of unfamiliar terminologies 4. Guidance to ignore data accuracy and focus on the ability to report undermines goals for quality improvement American Hospital Association. A Study of the Impact of Meaningful Use Clinical Quality Measures. July Available at:

18 ecqm Impact Study: Challenges Technical Challenges Expectations that existing EHR data would suffice to calculate the ecqms were not realized 1. EHRs do not store entered data in readily retrievable form 2. EHRs are not designed to capture many of the elements in structured form to enable re-use for ecqm reporting 3. EHRs are not designed to capture information from other department information systems at the level of detail needed for ecqm reporting American Hospital Association. A Study of the Impact of Meaningful Use Clinical Quality Measures. July Available at:

19 ecqm Impact Study: Challenges Clinical Challenges 1. EHRs and certification requirements are not designed to support end-to-end patient care workflows to draw data as expressed in ecqms 2. Hospitals were unable to validate the ecqm results 3. Some ecqm specifications were out of date Strategic Challenges 1. Time and personnel requirements to implement ecqms were excessive and far beyond expectations 2. The time and effort provided no return on investment as results could not be validated and were therefore not useful for quality management. American Hospital Association. A Study of the Impact of Meaningful Use Clinical Quality Measures. July Available at:

20 ecqm Impact Study: Recommendations American Hospital Association. A Study of the Impact of Meaningful Use Clinical Quality Measures. July Available at:

21 Updated Process: A World of Dependencies Review Oversight - HHS Measure Authorin g Tool Value Set Authority Center Quality Data Model Standard Vocabulary Health Quality Measure Format Quality Report Document Architecture Clinical Document Architecture Courtesy: The Joint Commission American Hospital Association. A Study of the Impact of Meaningful Use Clinical Quality Measures. July Available at:

22 ecqm to CDS Example: Measure CMS 165 (NQF 0018) Trigger Condition Actions CONDITION: All are true: Patient is years Patient has diagnosis = Hypertension [ Active ] Patient does not have diagnosis = Pregnancy Patient does not have diagnosis = End Stage Renal Disease Patient has not had procedure during the measurement year = ESRDrelated procedures APPLIES WHEN: Diastolic Blood Pressure > 90 OR Systolic Blood Pressure > 140 during the last visit New data: Query Patient Registry q30 Days Advice: Provide list of patients with possible need for follow up AND hyperlink to NHLBI Guidelines sion/index.htm And provide patient education resources: Evaluate: 1. Currency of evidence 2. Feasibility of elements in clinical workflow 3. Value set content APPLIES WHEN: Diastolic Blood Pressure OR Systolic Blood Pressure are absent during the last visit Advice: Provide list of patients with indication blood pressure should be taken at each visit AND hyperlink to NHLBI Guidelines sion/index.htm And provide patient education resources:

23 Contact Info Floyd Eisenberg This presentation includes content from the American Hospital Association. A Study of the Impact of Meaningful Use Clinical Quality Measures. July American Hospital Association 2

24 Conflict of Interest Disclosure Floyd Eisenberg, MD, MPH, FACP No conflicts of interest to disclose 2014 HIMSS

25 Capturing EHR Enabled Quality Improvement- Decision Support, Measurement, and Reporting February 23, 2014 Ferdinand Velasco, M.D., FHIMSS Chief Health Information Officer, Texas Health Resources Chair, HIMSS Quality Cost Safety Committee DISCLAIMER: The v iews and opinions expressed in this presentation are those of the author and do not necessarily represent of f icial policy or position of HIMSS.

26 Organizational Background One of the largest faith-based, nonprofit health care delivery systems in the United States and the largest in North Texas in terms of patients served. The system's primary service area consists of 16 counties in north central Texas, home to more than 6.2 million people. 26

27 Texas Health Resources EHR highlights 2013 Davies Award All hospitals at HIMSS EMRAM Stage 6 or 7 Attested to Meaningful Use Stage 1 (3 years)

28 We are here Received 80% of anticipated EHR incentive funding from 3 years of Stage 1 MU Addressing challenges of meeting Stage 2 MU objectives Utilizing historical methods for HIQR and PQRS reporting Shifting organizational focus from acute care to population health management

29 Case studies Purpose: to illustrate the considerations of transitioning from Chart Abstracted Measures to emeasures ED throughput: median time from ED arrival to departure for admitted patients Ischemic Stroke: anticoagulation therapy for atrial fibrillation/flutter

30

31 ED throughput (NQF 0495) Source: CMS, Guidance/Legislation/EHRIncentivePrograms/Downloads/2014_CQM_EH_FinalRule.pdf

32 ED throughput (NQF 0495) Measure output relevant and meaningful both internally and externally emeasure specification relative simple Documentation of workflows able to capture discrete EHR Excellent correlation between manual abstraction and EHR method

33 ED throughput (NQF 0495) 350 ED ED

34 ED throughput (NQF 0495) Measure output relevant and meaningful both internally and externally emeasure specification relative simple Documentation of workflows able to capture discrete EHR Excellent correlation between manual abstraction and EHR method Potential opportunities for improvement More meaningful segmentation Correlation with ED / hospital census ED wait times ED staffing ratios Syndromic surveillance Patient satisfaction Use of realtime location sensing technology to eliminate manual time stamps in EHR Consider similar measures for inpatient, OR, ambulatory throughput

35

36 Source: Texas Dept. of State Health Services, Texas Influenza Surveillance Report Season

37 Stroke: anticoagulation therapy for atrial fibrillation/flutter (NQF 0436) Source: CMS, Guidance/Legislation/EHRIncentivePrograms/Downloads/2014_CQM_EH_FinalRule.pdf

