WILL EQUITY BE ACHIEVED THROUGH HEALTH CARE REFORM? John Z. Ayanian, MD, MPP

WILL EQUITY BE ACHIEVED THROUGH HEALTH CARE REFORM? John Z. Ayanian, MD, MPP Brigham and Women s Hospital Harvard Medical School Harvard School of Public Health BWH Patient-Centered Outcomes Seminar April 23, 2013

OBJECTIVES 1) Introduce conceptual framework for understanding health care disparities 2) Highlight key research & reports on disparities in health care 3) Consider policy approaches for achieving equity through health care reform under ACA: Insurance coverage Coordination of care Performance measurement & feedback

CROSSING THE QUALITY CHASM REPORT Equitable care is one of 6 core aims for improving health-care systems *Along with effective, efficient, timely, safe & patient-centered care Institute of Medicine, 2001

DIFFERENCES, DISPARITIES & DISCRIMINATION Difference Clinical Need, Appropriateness, Patient Preferences Healthcare Systems & Legal / Regulatory Systems Discrimination: Bias, Stereotyping, & Uncertainty Disparity Institute of Medicine, 2003

Conceptual Model for Health Care Disparities Patient Factors: Race/ethnicity, age, insurance, SES Community Factors Quality of Care Received Outcomes Physician, Hospital & Health System Factors Ayanian World J Surg 2008

1) Describing the problem 2) Understanding mediators & outcomes HISTORY OF RESEARCH ON HEALTH CARE DISPARITIES 1985 1990 1995 2000 2005 2010 2015 2020 3) Determining? effective interventions Ayanian World J Surg 2008

THE CAST Simulated Patients with Chest Pain in a Video About Referral for Cardiac Catheterization Schulman et al. N Engl J Med 1999

HEADLINES ABC Nightline A Recent Study Shows that Doctors Diagnose Black and White Patients Differently Los Angeles Times Heart Study Points to Race, Sex Bias New York Times Doctor Bias May Affect Heart Care, Study Finds

Pittsburgh Post-Gazette

Setting the Context Minority Americans less likely than whites to get many effective medical & surgical services Increasing attention to reasons for differences: Patient preferences? Clinical factors? Socioeconomic factors? Communication and trust? Physician bias? Fragmented systems of care? Research vital for clinical & policy solutions

Trends in US Life Expectancy by Race & Sex,1975-2003 Harper et al. JAMA 2007

Chicago Sun-Times 2007

UNEQUAL TREATMENT REPORT Racial & ethnic disparities in care associated with worse outcomes, thus unacceptable Disparities reflect broader inequality & discrimination in American society Institute of Medicine, 2003

Insurance Coverage

The Diagnosis of Exclusion Unfortunately you have what The New Yorker we call no insurance.

Jim Waterhouse s Story * Diagnoses: hypertension, diabetes, heart disease, sleep apnea While uninsured in his early 60 s: Reduced visits to primary physician and cardiologist Reduced glucose monitoring Ignored worsening dyspnea Discontinued CPAP for sleep apnea Deferred colonoscopy for new anemia Complications: Delayed visit to primary MD led to $46,000 admission * Hayes et al. Too Sick to Work, Too Soon for Medicare, Commonwealth Fund 2007

I tried to put off medical care until I became eligible for Medicare, which complicated my condition. --Jim Waterhouse, after gaining Medicare coverage

Rates of Uninsurance by Race & Ethnicity US Non-Elderly Population, 2011 Uninsured Rate (Percent) 40 30 20 10 0 18.0 General population under age 65 13.0 Non- Hispanic White 18.0 Asian American and Pacific Islander 21.0 African American 27.0 American Indian and Alaska Native 32.0 Hispanic Current Population Survey, 2012

90 Access to Effective Primary Care Before & After Medicare Eligibility Cholesterol Testing for Adults with Diabetes or Hypertension 84.1 80.4 88.0 % 60 47.2 30 0 Before 65 After 65 Previously Uninsured Previously Insured McWilliams, Zaslavsky, Meara, & Ayanian. JAMA 2003

Nonwhite White Difference in HbA1c (%) 1.6 Glucose Control for Diabetes US Black & Hispanic vs. White Adults NHANES, 1999-2006 1.4 1.2 1 0.8 Difference from pre-65: - 0.7% (P<0.001) 0.6 0.4 0.2 0 40-59 60-64 65-69 70-85 Age McWilliams, Meara, Zaslavsky & Ayanian. Ann Intern Med 2009

Worse Health Score Better Trends in Health of Adults with Cardiovascular Disease or Diabetes 30 24 23 22 Differential Change in Trend: +0.26 (P=0.006) 21 20 19 18 17 0 16 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 55 57 59 61 63 65 67 69 71 Age Uninsured Before 65 Continuously Insured Before 65 Expected Trend McWilliams, Meara, Zaslavsky & Ayanian. JAMA 2007

Benefits & Challenges of Health Care Reform 16 million low-income adults newly eligible for Medicaid 16 million middle-income adults newly eligible for private coverage through employers or insurance exchanges ************************************************************** How many states will expand Medicaid? How quickly will people be enrolled in new coverage? Will they have access to effective primary & specialty care, medications, and preventive services? Insurance coverage alone won t eliminate disparities!

