Racial and Ethnic Disparities in Health and Health Care

1. Racial and Ethnic Disparities in Health and Health Care Olveen Carrasquillo, MD, MPH Assoc Prof Medicine & Health Policy Director, Columbia Center for the Health of Urban Minorities

2. Outline What is race / ethnicity + demographics Genetics (very little) Disparities in health (too much) Disparities in health care (too much) Disparities Research at CUMC / CHUM Interventions that work CBPR

3. Do not take notes-Email me!!!! oc6@columbia.edu

4. Resources for Health Disparities www.kaiseredu.org Tutorials: Race, Ethnicity and Health Care Reference Libraries �Immigrants: Coverage & Access to Care�Race, Ethnicity, and Health Care: The Basics www.cmwf.org Care of underserved, cultural competency, health disparities http://www.improvehealthcare.org/ Health disparities (cases) www.ahrq.gov Health disparities report ( (

6. OMB Directive No. 15  Separate questions used for reporting race and ethnicity Ethnicity (1st)- cultural identity -- Hispanic or Latino -- Not Hispanic or Latino Race- physical characteristics/ geographic origin -- American Indian or Alaska Native -- Asian -- Black or African American -- Native Hawaiian or Other Pacific Islander -- White Respondents have the option of selecting one or more racial designations. Reporting: Except when the collection involves a sample of such size that the data on the smaller categories would be unreliable, or when the collection effort focuses on a specific racial or ethnic group.

7. Combined format Six minimum categories: -- American Indian or Alaska Native -- Asian -- Black or African American -- Hispanic or Latino -- Native Hawaiian or Other Pacific Islander -- White Never use whites vs nonwhites

8. Definitions White. A person having origins in any of the original peoples of Europe, the Middle East, or North Africa Black or African American. A person having origins in any of the black racial groups of Africa. Terms such as "Haitian" or "Negro" can be used in addition to "Black or African American." West Indian, African American Indian or Alaska Native. A person having origins in any of the original peoples of North and South America (including Central America), and who maintains tribal affiliation or community attachment. -- Native Hawaiian or Other Pacific Islander. A person having origins in any of the original peoples of Hawaii, Guam, Samoa, or other Pacific Islands. -- Asian. A person having origins in any of the original peoples of the Far East, Southeast Asia, or the Indian subcontinent including, for example, Cambodia, China, India, Japan, Korea, Malaysia, Pakistan, the Philippine Islands, Thailand, and Vietnam. -- Hispanic or Latino. A person of Cuban, Mexican, Puerto Rican, Cuban, South or Central American, or other Spanish culture or origin, regardless of race. The term, "Spanish origin," can be used in addition to "Hispanic or Latino."

9. Olveen comment: never compare across races!!! use a referent usually NHWS Not chi square across groups It is Hisp vs NHWs and Blacks vs NHWs Can do Hisp vs Blacks if part of hypothesis

14. Citizenship Status of Latinos

16. Dominicans in US US Census �official figures�- short form 1980= 190,280 1990= 520,151 2000= 764,945 Ancestry Question 2000= 908,531 Ancestry + Place of Birth 2000= 1,111,142

17. Projections Latino Elderly Pop Currently 4.6 million age 55+ (7% pop) 2.4 million age 65+ (6% pop) 2050 22 million age 55+ (18% pop) 13 million age 65+ (13% pop)

18. Birth rate: United States, 2000

20. Genetics Specific differences in genes are still under study In the case of diabetes in Mexican-Americans, genetic admixture seems to have a clear role American indigenous people have very high prevalences of diabetes compared to Whites There is a paucity of data regarding genetic differences among Latino subgroups

22. Studying genetics in AAs Can trace certain ancestral alleles to Central Africa Can follow single gene mutations that are Mendelian Not very meaningful in genetically complex disorders Hypertension in AAs Htn in Africa ? Gene environment intercation K channel Related to higher salt diet

