Latino Health Latino Health Disparities: Disparities: A Cultural Paradox? A Cultural Paradox? Eliseo J. Pérez-Stable, M.D. Eliseo J. Pérez-Stable, M.D. Professor of Medicine Professor of Medicine Division of General Internal Medicine Division of General Internal Medicine Department of Medicine, UCSF Department of Medicine, UCSF October 27, 2005 October 27, 2005
Latino Health Disparities: A Cultural Paradox?. Eliseo J. Pérez-Stable, M.D. Professor of Medicine Division of General Internal Medicine Department of Medicine, UCSF October 27, 2005. Disparities and Differences. - PowerPoint PPT Presentation
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Latino Health Disparities: Latino Health Disparities: A Cultural Paradox?A Cultural Paradox?
Eliseo J. Pérez-Stable, M.D.Eliseo J. Pérez-Stable, M.D.Professor of MedicineProfessor of Medicine
Division of General Internal Medicine Division of General Internal Medicine Department of Medicine, UCSFDepartment of Medicine, UCSF
October 27, 2005October 27, 2005
Disparities and DifferencesDisparities and Differences
• Disparities implies a difference that Disparities implies a difference that demonstrates a disadvantage for a group demonstrates a disadvantage for a group that has been traditionally disenfranchisedthat has been traditionally disenfranchised
• Some differences may not be classified as Some differences may not be classified as disparities–White men have more CADdisparities–White men have more CAD
• Disadvantaged groups may have better Disadvantaged groups may have better outcomes for leading causes of death–outcomes for leading causes of death–Latinos and AsiansLatinos and Asians
Race or Ethnicity?Race or Ethnicity?• Racial categories fit geographic
origins of humans
• Ethnicity refers to self-identity with a national origin or cultural group
• Admixture may confound categories
• Census uses racial categories and subgroups and Hispanic ethnicity
• Self identification = gold standard
Social Class and Race/EthnicitySocial Class and Race/Ethnicity
• Race has been a substitute for Race has been a substitute for defining social class in the U.S.defining social class in the U.S.
• Gradient of health outcomes at all SES Gradient of health outcomes at all SES levels comparing Blacks and Whiteslevels comparing Blacks and Whites
• Education and income are insufficient Education and income are insufficient measuresmeasures of social classof social class
• Measures of wealthMeasures of wealth• Generation of social classGeneration of social class• Community measures for segregation, Community measures for segregation,
Definition of Epidemiologic Definition of Epidemiologic ParadoxParadox
• Outcomes are better than Outcomes are better than expected based on the known or expected based on the known or standard predictive risk factorsstandard predictive risk factors
• Low SES does not always Low SES does not always translate to worse outcomestranslate to worse outcomes
% LBW Rates by Ethnicity% LBW Rates by Ethnicity
23456789
101112131415
African American Latina White
Study
Fuentes-Afflick E and Lurie P, Arch Pediatr Adolesc Med 1997
Death Rate by Ethnicity, US 2000Death Rate by Ethnicity, US 2000
U.S. Asthma Mortality 1990-1995U.S. Asthma Mortality 1990-1995Average Annual Rates per Million
11.315
40.75 40.9
0
10
20
30
40
50
Mexican White African
American
Puerto Rican
Homa et al. 2000Homa et al. 2000
Adverse Demographic Profile Adverse Demographic Profile for Latinosfor Latinos
• Less household income on averageLess household income on average
• About 30% live in poverty and have less About 30% live in poverty and have less wealth at every level of incomewealth at every level of income
• Fewer average years of education and Fewer average years of education and proportion of college graduatesproportion of college graduates
• Fewer than half of Latinos 25 years or Fewer than half of Latinos 25 years or older completed high school compared older completed high school compared with 77% of Whiteswith 77% of Whites
• More single-parent householdsMore single-parent households
Source: McGinnis JM, Russo PG, Knickman, JR. Health Affairs, April 2002.
