Project BOOSTReducing Readmissions

Mark V. Williams, MD, FACP, FHMProfessor & Chief, Division of Hospital Medicine

Northwestern U. Feinberg School of MedicinePrincipal Investigator, Project BOOST

A Problem for a long time

Rosenthal, J. M. and D. B. Miller "Providers have failed to work for continuity." Hospitals 53(10): 79-83. Continuity of patient care between different health care settings has been advocated for nearly 20 years, but little has been done to effect it. The study described here emphasizes the current lack of effort by health care providers in hospitals and nursing homes to find a workable solution.

1979

June 2007 MedPAC Report

Medicare pays for ALL admissions regardless Initial stay or readmission for same condition

17.6% of admissions result in re-admissions within 30 days (6% in 7 days) = $15 billion in spending

Future “CMS proposes to require that all general acute

hospitals conduct a CARE assessment on every Medicare beneficiary being discharged.”

Continuity Assessment Record and Evaluation Public Disclosure of readmission rates Lower case payments for readmissions

• 1 in 5 Medicare patients rehospitalized in 30 days• Half never saw outpatient doc• 70% of surgical readmissions–chronic medical conditions• Costs $17.4 billion

Jencks S, Williams MV, Coleman EA. et al. N Engl J Med 2009;360:1418-1428

Rates of Rehospitalization within 30 Days after Hospital Discharge

Health Affairs 2010; 29:57-64

Average LOS: US Hospitals

DeFrances et al, Adv data, 2007 Jul 12;(385):1-19

> 65 = 12.6 to 5.5 days

Observational study of 6,955,461 Medicare FFS hospitalizations for HF; 1993 and 2006, with 30-day f/u. Mean age = 80 52% Htn, 38% DM, 37% COPD

LOS 8.8 days down to 6.3 In-hospital mortality declined from 8.5% to 4.3%30-day mortality declined from 12.8% to 10.7%Discharges to SNF increased from 13% to 20%

Discharge to home decreased from 74% to 67%

30 day readmission increased from 17.2% to 20.1% Post-discharge mortality increased from 4.3% to 6.4%

Harlan M. Krumholz, MD, SM research group

Preventable Admissions Hospital inpatient care is the most expensive type

of health care > 4 million Preventable Admissions

Cost nearly $31 Billion

Heart Failure and Pneumonia Half of the $ problem

COPD – 16% Diabetes – 13%

Elderly – 2/3 of these hospitalizations - 1 in 5 Medicare admissions

Care Coordination Failure?

5 commercial disease management companies, 3 community hospitals, 3 AMCs, 1 integrated delivery system, 1 hospice, 1 long term care facility, 1 retirement community across U.S.

No cost savings2 reduced hospitalizationsSickest patients benefited

HospitalCompare.gov

Readmission ReductionCBO - $7.1B savings over 10 yrs

Hospital Quality & Performance Based Payments

All DRG payment amounts in hospitals with excess readmission are reduced by a factor determined by the level of “excess, preventable readmissions” Effective 2013 Excess = ratio of actual to expected (risk-adj) Reduction of 1%, 2%, and 3% first 3 years

Readmission Reduction ProgramNQF endorsed measures

Initially AMI, HF, pneumonia Expand in 2015 to 4 more conditions

COPD, CABG, PTCA, Other Vascular

Measures must have exclusions for readmissions unrelated to prior discharge

e.g. transfers, planned readmissions

Readmission time window specified by Secretary 30 days in NQF measures

Report all-payer readmission rates publicly

Measures – AMA PCPI

Care Transitions Work Group Performance Measure Set

Reconciled medication listTransition recordTimely transmissionDischarge Planning/Post-Discharge Support

for Heart Failure Patients

Hospital Discharge - currently

“Random events connected to highly variable actions with only a remote possibility of meeting implied expectations.”

Roger Resar, MDAgent of Tremendous Change and Global Innovation SeekerLuther Midelfort – Mayo Health SystemSenior Fellow, IHI

Dangers of Discharge

•19% of patients had a post discharge AE

- 1/3 preventable and 1/3 ameliorableAnn Intern Med 2003;

Vol. 138

•23% of patients had a post discharge AE

- 28% preventable and 22% ameliorable

CMAJ 2004;170(3)

1095 of 2644 (41%) inpatients discharged with test result pending

 - 191 (9.4%) potentially required action  - Survey of MDs involved: almost 2/3 unaware of results  - Of these: 37% actionable and 13% urgent

Dangers of Discharge

Ann Intern Med 2005;143(2):121-8

Dangers of Discharge

¼ of discharged patients require additional outpatient work-ups

> 1/3 not completedIncreased time to post-discharge f/u

associated with lack of work-up completionAvailability of discharge summary increased

likelihood of work-up being done

Arch Intern Med. 2007;167:1305-1311

Hospitalist to PCP

Info transfer and communication deficits at hospital discharge are common Direct communication 3-20% Discharge summary availability at 1st post-

discharge appt 12-34%; 51-77% at 4 weeks Discharge summaries often lack info

Dx test results (33-63%), hospital course (7-22%), discharge meds (2-40%), pending test results (65%)

Follow-up plans (2-43%), Counseling (90-92%)

Kripalani S, LeFevre F, Phillips CO, Williams MV, Basaviah P, Baker DW JAMA 2007;297:831-41.