23. Why study Genetics in Latinos There is no Latino gene!!!!!! Latinos as a genetic group not cw evolution Latinos very genetically homogeneous PRs very different from Mexicans Mexican Spaniards very diff from Mayans Good model of genetics overall Large families Close connections Higher disease burden: Diabetes, Alzheimer�s Can study gene environment interactions Barrios Native countries

25. More on DM Pima Indians in Mexico vs US Pimas Why are poor whites genetically at risk for diabetes versus rich whites

26. What are disparities Differences in the health of racial or ethnic minorities versus non Hispanic whites not due to known reasons such as income, location etc? Differences in quality of health care received by racial or ethnic minorities versus non Hispanic whites Not due to clinical needs or preferences ? Not due to known reasons such as income ?

27. Why are disparities important? While great gains have occurred in improving overall health and reducing health disparities, the persistence of racial, ethnic, economic, or other social inequalities in health is unacceptable. Eliminating health disparities in New York City would save thousands of lives each year. Thomas R. Frieden, MD, MPH Commissioner, New York City Department of Health and Mental Hygiene

28. Infant Mortality Statistics from the 2003 Period Linked Birth/Infant Death Data Set 

29. Infant, neonatal, and post-neonatal deaths and mortality rates 2003 linked file 

31. Death Rate by Ethnicity, US 2000 W B L A/PI Heart Disease 130 191 89 72 Stroke 25 44 20 24 Diabetes 12 29 19 9 Age-adjusted per 100,000 NCHS

34. Prevalence of DM and IFG

36. Cancer Mortality Rates - Men

37. Cancer Mortality Rates Women

41. Rates (per 100,000 population) of AIDS, 2004�50 states and the District of Columbia

42. Rates (per 100,000 population) of HIV/AIDS, 2004�33 states with confidential name-based HIV infection reporting 

43. Other Important Health Differences Strokes/ HTN in African Americans Substance Abuse/ Violence Dental health Mental Health

44. Life Expectancy at Birth Born 2000 white 77.4 yrs Black 71.7 yrs Age Adjusted Death Rates (per 100,000) NHW 855 Black 1,126 Hispanic 670 Asian 517 US Health, 2003

47. Latino paradox Many studies link poverty to poor health Latinos are poorer than African Americans but have lower overall mortality rates, death from cancer and heart disease, infant mortality than AAs/ whites But--acculturation leads to poorer health outcomes

48. Latino paradox What causes the paradox? Theories: �Healthy immigrant�; �salmon� hypotheses Strong social/family networks Low tobacco and ETOH use especially in women Religiosity Traditional healing practices Traditional diet

51. Disparities in Health Care Disparities in Access to Care Health Insurance Regular Provider Disparities in Receipt of Quality Health Care

