1. The Latino Mortality Paradox Revisited: Is Acculturation Bad for Your Health? Ana F. Abraído-Lanza, Ph.D. Columbia University Mailman School of Public Health

2. Purpose The Latino Mortality (Epidemiologic) Paradox The Latino Mortality (Epidemiologic) Paradox  Brief overview: Sociodemographic characteristics of Latinos in the U.S. Hypotheses regarding the paradox Hypotheses regarding the paradox Health behaviors and acculturation Health behaviors and acculturation  Key issues and applications Directions for future research Directions for future research

3. Heterogeneity of Latino Groups Source: Current Population Survey, March 2000

4. Age Differences (median) Mexican American 24.4 Puerto Rican 28.0 Cuban 41.4 Central and South American 29.9 Other Latino 29.1 Non-Latino white 37.9 Source: US Bureau of the Census, CPS March, 2000

5. Nativity Status: Latinos (in percent) Source: US Bureau of the Census, Current Population Reports, 1998 *Figures for “All Latino” exclude Puerto Ricans

6. Median Income (in thousands, 1999) Source: US Bureau of the Census, CPS March, 2000

7. Percent of Families Living Below Poverty Level Source: US Bureau of the Census, CPS March, 2000

8. Education: High School or More (Percent of persons age > 25 years) Source: US Bureau of the Census, CPS March, 2000

9. Occupations (Civilians age > 16 years) Source: US Bureau of the Census, CPS March, 2000

10. Access to Health Care: Health Insurance Latinos are more likely than whites to lack health insurance Latinos are more likely than whites to lack health insurance Gender and age Gender and age  Latino males  young adults, children Health coverage varies by Latino group Health coverage varies by Latino group

11. Geographic Distribution > 20 15-20 5-10 % of State Population Latino < 5 TX NM CA CO NV NY NJ CT FL AZ WY ID IL 10-15 OR WA UT KS

12. Reasons for sociodemographic differences Political factors Political factors Immigration Immigration  Reasons for immigrating  Patterns  History Economic factors Economic factors Social factors Social factors

13. Latino Mortality: An Epidemiologic Paradox Inverse relationship between socioeconomic status and morbidity and mortality Inverse relationship between socioeconomic status and morbidity and mortality Relative to non-Latino whites, Latinos in the United States have a worse socioeconomic status profile Relative to non-Latino whites, Latinos in the United States have a worse socioeconomic status profile Yet, Latinos have a lower all-cause mortality rate than do non-Latino whites Yet, Latinos have a lower all-cause mortality rate than do non-Latino whites

14. Hypotheses for the Paradox The lower mortality is The lower mortality is  Genuine  Health behaviors  Social networks  Other protective factors  Artifact of migratory processes  The “salmon hypothesis”  The healthy migrant effect

15. Tests of the hypotheses “Salmon bias” hypothesis “Salmon bias” hypothesis  Plausibility of hypothesis: different groups  Cubans vs. Puerto Ricans  US-born Latinos: ties to the United States Healthy migrant hypothesis Healthy migrant hypothesis  Selection factors: not relevant to US-born

16. Mortality: Latinos compared with Non-Latino whites (Cox Prop. Hazards) MenWomen Mexican American.57.60 Puerto Rican.63.45 Cuban.53.47 Central/South and Other Latino.61.56 Other Latino.61.56 All Latinos/Latinas.59.56 Source: Abraído-Lanza, Dohrenwend, Ng-Mak & Turner (1999)

17. Mortality: Analyses of US-born and Foreign-born (Cox Prop. Hazards) MenUS-born: Latinos vs. whites Foreign-born: 25-44 years 25-44 years.59 (.44,.80).80 (.44, 1.45) 45-64 45-64.60 (.51, 71).50 (.32,.78) > 65 > 65.62 (.53,.72).62 (.48,.80) Women 25-44 25-44.49 (.33,.73).66 (.32, 1.39) 45-64 45-64.65 (.52,.79).52 (.30,.91) > 65 > 65.59 (.49,.71).57 (.42,.73)

18. Conclusion The salmon and healthy migrant hypotheses The salmon and healthy migrant hypotheses  Inadequate explanations Other factors account for the paradox Other factors account for the paradox