Discharge Summary

J Gen Intern Med 2009;24:1002-6

“Discharge summaries are grossly inadequate at documenting both tests with pending results and appropriate f/u providers.”

Northwestern Solution

Significantly improved the quality and timeliness. Better documentation of f/u issues, pending tests, and

info provided to patients and/or family.PCPs more satisfied with timeliness and quality>95% of discharge summaries completed in < 1 week

Journal of Hospital Medicine 2009;4:219

Discharge Planning - is it THE answer?

21 RCTs: 4509 medical, 2285 med-surg; 440 Ψ LOS: mean decrease -0.91 (95% CI: -1.55 to -0.27) Readmission rates: RR 0.85 (0.74 to 0.97) Elderly medical pts: mortality RR 1.04 (0.74 to 1.46) Discharged to home: RR 1.03 (0.93 to 1.14) Improved patient satisfaction Subset analysis: improved functional status

Cochrane Database of Systematic Reviews 2010;1

Randomized 363 patients age > 65“Comprehensive discharge planning” and home

follow-up with APNs~70% completion rateReadmissions at 24 weeks 20% vs 37%

Reduced multiple readmissions 6.2% vs 14.5% Prolonged time to first readmission Medicare reimbursements cut in half

Elderly patients transitioning to SNF/homeRandomized: Intervention group paired with

“Transition Coach” vs. standard careEmpowerment and education: 4 pillars

Facilitate self management/adherence Maintain a personal health record Timely follow-up Knowledge and management of complications

Education during hospitalization including meds and med reconciliation

Phone calls and personal visits by TC post dischargeReduced rehospitalization and costs

Arch Intern Med 2006;166:1822-1828

ResultsRehospitalization Interv Cont P(adj) OR (95%CI)

Within 30d 8.3 11.9 0.048 0.59 (0.35-1.00)

Within 90d* 16.7 22.5 0.04 0.64 (0.42-0.99)

Within 180d* 25.6 30.7 0.28 0.80 (0.54-1.19)

Costs($) Interv Cont Unadj Log Transformed

At 30d 784 918 0.048 0.06

At 90d 1519 2016 0.02 0.02

At 180d 2058 2546 0.04 0.049

Arch Intern Med 2006;166:1822-1828

*Also significantly improved for

“Rehospitalization for same diagnosis as index admission.”

Or should it be a Pharmacist?

N=221 randomized at UCSFAll receive pharmacist facilitated discharge110 got 2 day phone call by pharmacist:

Check on clinical status Remind about follow-up Check on medications (did they obtain them; any

problems taking them; any side effects; did they know which to take and how; etc…)

Am J Med 2001;111(9B):26S-30S

Results

Contacted 79 or 110 25% had questions about their meds 11% had questions about their care 11% had questions about follow-up 19% had been unable to get their meds 15% reported new problems Greater satisfaction in intervention group:

86% vs. 61% very satisfied (p=0.007) 10% vs. 24% patients came to ED at UCSF at 30d

(p=0.005) 15% vs. 25% rehospitalized at 30d (p=0.07)

Pharmacy Literature

Schnipper et al: N = 178 medical patients randomized Intervention:

Med reconciliation done at d/c by Pharmacist Pharmacist counseling at d/c and 3day follow-up call At d/c, pharmacist recommended med changes in 60% At 3d call, unexplainable discrepancies between d/c meds

and reported home meds in 29% At 30d

Fewer preventable ADEs: 1% vs. 11% (p=0.01) Fewer preventable med related ED visits: 1% vs. 8% (p=0.03) 49% had med discrepancies! No difference in total ADEs, health care utilization, patient

satisfaction, or med adherenceArch Intern Med 2006;166:565-71

Pharmacists Work!

Swedish ward-based pharmacists 16% reduction in hospital visits 47% reduction in ER visits Drug-related readmissions reduced 80% Intervention group cost < control

Arch Intern Med. 2009;169(9):894-900

Project REDRCT of 749 hospitalized adultsIntervention

Nurse Discharge AdvocateF/U appt, Medication ReconciliationPatient education

Individualized instruction booklet Pharmacist call 2-4 days post-discharge

Review medications

Limitations Urban, academic, safety net hospital

Project RED Outcomes

Intervention(n = 370)

Control(n = 368)

ER Visits* 16.5% 24.5%

Rehospitalization** 15% 21%

PCP f/u in 30 days* 62% 44%

Prepared for Discharge* 65% 55%

*p < 0.05

**p = 0.09

Low-cost Intervention

“user-friendly” Patient Discharge FormTelephone outreach from a nurse post-

dischargeImproved outpatient follow-upReduced ER visits and

rehospitalizations from historical controls

JGIM 2008

1. Med Rec by PharmD

2. RN Care Coordinator D/C Planning

3. Phone Follow-up

4. PHR, Supplemental Discharge Form Reduced ER visits, Reduced Readmission

SHM Initiatives

Discharge Checklist Halasyamani L et al. Transition of care for hospitalized elderly patients--development of a discharge checklist for hospitalists. J of Hosp Med 2006:354.