Adverse Social and Access FactorsAdverse Social and Access Factors
• Lower functional health literacyLower functional health literacy• Limited English proficiency–25%Limited English proficiency–25%• Lowest health insurance coverage– 40% Lowest health insurance coverage– 40%
between 18-64 y are uninsuredbetween 18-64 y are uninsured
• Mexicans have the lowest insurance Mexicans have the lowest insurance coverage of any national origin group coverage of any national origin group
• Less access to primary care MDLess access to primary care MD
• Twice as likely to report using ER as Twice as likely to report using ER as primary source of careprimary source of care
Access to Markets with Healthy Access to Markets with Healthy Foods for Diabetics in New YorkFoods for Diabetics in New York
• 173 stores in East Harlem and 152 stores in 173 stores in East Harlem and 152 stores in Upper East SideUpper East Side
• Had all 5 categories: 9% vs. 48%Had all 5 categories: 9% vs. 48%• More likely to live on a block with no store More likely to live on a block with no store
selling foods in E Harlem–50% vs. 24%selling foods in E Harlem–50% vs. 24%• Example of disparities in environmental Example of disparities in environmental
Diabetes, Hypertension Diabetes, Hypertension and Cigarette Smokingand Cigarette Smoking
Do these risk factors or Do these risk factors or conditions explain the conditions explain the
paradox?paradox?
Diabetes Prevalence in Latinos
• NHANES III: 20% Mexican Americans vs. 11% Whites have DM
• Increase of 20% to 35% in 15 years• Undiagnosed diabetes 4%• Up to half of Latinos unaware of DM• 95% of diabetes is type 2• Prevalence in Puerto Ricans similar
Disparities in Diabetes Treatment and Outcomes
• CDC report--compared to Whites, CDC report--compared to Whites, Latinos were less likely to have:Latinos were less likely to have:– Dilated eye exam(56% vs. 60%)Dilated eye exam(56% vs. 60%)– Foot exam (47% vs. 56%)Foot exam (47% vs. 56%)– A1C test (18% vs. 27%)A1C test (18% vs. 27%)
• Latinos have more LE amputationsLatinos have more LE amputations• Mexican Am have more retinopathyMexican Am have more retinopathy• More proteinuria and ESRDMore proteinuria and ESRD
Hypertension Control in Latinos - Have We Made Any Progress?
• HHANES 1982-8420% controlled at <140/90
• HHANES 1988-199224% controlled
• San Antonio and LaredoFewer aware, treated and controlled
• South Bronx 23% Puerto Ricans controlled
• Less knowledge about CAD prevention
Hypertension Awareness Hypertension Awareness and Control, 1999-2000and Control, 1999-2000
• Awareness: 58% Mex Am vs. 68% Whites• Only 50% of Mex Am men were aware• Similar awareness among women• Treatment rates lower: 39% vs. 59%• Control among those treated: 40% vs. 54%• Only 33% of Mex Am men at goal• +60 y more aware, treated, less control• Slow improvements in 1990s
Behavioral FactorsBehavioral Factors
• Less cigarette smokingLess cigarette smoking
• More alcohol consumed - menMore alcohol consumed - men
• Nutritional habits less healthyNutritional habits less healthy
• Less physical activity - womenLess physical activity - women
• More violence - DV plus More violence - DV plus
• Less adherence to medicationsLess adherence to medications
Cigarette Smoking in the U.S. – 2002Cigarette Smoking in the U.S. – 2002National Health Interview SurveyNational Health Interview Survey
% Men % Women
WhiteAfrican AmLatinoAsianAm Indian
25.527.122.719.040.5
21.818.710.8 6.540.9
8 years or less9-11 yrs schoolhigh school diplCollege degree
25.438.129.813.6
13.530.922.110.5
Below poverty 36.9 30.1
MMWR 2004;53:427-431
Biochemical Smokers in Biochemical Smokers in Mexican American LatinosMexican American Latinos
• Underreporting occurred in up to Underreporting occurred in up to 25% of Mexican American smokers25% of Mexican American smokers
• Former smokers misclassified - 11%Former smokers misclassified - 11%
• Never smokers misclassified in 4%Never smokers misclassified in 4%
• 12.1% of smokers had non-smoker 12.1% of smokers had non-smoker cotinine levelscotinine levels