53. Forty-Four Million Uninsured In 1998, 44 million Americans were uninsured, more than at any time since the passage of Medicare and Medicaid in the mid 1960's. While 44 million were uninsured at any one time during the year, about 55 million people lacked coverage for at least one month. Over the course of 28 months more than 67 million are uninsured for at least one month. Hence, about one quarter of the population has experienced a recent bout of "uninsurance.� The situation is particularly bad for young people. About 11.1 million (15.0%) children under 18 are uninsured at any one time. Among young adults age 18-24, 30.1% are uninsured. During 1995-1996 23.1 million of the total of 70.8 million children in the U.S. went without health insurance for at least one month. � Poorer families have the highest uninsurance rates, but even the well-to-do are at risk. 25.2% of persons in households with annual incomes below $25,000 were uninsured in 1998, vs. 8.3% of those with household incomes greater than $75,000. People with serious illnesses or disabilities depend predominantly on public programs. While 46% of health costs for persons without disability are paid by private insurance, private insurance accounts for only 27% of spending for those with disabilities; 18% is paid out-of-pocket, 30% by Medicare, 10% by Medicaid, 10% by other public programs, and 4% by other sources. More than half of all HIV positive Americans are covered by public insurance policies, while private insurance covers only 19%; 29% are uninsured. � Uninsurance rates are highest in the South and West. Texas had the highest rate, 24.5%. Arizona, California. Mississippi, Nevada, and New Mexico also had uninsurance rates of 20% or higher. Hawaii, Iowa, Minnesota, Nebraska, Rhode Island and Vermont had the lowest rates, 9.0% to 10.0%. Forty-Four Million Uninsured In 1998, 44 million Americans were uninsured, more than at any time since the passage of Medicare and Medicaid in the mid 1960's. While 44 million were uninsured at any one time during the year, about 55 million people lacked coverage for at least one month. Over the course of 28 months more than 67 million are uninsured for at least one month. Hence, about one quarter of the population has experienced a recent bout of "uninsurance.� The situation is particularly bad for young people. About 11.1 million (15.0%) children under 18 are uninsured at any one time. Among young adults age 18-24, 30.1% are uninsured. During 1995-1996 23.1 million of the total of 70.8 million children in the U.S. went without health insurance for at least one month. � Poorer families have the highest uninsurance rates, but even the well-to-do are at risk. 25.2% of persons in households with annual incomes below $25,000 were uninsured in 1998, vs. 8.3% of those with household incomes greater than $75,000. People with serious illnesses or disabilities depend predominantly on public programs. While 46% of health costs for persons without disability are paid by private insurance, private insurance accounts for only 27% of spending for those with disabilities; 18% is paid out-of-pocket, 30% by Medicare, 10% by Medicaid, 10% by other public programs, and 4% by other sources. More than half of all HIV positive Americans are covered by public insurance policies, while private insurance covers only 19%; 29% are uninsured. � Uninsurance rates are highest in the South and West. Texas had the highest rate, 24.5%. Arizona, California. Mississippi, Nevada, and New Mexico also had uninsurance rates of 20% or higher. Hawaii, Iowa, Minnesota, Nebraska, Rhode Island and Vermont had the lowest rates, 9.0% to 10.0%.

57. Change in # Uninsured (1,000)

58. NHWs: No longer a majority of the uninsured:Trends in composition of uninsured population 1987 NHWS 58% Blacks 19% Hispanics 19% Asians 3% 2006 NHWS 45% Blacks 16% Hispanics 32% Asians 5%

62. Health Insurance: Summary Most important determinant of access to the health care system Glaring, horrible racial/ethnic disparities We need: NATIONAL HEALTH INSURANCE Does not explain all of disparities

64. Racial/Ethnic Disparities in access to Cardio-Vascular procedures CV disease is number one killer in America Useful to look at because it addresses a continuum of care Multiple Reviews Annals of Internal Medicine 2001;135:352-366

65. Racial/Ethnic Disparities in access to Cardio-Vascular procedures 27 studies using administrative data OR for blacks getting cath (.41-.94) CABG (.23-.68) 28 studies with detailed clinical data Cath (.03-.85) CABG (.22-.68) 14 studies examining why not done Some due to pt refusal �education imp Physician bias still caused a lot of variation

68. Implicit Bias among MDs Implicit Association Tests Implicit stereotypes of blacks as less cooperative 58% offered TPA to whites versus 42% blacks As pro-white bias increased disparity increased Unconscious bias contributes to disparities

69. Sex and Racial Differences in the Management of Acute Myocardial Infarction, 1994 through 2002 Rates of reperfusion therapy, coronary angiography, and in-hospital death after myocardial infarction, vary according to race with no evidence that the differences have narrowed in recent years. CABG OR (.74 black men, .69 black women) Use of aspirin and beta-blockers showed much less variation NEJM 2005;353:671-82

70. Is it just the heart and lungs??

71. Is it just the heart and lungs??

76. Trends in the Quality of Care and Racial Disparities in Medicare Managed Care  NEJM 2005;353:692-700.

77. Separate and Unequal Health Care Systems Black and white patients to a large extent are treated by different physicians. 80% of visits by African-Americans were made to 22% of physicians. Such doctors were less likely to be board certified and reported less access to specialists, diagnostic procedures, and non-emergency hospital admissions. NEJM 2004;351:575-84

78. Separate and Unequal Health Care Systems 28,000 patients in New York State who underwent coronary artery bypass graft surgery in 1996-97 African Americans were treated by surgeons with risk-adjusted mortality rates 13.8 percent higher than surgeons who treated whites. Hospital was the primary factor explaining the disparities, suggesting that physician referral patterns may be important determinants of where minorities receive treatment.