19. Generality of Mortality Paradox Mortality from different causes Mortality from different causes  Relative to whites, Latinos have higher rates of death from various diseases, including:  Diabetes  Certain types of cancer (e.g., cervical) Age and gender differences Age and gender differences Varies by cause of death and Latino group Varies by cause of death and Latino group  Diabetes complications  Mexicans vs. Cubans

20. Other hypotheses Health behaviors and acculturation: Health behaviors and acculturation:  Latinos have more favorable health behaviors and risk factor profiles than do non-Latino whites  Health behaviors and risk factors become more unfavorable with greater acculturation

21. Do Latinos Have More Favorable Health Behaviors Compared With Non-Latino Whites? Controlling for age and SES, Latinos compared with whites are: Controlling for age and SES, Latinos compared with whites are:  Less likely to  Smoke  Drink alcohol  Engage in leisure-time exercise activities  More likely to have a high Body Mass Index Health behaviors vary by gender Health behaviors vary by gender Mixed evidence: health behaviors hypothesis Mixed evidence: health behaviors hypothesis

22. The Health Behavior and Acculturation Hypothesis Assumption: Health Behavior Hypothesis Assumption: Health Behavior Hypothesis  Health behaviors reflect Latino cultural norms Among Latinos, do health behaviors worsen (become more risky) with greater acculturation? Among Latinos, do health behaviors worsen (become more risky) with greater acculturation?

23. Acculturation Acculturation Acculturation  Process by which immigrants adopt the attitudes, values, customs, beliefs and behaviors of a new culture Assumptions of most models Assumptions of most models  Linear, directional: loss of original culture with greater acculturation  Conflict, anxiety  Original cultural devalued Sources: Clark & Hofsess (1998), LaFromboise et al. (1993), Chun et al. (2003)

24. Alternation/Bicultural Models Possible to belong to two cultures Possible to belong to two cultures Alternate between cultures Alternate between cultures Assumptions Assumptions  Bidirectional, orthogonal relationships  Both cultures positive, valuable  Not necessarily equally valued

25. Acculturated Unacculturated High Low HighLow Native Culture New Culture Linear vs. Bicultural models BiculturalAcculturated Unacculturated Marginal

26. The Problem of “Acculturation”: Theory and Measurement Language (use, comfort with): speak, read, think Language (use, comfort with): speak, read, think Length (proportion) of time in the U.S. Length (proportion) of time in the U.S. Generation status, parents’ place of birth Generation status, parents’ place of birth Ethnicity of past, current friends Ethnicity of past, current friends Ethnic composition of past, current neighborhood Ethnic composition of past, current neighborhood Behaviors, preferences Behaviors, preferences  Music, radio, television, movies, books/newspapers, celebrating occasions, diet/food Unidimensional vs. multi-components measures Unidimensional vs. multi-components measures

27. Unidimensional Measure: Language Use (Marín et al., 1987) In general, what language(s) do you In general, what language(s) do you  Read and speak  Speak at home  Think  Speak with friends 1 = Only Spanish 1 = Only Spanish 2 = Spanish better than English 2 = Spanish better than English 3 = Both equally 3 = Both equally 4 = English better than Spanish 4 = English better than Spanish 5 = Only English 5 = Only English

28. Multi-component Measure Cuellar et al. (1980): 20 items Language Language Self-identification Self-identification Parents identification Parents identification Ethnic origin of childhood and teen friends Ethnic origin of childhood and teen friends Current friends Current friends Music, TV, movie, food preferences Music, TV, movie, food preferences Generation status Generation status Country in which raised Country in which raised Contact with Mexico Contact with Mexico Ethnic pride Ethnic pride

29. Health Behavior/Acculturation Hypothesis: Is Acculturation Bad for Your Health? Greater acculturation is associated with: Greater acculturation is associated with:  Increased rates of infant mortality, low birth weight  Overall cancer rates  High blood pressure  Adolescent pregnancy  Smoking, alcohol consumption, illicit drug use  Decreased fiber consumption  Depressive symptoms?