Cascade of Voltage Drops From Insurance to High-Quality Health Care Eisenberg & Power, JAMA 2000

Coordination of Care

PATIENT RATINGS OF OVERALL CANCER CARE AS EXCELLENT / VERY GOOD Northern California, 1999-2000 Percent 100 80 60 40 20 0 82 74 67 61 White Black Hispanic Asian English Non- English Ayanian et al. J Clin Oncol 2005 Coordination of care strongest factor in overall quality ratings in all groups Race/Ethnicity 81 52 Language

HOW DO FRAGMENTED SYSTEMS OF CARE CONTRIBUTE TO DISPARITIES? System deficits affect all segments of society, but especially non-white patients Disadvantaged patients fall through the cracks in complex systems of care Small disparities in multi-step processes create moderate to large disparities overall Disparities arise even when providers well intentioned

PATIENT NAVIGATORS Intuitively appealing to improve coordination of care Promising model for community engagement and serving disadvantaged communities Rigorous evaluation will be essential to build support and secure funding Promising results for cancer screening navigators in Cambridge Health Alliance and New York City

Randomized Trial of Patient Navigators for Colorectal Cancer Screening Cambridge Health Alliance Community Health Centers % Screened Intervention Control P All patients 33.6 20.6 <0.001 Race White 33.9 16.5 0.003 Black 39.7 16.7 0.004 Language English 26.8 21.4 0.35 Non-English 39.8 18.6 <0.001 Lasser, Ayanian, et al. Arch Intern Med 2010

New York City Department of Health public service announcement

Health Systems & Communities Partnering to Promote Screening & Eliminate Disparities New York City, 2003-2007 In my career, 3 things have surprised me: how quickly TB cases came down in NYC, how quickly tobacco use came down in NYC, and how quickly colon cancer screening went up in NYC, Dr. Frieden said. Even more surprising is the closing of the race and ethnicity gap. 100 2003 80 White Black Hispanic 60 48 40 35 38 20 0 2007 100 White Black Hispanic 80 62 64 63 60 40 20 0 New York Times, 2008

Use of Mammography by Race/Ethnicity Among Women Ages 65-69 in Medicare HMOs & Traditional Medicare, 2009 Percent P<0.001 P<0.001 P=0.003 P<0.001 P<0.001 P<0.001 Ayanian et al. 2013 (under review)

Performance Measurement

ADDRESSING DISPARITIES THROUGH QUALITY PERFORMANCE MEASURES Disparities are quality problem Current data inadequate Fiscella et al. JAMA 2000 Stratify quality measures by race/ethnicity & socioeconomic position

Percent BETA-BLOCKER USE AFTER ACUTE MI Medicare Managed Care, 1997-2002 100 80 60 40 20 0 11.8 4.7 White Black White-Black Gap -2.0 3.2 3.1 0.4 1997 1998 1999 2000 2001 2002 *Overall quality improved and racial disparity eliminated Trivedi, Zaslavsky, Schneider & Ayanian. N Engl J Med 2005

Percent CHOLESTEROL CONTROL FOR HEART DISEASE Medicare Managed Care, 1999-2002 100 80 60 White Black White-Black Gap 40 20 13.2 12.7 13.3 16.3 0 1999 2000 2001 2002 *Overall quality improved but racial disparity persisted Trivedi, Zaslavsky, Schneider & Ayanian. N Engl J Med 2005

CHOLESTEROL CONTROL FOR HEART DISEASE Medicare HMOs by Region, 2010 Percent Black (N) 1,617 1,470 3,321 3,340

Baseline Racial Disparities in Diabetes Care Harvard Vanguard Medical Associates, 2006 White (n=4,858) Black (n=2,699) P Process measures, % Annual HbA1c test Annual LDL test Annual BP check 87 83 95 89 83 94 0.14 0.99 0.04 Outcomes measures, % HbA1c < 7% LDL < 100 mg/dl BP < 130/80 mmhg 46 55 32 40 43 24 <0.001 <0.001 <0.001 Sequist, Ayanian et al. Ann Intern Med 2010

HVMA Diabetes Intervention Components 31 primary care teams randomized to usual care or: Cultural competency training 1 or 2 day course (89% team attendance) Monthly educational materials developed from patient surveys and focus groups + Clinical performance feedback Monthly distribution Race-stratified (HbA1c<7%, LDL<100, BP<130/80) Provider-specific (benchmarked to practice)

Sample Diabetes Performance Report

% Very or Somewhat Often Impact on Clinician Awareness at HVMA Do racial disparities in diabetes care exist in. 100 80 60 40 20 0 P=0.003 P=0.02 P=0.04 82 59 All health centers 70 51 Your health center 63 43 Your patients Intervention Control Sequist, Ayanian et al. Ann Intern Med 2010

Disease Control Among Black Patients with Diabetes One-Year Follow-up Intervention (%) Control (%) P HbA1c <7% 49 47 0.36 LDL <100 mg/dl 50 50 0.68 BP <130/80 mmhg 23 24 0.73 Sequist, Ayanian et al. Ann Intern Med 2010

% 'Very Effective' Clinicians Views of Study Intervention Are these strategies effective to reduce racial disparities? 50 40 P=0.04 P=0.05 34 33 30 20 10 17 11 0 Sequist, Ayanian et al. Ann Intern Med 2010 Cultural Competency Training Intervention Performance Feedback Control

Engaging Physicians & Health Care Systems physicians & the health care systems in which they operate are key to making sure that all patients get the very best care. www.kff.org/whythedifference

LESSONS FOR HEALTH CARE REFORM (1) Performance reports by race/ethnicity 2009 IOM report underscores importance of consistent reporting of quality data by race and ethnicity Broad quality improvement efforts reduce disparities in processes of care Disparities diminish in more organized systems: e.g. Medicare HMOs, integrated medical groups

LESSONS FOR HEALTH CARE REFORM (2) Focused interventions in health systems & communities required to achieve equity in health outcomes: Universal insurance coverage is essential foundation More effective coordination of care Community engagement and policy interventions to address social determinants of health outcomes ACO s, patient-centered medical homes & managed care plans must strive to meet these challenges

Harvard Catalyst Health Disparities Research Program Catalyst.harvard.edu/programs/disparities

Thank You