79. Nursing Homes Lower-tier facilities 85% or more residents covered by Medicaid, difficulty retaining staff members, have few financial resources and often restrain patients 40% of African-American nursing home residents nationwide reside in lower-tier homes, compared with 9% of white nursing home residents Competition works only in markets where consumers have choices, and unfortunately, many of these nursing home residents don't have much choice

81. AHRQ Disparities Report

84. What can de done:  Address social determinants of health Poverty, education, housing, environment, social welfare issues Single Payer Universal Insurance What else????

85. The IOM Report: A Landmark Document

86. Health care workforce diversity Latinos and African Americans account for 25% of the U.S. Population�. But represent only 6% of practicing physicians. In NYC , 54% of the population of New York City is black or Hispanic yet in 2002-03 only 13.5% of the 892 entrants to the six allopathic medical schools in NYC were black or Hispanic Let us now turn to the issue of health care workforce diversity read aboveLet us now turn to the issue of health care workforce diversity read above

87. Schools Self-Assessment of Success in Meeting Diversity Goals

88. Cultural Competency

89. Disparities Research Advocacy Oriented Research Race / Ethnicity stuff Acculturation Issues

90. Insurance Coverage among Non-citizen Latino immigrants

91. Health Care Expenditures of Immigrants

93. Results

94. Hispanics, Race and Life Chances� How race counts for Hispanic Americans: John R. Logan. Sage Race Relations Abstracts 2004;19:7-19 �On the basis of social similarity, if it is necessary to combine Hispanic blacks with another group, there is now better data to support the classification of black Hispanics as black rather than as Hispanics�

95. Racial classification among Hispanics and health and well-being: A Conceptual ModelAmerican Journal of Public Health 2005;95:379-81 This model was published early on 2005 and has been modified through feedback in presentations and conversations with colleagues. And as I mentioned before, this model reflects the interaction of factors through different levels. This model was published early on 2005 and has been modified through feedback in presentations and conversations with colleagues. And as I mentioned before, this model reflects the interaction of factors through different levels.

96. Crude and adjusted odds ratios (OR)* for diabetes by race/ethnicity among adults =18 years of age: NHIS 2000- 2003 After adjustment for selected covariates (Model 3), Hispanics, regardless of their race, were more likely to report having diabetes than non-Hispanic whites. Specifically, Hispanic whites and blacks were 1.56 and 2.64 times, respectively more likely to report having diabetes than non-Hispanic whites after adjusting for selected covariates. The odds ratio for non-Hispanic blacks was 1.45 (95% CI 1.29-1.64). Non-Hispanic Black 1.46 (1.29-1.64) All Hispanic 1.60 (1.38-1.87) When Hispanic blacks were excluded: Non-Hispanic Black 1.46 (1.29-1.64) All Hispanic 1.57 (1.35-1.82) White Hispanics 1.55 (1.31-1.82) Black Hispanics 1.64 (0.66-4.05) Hispanic Black 1.76 (0.76-4.08) Non-Hispanic Black 1.43 (1.27-1.61) P-Interaction: 0.72 After adjustment for selected covariates (Model 3), Hispanics, regardless of their race, were more likely to report having diabetes than non-Hispanic whites. Specifically, Hispanic whites and blacks were 1.56 and 2.64 times, respectively more likely to report having diabetes than non-Hispanic whites after adjusting for selected covariates. The odds ratio for non-Hispanic blacks was 1.45 (95% CI 1.29-1.64). Non-Hispanic Black 1.46 (1.29-1.64) All Hispanic 1.60 (1.38-1.87) When Hispanic blacks were excluded: Non-Hispanic Black 1.46 (1.29-1.64) All Hispanic 1.57 (1.35-1.82) White Hispanics 1.55 (1.31-1.82) Black Hispanics 1.64 (0.66-4.05) Hispanic Black 1.76 (0.76-4.08) Non-Hispanic Black 1.43 (1.27-1.61) P-Interaction: 0.72