30. State of the Literature: Health Behavior and Acculturation Hypotheses Inconsistent support Inconsistent support After controlling for SES and other demographic variables, acculturation among Latinos is associated with increases in some risky health behaviors and risk factors, such as: After controlling for SES and other demographic variables, acculturation among Latinos is associated with increases in some risky health behaviors and risk factors, such as:  Smoking  Alcohol intake  Body Mass Index Acculturation, however, is also associated with increased exercise and lower fat consumption, which are health protective behaviors Acculturation, however, is also associated with increased exercise and lower fat consumption, which are health protective behaviors Effects of acculturation on health behaviors differ by gender Effects of acculturation on health behaviors differ by gender

31. Issues to Consider Is acculturation a proxy for other variables? Is acculturation a proxy for other variables?  Exposure to adverse circumstances, stressful events Adjustment for confounding variables (e.g., SES) Adjustment for confounding variables (e.g., SES) Age differences Age differences Different indicators of acculturation Different indicators of acculturation Acculturation as a risk or protective factor? Acculturation as a risk or protective factor?  Health-related and other beliefs, values, and behaviors  Effect may vary depending on outcome variable

32. Issues to Consider (cont’d) Do acculturation processes differ across groups (e.g., Cubans vs. Puerto Ricans)? Do acculturation processes differ across groups (e.g., Cubans vs. Puerto Ricans)? Acculturating to what? Assumption: majority (white) American culture Acculturating to what? Assumption: majority (white) American culture What is the role of socioeconomic status in acculturation? What is the role of socioeconomic status in acculturation? What is the impact (if any) of historical and political factors on acculturation processes? What is the impact (if any) of historical and political factors on acculturation processes?

33. Acculturation and Health Paucity of theoretical models Paucity of theoretical models Account for SES confounders Account for SES confounders Mediating mechanisms Mediating mechanisms  Beliefs, values, behaviors  Structural constraints (e.g., access to health care) Comprehensive measures of acculturation Comprehensive measures of acculturation

34. On Paradigms and Paradoxes

35. Paradox and Health Disparities: Complex Issues and Paradigms Social/structural factors Social/structural factors  Socioeconomic factors (SES), social capital  Health care (e.g., access, quality)  Racism Individual-level factors Individual-level factors  Coping, mobilization of support systems Cultural factors Cultural factors  Attitudes, values, customs, beliefs, behaviors  Level of acculturation

36. Latino Culture: Values, Customs, Beliefs Respeto Respeto Cortesía Cortesía Personalismo, Simpatía Personalismo, Simpatía Personas de confianza Personas de confianza Familism Familism Religiosity Religiosity Fatalism vs. destino Fatalism vs. destino

37. Application: Breast Cancer Mortality*5-yr survival Latina 1576% Non-Latina white women 27.785% *Per 100,000 population Hypotheses Health-related behaviors, major focus: screening Health-related behaviors, major focus: screening Cultural vs. Structural factors Cultural vs. Structural factors Acculturation (beliefs?, behaviors?) Acculturation (beliefs?, behaviors?) Other factors Other factors

38. Example: Theoretical Model Language barriers Health beliefs, values, knowledge Behavior Diet, parity, smoking, risky behaviors Acculturation SES Income, Education, Occupation Breast Cancer (Screening, Mortality, Survival) Health Care - Access - Quality PredictorsMediators Sociocultural Stressors (racism, etc) Age Source: Adapted from Abraído-Lanza, Chao & Flórez (under review)

39. State of the Literature: Acculturation and Breast Cancer Screening Level of acculturation vs. structural factors Level of acculturation vs. structural factors Access to and quality of health care Access to and quality of health care Role of acculturation is not clear Role of acculturation is not clear  Conflicting results  Measures differentially predict outcomes Lack of theoretical models (e.g., beliefs about survival) Lack of theoretical models (e.g., beliefs about survival)

40. Future Research: The Paradox Generality of the paradox Generality of the paradox Acculturation: risk and protective factors Acculturation: risk and protective factors  Role of ethnic enclaves  Social networks, social capital, access to cultural and other resources  Values and Behaviors: diet, parity, other factors  Health promoting (e.g., exercise) vs. disease detection behaviors (e.g., cancer screening)  Beliefs, values (e.g., “fatalism”??) Theoretical models Theoretical models  Resource vs. Deficit models