Identification of patient streams

Transcription

1 Reducing Avoidable ED Visits: IHI Triple Aim Prototyping Community Approach Alan Glaseroff, MD, CMO Humboldt Del Norte IPA CIN Webinar 9/28/11

2 IHI Framework Mission High Level Coalitions Identification of patient streams Refined Coalitions based on Strategies Reducing Avoidable ED Visits High level l population groupings Necessary for funding grants and business plans Data collection on patients (Uncover key defects in the healthcare/community resources) Strategies based on data collection may require additional partners Refined coalition establishes aims, measures initial design, concepts

3 Building a Learning Collaborative Charter Meeting: Agree on Vision and Core Principles Freedom to explore how principles implemented based on context Step into the work collectively: l Breakthrough Series Collaborative with pilot care teams Create emergent new knowledge through practice Etblih Establish a learning system Lead with principles, follow with tools and measures Emphasis on high yield change methods Model for Improvement/ PDSA cycles

4 Population and Baseline 7,000 managed care (HMO) patients + 3,000 self-insured (PPO), all HDNFMC plans ED use was 161 PKPY (HMO) compared to state average of 117 PKPY in 2008 Only 1 Urgent Care Center (hospital based) and it is about to close

5 Baseline Data HUMBOLDT-DEL DEL NORTE INDEPENDENT PRACTICE ASSOCIATION (0KR) Claims Incurred Oct 06 - Sep 07, Paid through Dec 07 (Year 2008 Base Line) PMG ER Visit Per Network ER Visit Per ER Visit Distribution by Day of the Week 20% 17% 17% 15% 16% 15% 15% 14% 14% 14% 13% 13% 13% 13% 13% 14% 10% 5% 0% Sunday Monday Tuesday Wednesday Thursday Friday Saturday HUMBOLDT-DEL NORTE INDEPENDENT PRACTICE ASSOCIATION (0KR) HMO Network ER Visit Distribution by Age Band Age Band # of ER Visits % of ER Visits Network Distribution <1 1 0% 1% % 9% % 7% % 7% % 9% % 10% % 10% % 9% % 8% % 11% % 7% % 6% % 4% % 2% Total % 100%

6 ED Fishbone Diagram

7 Patient Interviews & Strategies Approach to interviews Motivational interviewing techniques with patients who fit stream criteria i Interviews conducted by telephone or mail Open-ended questions and surveys with an area for general/spontaneous comments encouraged patient stories for qualitative measures

8 Patient Streams Large enough group for measurement Access to timely data stream Likely to have success Can be defined by whichever characteristic is useful (such as diagnosis, demographic, geographic, insurance type)

9 Six Streams 1. HMO members who use ED for non-urgent care 2. PPO members who use ED for primary care 3. Infants aged 0-1 who may be taken to ED for nonurgent conditions by anxious parents 4. Children aged 1-13 who have been seen in ED for non-urgent conditions (and tend to repeat) 5. Hospital employees who fast-track to ED when ill 6. Young adults aged who appear to use ED rather than primary care

10 Three Key Lessons from Interviews Lesson 1: PCP helpful only = 33 % There is perceived lack of access to PCP. 55% of patients did not contact PCP prior to ED visit Access not seen as PCP by problem by PCPs My job is to see the patients on my schedule. Sample comments: I didn t think it was appropriate to call after hours. My doctor is always too busy. My doctor s office staff always says they can t get me in for days.

11 Three Key Lessons from Interviews: Lesson 2: Patients lack information on alternatives to ED and/or have little confidence in their own judgment when faced with unexpected illness/injury. 55% hdli had limited i dif information i about resources other than ED. Sample comments: I didn t know what else to do. Idid didn t tk know about. t I didn t think about urgent care or the nurse line.

12 Three Key Lessons from Interviews: Lesson 3: ED is used for convenience and patients are unaware of increased cost. 50% said ED was close and convenient. Sample comments: I work at the hospital and if I come from the back, I don thavetowait to wait. I didn t realize ED was not the same as urgent care. I didn t know that ED costs more or that the co-pay is higher.

13 Primary Care System Breakdown Patients mostly made the best choice available based on their level of confidence in self-assessment, the knowledge of the system, and current incentives. No clear established set of mutual expectations between patient and practice What to do after hours? PCPs often don t see decreasing waste as their job 50% ED visits are during office hours and are paid anyway!!! Access and convenience Express Lube vs. 60,000 checkup

14 First Strategy Patient Information! PDSA: Resource list to patients with recent ED visits (along with PCP notification): 83% said resource list was very helpful but untimely (sent 8-16 weeks after ED visit) Avoided 4 repeat ED visits Received ei ed positive feedback from PCPs (4 calls,1 great story)

15 Second Strategy Timely notification to IPA! Results: Receiving daily list of ED patients directly from hospital is achievable, but takes constant nt vigilance to maintain

16 Third Strategy IPA contact patients! Results: Only 28% of patients responded to follow-up phone calls from IPA

17 Fourth Strategy Get PCP involved! PDSA: : Can one PCP office contact all patients on previous day s ED list within 24 hours? Results: Participating office had 100% success in contacting SJHS for list of patient names who had been to ED (continued each morning for 1 week) All 14 names on list were called within 24 hours (8 contacted, 3 needed follow-up appointments) Comments: Many patients didn t know about extended hours or that it was alright to call for advice. 100% of patients contacted were very pleased with follow-up calls.

18 Stream 1 : HMO Member ED visits July 1toDecember 31, Member Visits PKPY July 2009 August 2009 September 2009 October 2009 November 2009 December 2009 Results flat since Lack of payment py leverage: HMO 5% of market No PCMH payment pilot Access crisis FQHCs and private practices overwhelmed with demand

19 Medical Homes Defining Value Why pay differently? Reduce ED use and admissions Improve HEDIS/Patient Experience scores Attract more medical students into primary care in order to improve HEDIS and overall cost of care long term strategy ED use and PCP role Why is this important to me? Comparative reporting Changing patient behavior one-patient-at-a-time a time It s the relationship that matters being able to speak to a trusted source when anxiety-producing symptoms manifest; mutual expectations made explicit Rewarding relationship-based care Care management when more is needed

20