97. When compared to non-Hispanic Whites, Hispanic Whites were less likely to have hypertension before and after adjustment for selected covariates. However, Hispanic blacks were not different from non-Hispanic whites. We repeated the analyses by ethnicity and race: The pattern for non-Hispanics and Hispanics was very similar. Specifically, blacks exhibited higher odds of hypertension than whites. Moreover, Hispanics whites had lower odds of hypertension than non-Hispanic whites. However, there was no difference between non-Hispanic and Hispanic blacks. When compared to non-Hispanic Whites, Hispanic Whites were less likely to have hypertension before and after adjustment for selected covariates. However, Hispanic blacks were not different from non-Hispanic whites. We repeated the analyses by ethnicity and race: The pattern for non-Hispanics and Hispanics was very similar. Specifically, blacks exhibited higher odds of hypertension than whites. Moreover, Hispanics whites had lower odds of hypertension than non-Hispanic whites. However, there was no difference between non-Hispanic and Hispanic blacks.

98. Latino paradox Many studies link poverty to poor health Latinos are poorer than African Americans but have lower overall mortality rates, death from cancer and heart disease, infant mortality than AAs/ whites Latino Paradoxes Infant mortality Cardiovascular disease Cancer

99. Salmon Hypothesis:Mortality of Latinos vs NHWs

100. Latino Healthy Behaviors Latinos relative to non-Latino whites (controlling for SES) were less likely to smoke drink alcohol, But less likely to engage in any exercise more likely to have a high BMI

101. Acculturation Higher acculturation was associated with greater (After adjusting for age and SES) alcohol intake (esp higher educated women) smoking (women) BMI But greater likelihood of exercise

103. Is Acculturation Bad for Your Health? The Association between Acculturation Status and cardiovascular disease (CVD) risk factors.

104. What Works IDEATel: a randomized, controlled trial comparing telemedicine case management to usual care= $30million In the intervention group (n = 844), mean HgbA1c improved over one year from 7.35% to 6.97% In the usual care group (n = 821) mean HgbA1c improved over one year from 7.42% to 7.17%. Net HgbA1c, 0.18% (p = 0.006) Net LDL cholesterol 9.5 mg/dL (p < 0.001).

105. A Systematic Review of Interventions to Improve Diabetes Care in Socially Disadvantaged Populations 7 databases searched for articles 1986- 2004, 17 studies found Interventions that were consistently associated with the largest negative outcomes: those that used mainly didactic teaching focused only on diabetes knowledge.

106. What works? Features most consistent positive effects cultural tailoring of the intervention community educators or lay people leading the intervention one-on-one interventions with individualized assessment and reassessment incorporating treatment algorithms focusing on behavior-related tasks providing feedback high-intensity interventions (>10 contact times) delivered over a long duration ( 6 months)

107. What works:Community Based Participatory Research

108. Why Should Investigators be Interested in Community Engagement Need for recruitment and retention Increasing Political Advocacy for Community Participation in Research: NIH Roadmap Hispanic Community Health Study Clinical Translation Science Award Some areas will be mandated to have community input / collaboration genomics Increasing Community Sophistication

109. community-based research Research conducted in a community as a place or setting Research primarily driven by the academic institution May address areas of importance for community Limited involvement of community members Study subjects Recruiters, community liaisons, RA�s

110. COMMUNTIY BASED PARTICIPATORY RESEARCH (CBPR) Collaborative approach to research that equitably involves all partners in the research process and recognizes the unique strengths that each brings Begins with a research topic of importance to the community Has the aim of combining knowledge with action and achieving social change to improve health outcomes [and eliminate health disparities]

