Racial and Ethnic Disparities in 30-Day Readmissions -...
Transcript of Racial and Ethnic Disparities in 30-Day Readmissions -...
Racial and Ethnic Disparities in 30-Day Readmissions An Analysis of Connec.cut Hospital Discharge Data
Connec.cut State Medical Society and Connec.cut Health Founda.on Symposium
Tackling Racial and Ethnic Dispari3es in Connec3cut’s Acute
Care Hospitals
Robert H. Asel.ne, Jr., PhD February 4, 2015
Collaborators and Contributors
• MaG Katz, CSMS
• Catherine Wagner, CSMS
• Jun Yan, UCONN
• Greg MaGhews, Loyola University
• Claudia Gruss, ProHealth
• Ross Benthien, HarTord Hospital
• Courtland Lewis, HarTord Hospital
• Steve Fleischman, Yale-‐New Haven
• Mark DeFrancesco, ACOG • Natalie Achong, Yale
University
Background • Race and ethnic dispari.es in hospital readmissions well
documented: – Blacks higher than Whites for diabetes, AMI, CHF, pneumonia,
cirrhosis, TJA, and following ambulatory surgery
– Hispanics higher than Whites for diabetes, CHF, AMI
– Rates of 30 day readmission are generally 10-‐20% higher among racial and ethnic minori.es
• CT appears to have substan.al dispari.es in health outcomes and access to care – Connec.cut Health Founda.on report (2010) – 10 year analysis – Connec.cut Health Care Survey (2014)
Overview of CSMS Connec.cut Hospital Readmission Ini.a.ve
• Calculate 30 day readmission rates across range of condi.ons/procedures using hospital discharge data – Expand condi.ons to those not previously studied and those affec.ng younger pa.ents
– Control for pa.ent characteris.cs confounded with race and ethnicity:
• Pa.ent sickness and comorbidi.es
• Pa.ent SES • Insurance coverage (public vs. private)
Overview of CSMS Connec.cut Hospital Readmission Ini.a.ve
• Conducted focus groups with specialists in OB-‐GYN, orthopedic surgery, gastroenterology, cardiology – Discussed role of pa.ent sickness, insurance type in fostering dispari.es
– Discussed unmeasured factors (e.g., pa.ent engagement, connec.on to primary care, pa.ent – provider communica.on)
– Discussed poten.ally strategies to address dispari.es
Overview of CSMS Connec.cut Hospital Readmission Ini.a.ve
• Reported hospital specific data to 7 partner ins.tu.ons
– Norwalk Hospital, Milford Hospital, Yale-‐New Haven Hospital, Bristol Hospital, St. Mary’s Hospital, Rockville General Hospital, Manchester Memorial Hospital
– Discussed factors that might account for beGer/worse performance than State
• Disseminate findings in peer-‐reviewed manuscripts
Measures 30 Day Readmissions – Diagnosis-Related Groups (DRGs)
DRG Condi*on Number of Admissions
% 30 Day Readmissions
775 Vaginal Delivery 167,807 .8%
766 Cesarean Delivery 75,552 1.5%
291,292 Heart Failure 39,985 23.1%
312, 313 Chest Pain 23,450 9.8%
332-‐395 Esophagi.s, Gastroenteri.s, and Misc Diges.ve Disorders 284,471 11.6%
470 Total Joint Replacement (ICD 9 CM 81.51, 81.52, 81.54, 81.56) 51,843 5.2%
191-‐193 Pneumonia 44,616 15.3%
870-‐872 Sep.cemia 40,306 18.0%
602, 603 Celluli.s 27,943 11.9%
689,690 Kidney and Urinary Infec.ons 31,357 16.3%
Measures • Race/ethnicity: pa.ent or facility reported • Pa.ent Sickness: Elixhauser AHRQ comorbidity index