• Cotinine measure may be betterCotinine measure may be better
Ethnic Differences in Serum Cotinine Levels: NHANES 3
>15 ng/mlpercent
≤15 ng/mlpercent
African Amssmokernon-smoker
962
498
Whitessmokernon-smoker
942
698
Mexican Amssmokernon-smoker
721
2899
JAMA 1998;280:135-139
Nicotine Metabolism in Blacks, Nicotine Metabolism in Blacks, Whites, Chinese and LatinosWhites, Chinese and Latinos
• Metabolic clearance of nicotine & cotinine Metabolic clearance of nicotine & cotinine in Latinos was similar to Whites, higher in Latinos was similar to Whites, higher among Blacks and lower among Chineseamong Blacks and lower among Chinese
• Intake of nicotine(mg) per cigarette:Intake of nicotine(mg) per cigarette:– Chinese: Chinese: 0.73 0.73 – Latinos: Latinos: 1.051.05– Whites Whites 1.101.10– Blacks Blacks 1.41 1.41
• Prevalence of smoking is lowerPrevalence of smoking is lower• Hypertension and lipids similarHypertension and lipids similar• Obesity more commonObesity more common• Physical inactivity more commonPhysical inactivity more common• Less BP & DM awareness and controlLess BP & DM awareness and control• Diabetes rate is 2-4 timesDiabetes rate is 2-4 times• Lower SES by income, educationLower SES by income, education• Fewer heart attacksFewer heart attacks• Fewer procedures to treat CADFewer procedures to treat CAD
CHD Prediction Scores By EthnicityColor in Framingham?
• Applied sex specific CHD functions to 6 ethnically diverse cohorts
• White and Black men and women prediction of CHD events works well
• Japanese & Latino men and American Indian men & women–risk is overestimated
• Adjust for different rates of risk factors and underlying rate of CHD
JAMA 2001; 286:180-7
Is culture a protective factor?Is culture a protective factor?• Lower heart disease mortality rates
despite higher or similar prevalence of cardiovascular risk
• Unidentified factors that are protective against chronic diseases
• More social support through community or social networks?
• Genetic factors?
Proposed Explanations of Proposed Explanations of ParadoxParadox
• Healthy immigrant effectHealthy immigrant effect• Salmon hypothesis–return to die Salmon hypothesis–return to die
at home and deaths not recordedat home and deaths not recorded• Misclassification of ethnicity in Misclassification of ethnicity in
diagnosis and deaths––Latinos diagnosis and deaths––Latinos misclassified as Whitesmisclassified as Whites
• Census undercounts (increase)Census undercounts (increase)
What About What About Acculturation?Acculturation?
Role of Acculturation?
• NHANES III: Mexican Americans born in the US and speaking Spanish have higher adjusted SBP than English speaking counter parts - 123.9 vs. 121.5 mm Hg
• US born Spanish speaking was significant in logistic regression models for men and women for SBP, BMI and current smoking
• Bicultural Latinos at highest risk?
Sundquist, AJPH 1999; 89:723
Are Latina Women at Higher Risk?Are Latina Women at Higher Risk?
• Women 25 to 64 years showed adjusted SBP higher for Mexican Americans in HANES III
• SBP was intermediate between Whites and African Americans
• Not observed for women 18 to 24 years of any ethnic group
Sacramento Area Latino Study on Aging: Sacramento Area Latino Study on Aging: Cohort StudyCohort Study
Study PopulationStudy Population– 1,789 Latinos aged 60+, Mexican ancestry (85%) 1,789 Latinos aged 60+, Mexican ancestry (85%) – Mean age at baseline: 71 (60-101); 58% womenMean age at baseline: 71 (60-101); 58% women– 51% born in Mexico or another Latin American 51% born in Mexico or another Latin American
country and were Spanish speakingcountry and were Spanish speaking– Baseline: 1998-99 & 4 –year follow upBaseline: 1998-99 & 4 –year follow up
In home clinical evaluations and interviewIn home clinical evaluations and interview– Cultural orientation assessed by the Cuellar Cultural orientation assessed by the Cuellar
scalescale • language, contact with own ethnic group vs. language, contact with own ethnic group vs.