111. Mandates Participatory models of research, in which communities are actively engaged in the research process through partnerships with academic institutions, have become central to the national prevention / disparities research agenda Calls by IOM, NIH, CDC, AHRQ�� Increasing Evidence Base for CBPR

112. PRINCIPLES OF CBPR Facilitates collaborative, equitable involvement of all partners in all phases of the research. collaborative partnership in which all parties participate as equal members share control over all phases of the research process, e.g., problem definition, data collection, interpretation of results, and application of the results to address community concerns

113. CBPR Involves a collaborative partnership in a cyclical, iterative process in which communities of identity play a lead role in identifying community strengths and resources selecting priority issues to address collecting, interpreting, and translating research findings in ways that will benefit the community Emphasizes the reciprocal transfer of knowledge, skills, capacity and power. The focus of the partnership is driven by issues and concerns identified by members of the community of identity.

114. Challenges: Community Distrust of AHCs Traditional mistrust of research Guinea Pig phenomenon Abandonment �Not in loop� Failure to carry out with policy / interventions Not sharing $$$$$

115. Challenges Academic Distrust of Community Public relations Politics Fiscal Integrity Foreign Culture Community Capacity to conduct Research $$$$$$ / Indirect issues Evaluation

117. Consists of Nine Cores Administrative Core Research Training Core Community Action Core Health Disparities Core (Cultural Competency Core) Five Research Cores (Access to Care, Cardio-vascular Disease, Mental Health, Injury Prevention, Diabetes)

118. Community Core Specific AIMS: Integrate CBPR into each of CHUM Research Cores/ work of investigators Partial Success (depends on core leaders/ investigators) Some cores already doing CBPR Injury Prevention Most were doing excellent Community Linkages and continue doing so Access core, Diabetes Core Most were not doing and will not do CBPR Impractical to do CBPR in secondary data analysis Multi site NIH type Clinical Research hard add CBPR Highly Successful Investigators do not need to do CBPR

119. Develop Partnerships between CUMC Investigators and CBOs CBPR Clearing House / Match making Partial success Investigators present to planning council they decide to participate Often approach for recruiting after study funded Mothers using cocaine Subjects for HIV at risk studies Some success at investigators approach at start and share resources

120. What Happens When you decline Does your man shoot up? Are you having sex Raw?

123. 4 awards Senior Center Based Walking Club Outcomes of CQI intervention on client flow CHW led DM education for mothers with gestational diabetes Qualitative study on community perceptions of depression and genetics

125. NYS DOH OMH Using community based CHWs to recruit into cancer clinical trials Get 15 CHWS HIPAA/ IRB certified Train them on recruitment 101 See if they can include recruitment as part of their ongoing activities at CBO

126. How did NMPP compare to HICCC? HICCC 100% FTE 1 month: 1 completed outreach data forms NMPP 5 CHWs @ 5% effort = 25% FTE 3 month: 183 completed outreach data forms Alianza- 2 months, 120 forms

127. CHUM 2 Research: Project Overview Project 1: A randomized controlled clinical trial (RCT) of 360 poorly controlled diabetic patients aged 35-70 to examine the effectiveness of a community based Community Health Worker (CHW) intervention in addressing the ABCs of diabetes care (HgA1c, Blood Pressure[BP], Cholesterol). Project 2: A prospective study examining the impact of glycemic control on cognitive function among Latino elders with diabetes. Project 3: An RCT to evaluate the effect of a community-based comprehensive therapeutic lifestyle intervention that includes group sessions and motivational interviewing on blood pressure among Latino elders.

128. Random Thoughts Remember why community wants more minority investigators Issues not addressed by traditional researchers New ideas, new approaches Research to action/ advocacy Must know much more than others Not only your field but also demographics, other minority health/ disparities issues Minority tax- must pay some of it

129. E-mail me oc6@columbia.edu