Resource RoomSafe STEPs

Project BOOST Better Outcomes for Older adults through

Safe Transitions John A. Hartford Foundation $1.4 million

Safe STEPsSafe and Successful Transitions for

Elderly PatientsJohn A. Hartford Foundation Grant

Safe STEP Interventions

Medication reconciliation Pharmacy reviews: admission and d/c Geriatric friendly medication forms

Education Patients: pre-d/c appointment Providers: geriatric h&p

PCP communications “Fast facts”

Safe STEPs

237 elderly patients at three hospitals Academic, community

5 component intervention Admission form with geriatric cues Fax to PCP Interdisciplinary worksheet Pharmacist-physician medication reconciliation Pre-discharge planning appointments

Reduced ED visits and readmissions by 1/3

Project BOOST Team

• Tina Budnitz, MPH• Eric Coleman, MD, MPH• Jeff Greenwald, MD• Eric Howell, MD• Lakshmi Halasyamani, MD• Mark V. Williams, MD

• Janet Nagamine, MD• Dan Dressler, MD, MS• Kathleen Kerr• Greg Maynard, MD• Arpana Vidyarthi, MD

Advisory Board

Social workCase managementClinical pharmacyGeriatric medicineGeriatric nursingHealth ITBlue Cross/Blue ShieldUnited Health

Health systemsNQFAHRQTJCCMSNational Consumer’s

LeagueOther content experts

Chair: Eric Coleman, MD, MPH

Co-Chair: Mark V. Williams, MD

with organizational representatives from:

www.hospitalmedicine.org/BOOST

What is BOOST Today?

Intervention Tailored clinical Tools:

Comprehensive Risk Assessment Team-based care Patient centered discharge process 72 Hour follow-up call for “high-risk” patients Scheduled outpatient follow-up visits Standardized PCP Communication

Tailored processes, work-flow Project management tools

BOOST components (cont)

Technical Support Mentors calls, email, resources Teleconferencing across sites Education (webinars, newsletters) Enduring Materials (Teachback DVD)

Peer Support Listserv Document sharing Moral support

Infrastructure Development Train the trainer curricula Mentor Guides Mentor University

NEW CONCEPT: Health information,

advice, instructions, or change in management

Adherence / Error reduction

Explain new concept / Demonstrate new skill

Patient recalls and comprehends /

Demonstrates skill mastery

Assess patient comprehension /

Ask patient to demonstrate

Clarify and tailor explanation

Re-assess recall and comprehension /

Ask patient to demonstrate

Teach Back

Modified from Schillinger, D. et al. Arch Intern Med 2003;163:83-90

Implement intervention

Keep stakeholders informed

Monitor core elements

Analyze data

Adjust intervention components

Report to stakeholders

Spread gains

Life-Cycle Project BOOST

Training & Preparation Individualized Mentoring

Training-6months 6-9 months 9-12 months

BOOST Network

BOOST eNewsletter Key milestones BOOST updates Site status reports, aggregate outcomes Forum for sharing ideas, challenges, mini

studies

BOOST Network E-mail, call between sites BOOST listserv

End-Result

Network of Institutions using the guide and interventions

Understanding Impact of Interventions

Understanding Implementation facilitating factors and barriers

BOOST Mentor Sites

Projected Growth

Cohort 1: 9/08 6 sitesCohort 2: 3/09 24 sitesMI Collaborative 5/10 14 sitesTuition pilot 5/10 2 sites

CA Collaborative 20 sitesFall 10 Tuition Cohort 15 sites

Online in 2010 = 81 sites

So what happens to readmission rates?

Hierarchical time series analysis of readmission rates (one year prior to kick-off through one year post kick-off) 12/10

Cohort 1 (n=6) kickoff

Implementation Survey

Cohort 2 (n=24) kickoff

12/08 6/09 12/09 12/10

Prelim Results

Across all sites overall readmission rates decreased from 13% to 11%. BOOST Intervention Units 6 months post “go live”

Readmission rates rose in non-BOOST units by 2%

Marked increased patient satisfaction at some sites.

A Hospital Nurse

“Project BOOST brings me back to what I thought nursing was really about. BOOST helps patients and families understand what they need to do to go home. This is why I went into nursing.”

THANKS!!!

The John A. Hartford Foundation