• 29 discrete comorbid and/or complica.ng medical condi.ons
• Up to 10 primary and secondary diagnoses were recorded for each hospitaliza.on
• Pa.ent age, sex, length of stay, facility volume, payer
Methods
• Up to 8 years of data from HIDD (2005 -‐ 2012)
• Es.mated logis.c regression models – Used generalized es.ma.ng equa.ons to account for clustering of observa.ons within person (mul.ple admissions)
Results Race and Ethnic Differences in 30 Day Readmissions in
Connecticut Hospitals
2.05
1.85
1.62
1.35
1.27
1.39
1.12
1.03
1.19
1.13
1.09
0.0 0.5 1.0 1.5 2.0 2.5
Cesarean Sec.on
Vaginal Delivery
Total Joint Arthroplasty
Sep.cemia
Celluli.s
Gastroenteri.s -‐ Surgical
Gastroenteri.s -‐ Medical
Heart Failure and Shock
Chest Pain
Kidney Disease
Pneumonia
Black vs. White – Rela*ve Risk
*
**
**
**
*
*p < 0.05
30 Day Readmissions for 10 Common Reasons for Hospitaliza.on
30 Day Readmissions for 10 Common Reasons for Hospitaliza.on
1.56
1.34
1.52
1.10
0.86
1.16
1.00
1.30
1.07
0.90
1.04
0.0 0.5 1.0 1.5 2.0 2.5
Cesarean Sec.on
Vaginal Delivery
Total Joint Arthroplasty
Sep.cemia
Celluli.s
Gastroenteri.s -‐ Surgical
Gastroenteri.s -‐ Medical
Heart Failure and Shock
Chest Pain
Kidney Disease
Pneumonia
Hispanic vs. White – Rela*ve Risk
*
*
*
*
*p < 0.05
Facility Level Differences Diseases of the Circulatory System
0.5
0.7
0.9
1.1
1.3
1.5
1.7
1.9
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29
Black/White OR
13
Race and Ethnic Differences in Insurance
Status and Patient Sickness
Race, Ethnicity, and Payer in CT Findings from the CTHCS (N = 4,608)
Race/Ethnicity % Covered by Medicaid
Hispanic or La.no 28.2
White, Non-‐Hispanic 9.4
Black, Non-‐Hispanic 24.3
Asian, Non-‐Hispanic 1.4
Other Race or Mul.-‐Racial 28.9
3x
2.5x
1.28
1.30
1.44
1.51
1.53
1.48
1.40
1.45
2.16
1.17
1.60
0.0 0.5 1.0 1.5 2.0 2.5
Cesarean Sec.on
Vaginal Delivery
Total Joint Arthroplasty
Sep.cemia
Celluli.s
Gastroenteri.s -‐ Surgical
Gastroenteri.s -‐ Medical
Heart Failure and Shock
Chest Pain
Kidney Disease
Pneumonia
Hispanic vs. White
All Stat Sig except Kidney Disease
30 Day Readmissions – Comparing Medicaid to Private Payers
Race and Ethnic Differences in Chronic Disease in CT Findings from the CDC
Race and Ethnic Disparities in 30 Day
Readmissions
The Role of Payer and Patient Sickness
30 Day Readmissions Comparing Black – White Differences with and without controls for payer and comorbidi.es
0.0 0.5 1.0 1.5 2.0 2.5
Cesarean Sec.on
Vaginal Delivery
Total Joint Arthroplasty
Sep.cemia
Celluli.s
Gastroenteri.s -‐ Surgical
Gastroenteri.s -‐ Medical
Heart Failure and Shock
Chest Pain
Kidney Disease
Pneumonia
Model 1 Model 2
* ** *
* ***
**
*
*
*p < 0.05
30 Day Readmissions Comparing Hispanic – White Differences with and without controls for payer and comorbidi.es
0.0 0.5 1.0 1.5 2.0 2.5
Cesarean Sec.on
Vaginal Delivery
Total Joint Arthroplasty
Sep.cemia
Celluli.s
Gastroenteri.s -‐ Surgical
Gastroenteri.s -‐ Medical
Heart Failure and Shock
Chest Pain
Kidney Disease
Pneumonia
Model 1 Model 2
* **
*
*
*
*
*
*
*p < 0.05
*
If Not Payer or Patient Sickness, Then What?