38 Stroke: anticoagulation therapy for atrial fibrillation/flutter (NQF 0436) Measure output relevant and meaningful both internally and externally Measure logic complex Documentation of workflows uneven Challenges with translating EHR data into discrete variables needed to generate CQM Modest success with reconciling abstracted and EHR-derived data

39 Source: AHRQ United States Health Information Knowledgebase,

40 Stroke: anticoagulation therapy for atrial fibrillation/flutter (NQF 0436) Source: AHRQ United States Health Information Knowledgebase,

41 Stroke: anticoagulation therapy for atrial fibrillation/flutter (NQF 0436) Measure output relevant and meaningful both internally and externally Measure logic complex Documentation of workflows uneven Challenges with translating EHR data into discrete variables needed to generate CQM Modest success with reconciling abstracted and EHR-derived data Potential opportunities for improvement Reduce measure logic complexity Fewer exclusions Be more parsimonious and prescriptive about definitive data sources Correlation with Patient education Use of secure messaging Long term anticoagulation effectiveness and safety Possible CDS application Checklist (pre-discharge)

42 Lessons learned / considerations Process Build CQM logic manually from measure specifications OR Utilize emeasure specifications from certified EHR technology provider Validation of CQMs Prioritization: CQM reporting for Internal process improvement External reporting Pay for performance Pay for reporting Measure overlap vs. separation EH: HIQR / VBP / MU EP: PQRS / ACO / MU

43 Conclusions Transition from manually abstracted measures to emeasures will be a long journey Approach needs to be meticulous and systematic Some emeasures are usable; most are not Abstraction will likely never be completely eliminated, at least for CQMs with complex measure logic Shift focus from retrospective chart abstraction to concurrent care management

44 Contact Info Ferdinand Velasco HIMSS Quality Cost Safety Committee 4

45 Conflict of Interest Disclosure Ferdinand Velasco, MD No conflicts of interest to disclose 2014 HIMSS

46 Integrating Quality Measurement and CDS-enabled Quality Improvement February 23, 2014 Jerome A. Osheroff, MD, FACP, FACMI Principal, TMIT Consulting, LLC Adjunct Associate Professor of Medicine, University of Pennsylvania DISCLAIMER: The v iews and opinions expressed in this presentation are those of the author and do not necessarily represent of f icial policy or position of HIMSS.

47 We are Here Strong/mounting pressure for measurable improvements Sub-optimal data to understand care process/outcomes Difficulty enhancing measurement/performance QM & CDS worlds both working on these Turn challenges to joint opportunities!

48 CDS Definition A process for enhancing health-related decisions and actions with pertinent, organized clinical knowledge and patient information to improve health and healthcare delivery. Improving outcomes with CDS, 2 nd Ed. HIMSS 2012 Reinforces inter-dependence of QM and QI

49 QI Success Framework: CDS Five Rights To improve targeted care processes/outcomes, get: the right information evidence-based, actionable [what] to the right people clinicians and patients [who] in the right formats documentation tools, data display, answers, order sets, alerts [how] through the right channels EHR, portals, smartphones, smart pill bottles/monitors [where] at the the right times key decision/action [when] 49

50 ONC Toolkit: Resources for Improving Care with CDS* *Posted at: bit.ly/cds4mu 5

51 Getting Better Faster Together (SM) : A Case Example

52 Warm-up Questions (To Engage Site Leads) 1. What % of our HTN patients have BP<140/90? 2. How are we supporting patient and clinician decisions and actions to drive improvement? 3. Are we using our EHRs (and other tools) to greatest benefit for workflow and outcomes? 4. How can we work as a group to get more efficient at Quality Improvement (QI) and Collaboration? 5

53 Improving BP Control What needs to happen? Decisions Actions Communication Data gathering In RCH health centers, today What information, Flows through which people, In what formats/channels, At which times? 5

54 Ambulatory Worksheet (Simplified) TMIT Consulting, LLC 5

55 Patient List/Registry: Powerful Non- Alert CDS Tools From Dr. Chris Tashjian, Ellsworth Medical Clinic, with permission 5

56 QI/CDS Worksheets: Tool to Get CDS 5 Rights Right Document target-focused information flow Invariably suggests potential enhancements Helps get QI team/practice on the same page Foster collaboration Provider <-> Provider (via Collaborative private site) Vendor <-> Provider client QI Experts (Million Hearts) <-> Implementers/Vendors Measure Developers <-> Others?! 5

57 Steps for CDS-enabled QI* 1. Shared understanding of CDS/QI concepts 2. Select improvement targets 3. Envision successful target-focused CDS/QI 4. Use CDS/QI worksheets 5. Make enhancements ( Plan-Do-Study-Adjust ) 6. Collaborate Role for QM Community?! *See bit.ly/cdsqisteps

58 Contact Info Jerry Osheroff ONC CDS for MU/QI Tools and Resources bit.ly/cds4mu CDS/PI Collaborative (Public) bit.ly/cdspicollab 5

59 Additional slide follows: Inpatient example of CDS/QM interplay: Clinical intervention contraindications (and CQM measure exclusion) captured during care delivery in a smart order set : simultaneously serves QI/CDS/QM purposes

60 An Inpatient Example: VTE Prophylaxis Best practice CDS/QI recommendations Based on Society of Hospital Medicine expertise Anchored by order sets that provide: Risk stratification (Hi/Med/Low) Orders pertinent to risk strata Opportunity to document contraindications Uses unit dashboards analogous to registry Simultaneously support best care, measurement, reporting For recommendations see:

61 Conflict of Interest Disclosure Jerome A. Osheroff, MD Ownership Interest TMIT Consulting, LLC provides healthcare quality improvement-related professional services, and helped developed the freely available ONC CDS/QI resources mentioned in this talk HIMSS