1. Hospital Readmissions: in search of potentially avoidable costs Bernard Friedman, PhD Center for Delivery, Organization, and Markets AHRQ Conference, 2009

2. Agenda Multiple uses of readmission data Multiple uses of readmission data – quality of inpatient care – effectiveness of management of chronic illness outside the hospital – efficiency in arrangements for post-hospital care – accountability for health plans: consumer choice and P4P Measurement choices depend on motives Measurement choices depend on motives – types of index admisssion, length of follow-up, eligible readmissions Tracking readmissions for the NHQR – it’s evolving Tracking readmissions for the NHQR – it’s evolving Recent research project: Contrast Medicare FFS vs. Advantage plan patients Recent research project: Contrast Medicare FFS vs. Advantage plan patients

3. Some AHRQ Published Studies on Readmissions 1.) Joanna Jiang was the lead author at AHRQ on several published studies of diabetes discharges. 1.) Joanna Jiang was the lead author at AHRQ on several published studies of diabetes discharges. – One finding was that half of the discharges or hospital costs in a year are for people with multiple discharges for diabetes and its complications. 2.) I examined (with Joy Basu) all readmissions within 6 months for people with 16 Potentially Preventable initial admissions. 2.) I examined (with Joy Basu) all readmissions within 6 months for people with 16 Potentially Preventable initial admissions. – Large variety of principal diagnoses for the RE-admissions – Just the re-admissions in the 16 categories of potentially preventable within 6 months had a projected national cost of about $1.4 Billion in 2008$. This covered only 4 states with 15% of the U.S. population.

4. Readmissions and Quality of Inpatient Care 3.) William. Encinosa and Fred Hellinger recently published “The Impact of Medical Errors on 90 Day Costs and Outcomes: An Examination of Surgical Patients”. Health Services Research, 2008 3.) William. Encinosa and Fred Hellinger recently published “The Impact of Medical Errors on 90 Day Costs and Outcomes: An Examination of Surgical Patients”. Health Services Research, 2008 – about $1.5 billion of cost in 3 months subsequent to the initial discharge due to safety events. Some of that was readmissions. 4.) B. Friedman, J. Jiang, W. Encinosa, R. Mutter, “Do patient safety events contribute to readmissions?” Medical Care, 2009. 4.) B. Friedman, J. Jiang, W. Encinosa, R. Mutter, “Do patient safety events contribute to readmissions?” Medical Care, 2009. – risk of a readmission within 1 month or 3 months after a surgical admission was raised about 20% by a safety event.

5. Effective Management of Chronic Conditions 5.) B. Friedman, with Joanna Jiang and Anne Elixhauser, 5.) B. Friedman, with Joanna Jiang and Anne Elixhauser, “Costly Hospital Readmissions and Complex Chronic Illness”, Inquiry, Winter, 2008/2009 – about 5 million adults were covered by the data – shows importance of the number of different chronic conditions in predicting readmission rates and annual cost. (“complexity”) – not easily “fixed” with disease-specific management protocols. But there is literature on demonstrations of other approaches. – 8% of the hospital costs for adults could be saved if you could bring down the extra readmissions for the 25% of hospitalized adulsts with 5 or more chronic conditions. – There have been a couple dozen demonstration projects of how to do that. It isn’t free, of course.

6. NHQR 2008 Readmissions Tracking system quality and system efficiency Tracking system quality and system efficiency Congestive Heart Failure, readmission for same. Congestive Heart Failure, readmission for same. readmission within 30 days (to any hospital) readmission within 30 days (to any hospital) – short enough to implicate the discharge planning, handoff, patient counseling – not apportioning blame (could be other factors) the national burden of readmissions: one person can have more than one readmit during the year qualifying to be counted (30 days from previous admit). the national burden of readmissions: one person can have more than one readmit during the year qualifying to be counted (30 days from previous admit). comparison of states within age groups (big difference between states, but not between age groups) comparison of states within age groups (big difference between states, but not between age groups)

7. Choices for future years NHQR Suggestions should go to Ernie Moy or Ryan or... Suggestions should go to Ernie Moy or Ryan or... Possibilities: Possibilities: – multiple index admissions, with statistical controls – readmission after elective treatment, after delivery – state or area rates with risk adjustment.

8. Do Medicare Advantage Patients Have Fewer Readmissions? Coauthors: B. Friedman, J. Jiang, John Bott, Claudia Steiner. Coauthors: B. Friedman, J. Jiang, John Bott, Claudia Steiner. Database: 5 states in HCUP with breakdown of type of Medicare coverage and with person identifiers. Database: 5 states in HCUP with breakdown of type of Medicare coverage and with person identifiers. Theory: superficially, it seems that the Advantage plans have both the motive (capitated revenue) and the means to reduce readmissions in comparison to FFS Medicare. Theory: superficially, it seems that the Advantage plans have both the motive (capitated revenue) and the means to reduce readmissions in comparison to FFS Medicare.

9. raw comparisons same 1-month rate of readmission (10%) same 1-month rate of readmission (10%) somewhat lower 3-month readmission rate (21% vs. 22.5%). somewhat lower 3-month readmission rate (21% vs. 22.5%). However, Advantage patients tend to be However, Advantage patients tend to be – a little younger – less severely ill even when hospitalized – less likely to have a major operative procedure.

10. Results Use risk adjustment and control for selection bias (predictors for joining an Advantage plan) Use risk adjustment and control for selection bias (predictors for joining an Advantage plan) Manuscript available on methods Manuscript available on methods Advantage patients are one third more likely to have a readmission (in 30 days, 13% vs. 10%; in 90 days, 30.5% vs. 22.5%). Advantage patients are one third more likely to have a readmission (in 30 days, 13% vs. 10%; in 90 days, 30.5% vs. 22.5%). How reconcile with incentives? How reconcile with incentives? – maybe we did something wrong.... – enrollees have no comparative data – FFS more discharges to LTC and other facilities – Advantage plans might be spending less on outpatient service and quality than we expected?