Exploring the Connection to Outpatient Care
Methods • Analyzed Medicaid hospitaliza.ons for PQI 01 & PQI 03 (Short and Long Term Diabetes Complica.ons)
• One year of Medicaid claims data: – Iden.fied hospitaliza.ons between 9/1/2012 and 2/28/2013
– Documented office visits 6 months prior to admission
2012 Hospitaliza.on among Medicaid Pa.ents PQI 01 & PQI 03 Short and Long Term Diabetes Complica.ons
Office Visit # of
Admissions % of
Admissions
PCP Only 198 31%
Endocrinologist Only 8 1%
PCP & Endocrinologist 8 1%
Neither 435 67%
Totals 649 100%
Conclusions • Substan*al race and ethnic dispari*es in 30 day readmissions across range of condi*ons – Especially for TJA, childbirth
• Substan*al variability in readmissions across payers – Rates much higher among Medicaid recipients compared to privately insured
• BUT – race differences not totally aYributable to payer or pa*ent sickness – YES: Chest Pain, GI, Celluli.s – NO: TJA, Childbirth, Sep.cemia, Heart Failure
Discussion • Explana*ons for dispari*es in readmission typically fall into 3 interrelated categories:
-‐ Hospital performance & prac.ces -‐ Pa.ent characteris.cs and behavior (e.g., Tx adherence) -‐ Community resources (e.g., access to outpa.ent care)
• This analysis touches on a few of these issues – pa.ent sickness, SES, payer, access to and u.liza.on of outpa.ent care
• Addi*onal presenta*ons, breakout sessions: Data will s.mulate discussion that leads to ac.on
Questions?
References q Kim H, Ross JS, Melkus GD, Zhao Z, Boockvar K. Scheduled and unscheduled hospital readmissions
among pa.ents with diabetes. Am J Manag Care. Oct 2010;16(10):760-‐767. q Joynt KE, Orav EJ, Jha AK. Thirty-‐day readmission rates for Medicare beneficiaries by race and site
of care. JAMA. Feb 16 2011;305(7):675-‐681. q McHugh MD, Carthon JM, Kang XL. Medicare readmissions policies and racial and ethnic health
dispari.es: a cau.onary tale. Policy Polit Nurs Pract. Nov 2010;11(4):309-‐316. q Alexander M, Grumbach K, Remy L, Rowell R, Massie BM. Conges.ve heart failure hospitaliza.ons
and survival in California: paGerns according to race/ethnicity. Am Heart J. May 1999;137(5):919-‐927.
q Ganesh S, Rogal SS, Yadav D, Humar A, Behari J. Risk factors for frequent readmissions and barriers to transplanta.on in pa.ents with cirrhosis. PLoS One. 2013;8(1):e55140.
q Menachemi N, Chukmaitov A, Brown LS, Saunders C, Brooks RG. Quality of Care Differs by Pa.ent Characteris.cs: Outcome Dispari.es Ater Ambulatory Surgical Procedures. American Journal of Medical Quality. November 1, 2007 2007;22(6):395-‐401.
q Rodriguez F, Joynt KE, Lopez L, Saldana F, Jha AK. Readmission rates for Hispanic Medicare beneficiaries with heart failure and acute myocardial infarc.on. Am Heart J. Aug 2011;162(2):254-‐261 e253.
q Smedley, Brian D., Adrienne Y. S.th, and Alan R. Nelson, eds. Unequal treatment: confronBng racial and ethnic dispariBes in health care (with CD). Na.onal Academies Press, 2009.