Anglo cultural orientation Anglo cultural orientation Protective of Cognitive DeclineProtective of Cognitive Decline
Incidence of Alzheimer’s Disease was 15.4% Incidence of Alzheimer’s Disease was 15.4% in Mexican born and 12.4% in US bornin Mexican born and 12.4% in US born
• Hazard of cognitive decline per point on Hazard of cognitive decline per point on cultural scale HR=0.98 95% CI (0.96-0.99) cultural scale HR=0.98 95% CI (0.96-0.99)
• 1 point increase means higher Anglo 1 point increase means higher Anglo cultural orientationcultural orientation
• Adjusted for age and gender, baseline Adjusted for age and gender, baseline diabetes and strokediabetes and stroke
Risk of dementia associated with combined Risk of dementia associated with combined income and education in study participantsincome and education in study participants
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Income <$1,000,Education <=8 years
Income <$1,000,Education >8 years
Income =>$1,000,Education <=8 years
Income =>$1,000,Education >8 years
Socioeconomic Status
HR
fo
r d
em
en
tia
/CIN
D
Adjusted for age, type 2 diabetes, stroke, gender, cultural orientationAdjusted for age, type 2 diabetes, stroke, gender, cultural orientation
Cancer
Cancer Incidence by Site and Ethnicity in Men, U.S. 2000
Language and Language and Cultural FactorsCultural Factors
What happens under the microscope
Language Concordance Matters
• Monolingual Spanish speaking patients with Spanish speaking physicians should do better
• Feel better with less pain, better health outlook, less symptoms
• Understand more of the physician instructions
• Better medication adherence?• Ask more questions–patient centered
How Much Fluency is Enough?How Much Fluency is Enough?
• Language fluency is a gradient
• Consider dialects
• Patients may avoid topics knowing limitations
• Evaluating cognition, mental health
• Literacy, understanding, jargon
Use of InterpretersUse of Interpreters
• Patients using interpreters ask less, say less, answer less
• Triangular interaction–space• Who translates matters: –
professional interpreter, trusted friend or family member
• Accuracy may be an issue
Translation of SurveysTranslation of Surveys• Standard technique: Forward/Backward• Forward from English to target language• Take new translation and back translate to
English• Different translators for each phase• Reconcile differences on review by fluent
speakers including both translators and investigators
• Colloquial use OK but has to be “correct”• Time consuming and Expensive
Health Related Quality of Life by Health Related Quality of Life by Ethnicity - Los Angeles 1999Ethnicity - Los Angeles 1999
Poor and Unhealthy Activity N Fair Health Days Limitation D
White 3376 13.1% 7.1 2.7
Latino 3267 35.6% 6.3 2.4
AA 835 21.2% 8.3 3.5
API 716 15.3% 4.7 1.7MMWR 2001; 50:556-9
How is your Health?How is your Health?
• Excellent
• Very Good
• Good
• Fair
• Poor
• Excelente
• Muy buena
• Buena
• Regular o Justa o Mas o Menos
• Mala (Pobre)
Ethnicity and Attitudes toward Ethnicity and Attitudes toward Patient Autonomy by EthnicityPatient Autonomy by Ethnicity
Tell Dx%
Tellprognosis
%
Life support%
European Am 87 69 65
Mexican Am 65 48 41
African Am 89 63 60
Korean Am 47 35 28
JAMA 1995; 274:820
Factors to Consider in Factors to Consider in Treatment of LatinosTreatment of Latinos
• FamilismoFamilismo - Helps Adherence? - Helps Adherence?• SimpatíaSimpatía Cultural Script Cultural Script for positive for positive