Objectives. We investigated differences in the development of disability in activities of daily living among non-Hispanic Whites, African Americans, Hispanics interviewed in Spanish, and Hispanics interviewed in English.

Methods. We estimated 6-year risk for disability development among 8161 participants 65 years or older and free of baseline disability. We evaluated mediating factors amenable to clinical and public health intervention on racial/ethnic difference.

Results. The risk for developing disability among Hispanics interviewed in English was similar to that among Whites (hazard ratio [HR]=0.99; 95% confidence interval [CI] = 0.6, 1.4) but was substantially higher among African Americans (HR=1.6; 95% CI=1.3, 1.9) and Hispanics interviewed in Spanish (HR=1.8; 95% CI=1.4, 2.1). Adjustment for demographics, health, and socioeconomic status reduced a large portion of those disparities (African American adjusted HR=1.1, Spanish-interviewed Hispanic adjusted HR=1.2).

Conclusions. Higher risks for developing disability among older African Americans, and Hispanics interviewed in Spanish compared with Whites were largely attenuated by health and socioeconomic differences. Language- and culture-specific programs to increase physical activity and promote weight maintenance may reduce rates of disability in activities of daily living and reduce racial/ethnic disparities in disability.

Disability among older adults (those 65 years or older) is a major health issue involving high personal and economic costs. The number of Americans 65 years or older with chronic disability exceeds 7 million.1 Maintaining the quality of life for older adults by delaying disability may be as important as prolonging life.2,3 Disability is more strongly associated with medical spending than with life expectancy.4,5 Long-term care expenditures for older people are projected to reach $161 billion per year by 2010, of which two thirds will be paid by government programs.6

The composition of the US population is changing. In 2000, 18% of people in the United States spoke a language at home other than English, up from 11% in 1980.7 The fastest-growing part of the older US population comprises minority groups, particularly African Americans and Hispanics.8 As the number of older people belonging to minority groups increases, there are growing public health concerns about racial/ethnic disparities in health outcomes.9 Although overall rates of disability among older Americans have declined over time, racial/ethnic disparities persist.1,1013 The literature on racial/ethnic disparities in disability mostly focuses on African Americans; national studies investigating disability among Hispanics are limited.1,12,14,15

Despite the common practice of conducting interviews in languages other than English to allow respondents to participate in their primary language, few national studies have considered the influence of language differences on health outcomes.16 There are known differences in health and mortality related to immigration and acculturation.1619 Use of an interview language other than that of the host culture may be a proxy for acculturation and a predictor of future poor health.2023 Insight from a broader investigation of risk factors that includes language differences in relation to disparities in the development of disability is essential to the development of population-based public health programs to help maintain independence among older adults.

We investigated racial/ethnic differences in disability among people 65 years and older using 6 years of data from the Health and Retirement Study (HRS),24 Finally, we investigated whether factors amenable to public health and policy intervention mediate minority differences in the development of disability among these Medicare-aged adults.

Study Population

A prospective cohort from a national sample of community-dwelling US adults was interviewed in 1998 (baseline) and biennially through 2004 as part of the HRS, which is sponsored by the National Institute on Aging and conducted by the University of Michigan.25 The HRS interview was translated into Spanish and back-translated into English question by question for meaning by multiple bilingual translators using established methods.26 To address dialect variations, experienced translators living in different parts of the country participated (written communication with HRS help site, May 3, 2000).

We based prevalence estimates of disability on a cohort of 10524 persons 65 years or older who participated in the 1998 HRS baseline interview and self-identified as Hispanic/Latino, Black/African American, or White. We limited analyses evaluating the development of disability to a cohort of 8161 people free of baseline disability who were alive at the subsequent scheduled interview (in 2000); excluded by design were 1304 people who reported baseline disability and 578 who died prior to 2000. For analytic purposes, we also omitted from incident disability analyses 426 HRS 2000 nonrespondents, 48 respondents with insufficient baseline information, and 7 respondents with missing follow-up disability information.

Outcome Variable

Disability in activities of daily living is consistent with the Institute of Medicine’s definition of disability, which is defined as the inability to carry out self-care tasks at the personal level.27 The presence or absence of disability in activities of daily living is a standard measure of disability in national studies.28,29 Basic activities of daily living monitored by the HRS include dressing, toileting, bathing, eating, walking across a room, and getting in and out of bed. Disability in an activity of daily living expected to last 3 months or more was determined from participants’ responses to questions regarding their inability to perform a certain task, avoidance of it, or the need for help or equipment in its performance. This assessment captures the chronic dependence in basic self-care tasks that could jeopardize a person’s ability to live independently. For the purpose of analysis, the development of disability in a particular activity of daily living was identified by the first report of that disability at a follow-up interview (2000, 2002, or 2004).

Explanatory Variables

Baseline (1998) demographic information included race/ethnicity, marital status, age, gender, and living situation (living alone or with others). Self-reported information on race/ethnicity was used to classify people into mutually exclusive groups: non-Hispanic African American, Hispanic, and non-Hispanic White. Hispanics were further divided through use of the baseline interview language as a proxy for acculturation or assimilation into Hispanics interviewed in Spanish or Hispanics interviewed in English.16,20 People from other racial/ethnic groups (n = 194) were excluded from analyses because of small numbers. Marital status and living situation were assessed at all biennial interviews (1998, 2000, 2002, and 2004). For purposes of analysis, age was categorized as 65–74, 75–84, and ≥ 85 years.

At all interviews, health factors were assessed from self-reported information on chronic conditions, functional limitations, and health behaviors. Chronic conditions were ascertained from a self-report of a physician diagnosis of conditions that included arthritis or rheumatism (hereafter called arthritis), cancer, cardiovascular disease (i.e., hypertension, heart attack, coronary artery disease, congestive heart failure, angina, other heart disease), diabetes, pulmonary disease (i.e., chronic bronchitis, emphysema), psychiatric problems (i.e., emotional, nervous, or psychiatric problems), and stroke. Bad vision was from a self-report of poor eyesight or being legally blind.

Both physical limitations and those involving instrumental activities of daily living were assessed at each interview. Physical limitations were determined by self-reports of difficulty in performing, inability to perform, or avoidance of 4 physical tasks: walking several blocks, climbing several flights of stairs without resting, pulling or pushing large objects, and lifting or carrying weights over 10 pounds. Limitations in instrumental activities of daily living were determined by self-reports of inability to perform, avoidance of, or receiving help with tasks such as preparing hot meals, shopping, using the telephone, taking medication, and managing money.

Health behavior information assessed at each interview included current smoking status, current alcohol consumption (such as beer, wine, or liquor), weight, and lack of regular vigorous physical activity. Substantial change in weight (more than 10 pounds in last 2 years) was noted; obesity and underweight were defined as a having a body mass index (weight in kilograms divided by height in meters squared) of 30 kg/m2 or more and less than 20 kg/m2, respectively, calculated from self-reported height and weight. Regular vigorous physical activity was defined as participation 3 times or more a week over the last 12 months in activities such as sports, heavy housework, or a job that involved physical labor.

Socioeconomic factors assessed at each interview included education, family wealth, household income, and health insurance. Education was dichotomized as higher (i.e., more than 12 completed years of schooling) and lower (i.e., 12 or fewer years of schooling) education. We used imputed estimates of family wealth and household income (i.e., all sources received by the respondent and spouse or partner during the preceding year) developed at the University of Michigan when only partial information was provided.30 For analytic purposes, income and wealth were dichotomized by use of the lowest HRS population-weighted quartiles at each interview. Health insurance was categorized into Medicare only, any private coverage, Medicaid or other government insurance, and no coverage or missing.

Statistical Analysis

The HRS is a national probability sample. We weighted all analyses and adjusted for the complex HRS sampling design through use of person-weights, strata, and sampling error codes for the 1998 HRS data, which were developed at the University of Michigan31 to provide valid inferences to the US population. For the analyses, we used SUDAAN software (Research Triangle Institute, Research Triangle Park, NC) except where noted. Statistical testing was done at a nominal significance level of α = .05. We used the χ2 test in baseline comparisons between the demographic, health, and economic characteristics of minority subgroups versus Whites. We compared baseline disability prevalences of minority subgroups versus Whites using associated 95% confidence intervals (CIs) based on the standard error of the difference.

We used survival analysis methods for discrete data to determine the effect of risk factors on developing disability; these data were restricted to persons without baseline disability. Discrete data methods modeled the development of disability ascertained at biennial interviews (i.e., in discrete time). Conceptually, a discrete hazard model is analogous to a Cox proportional hazard model for continuous data.32 Discrete hazard models account for repeated measures, use time-varying covariates, and do not require a proportional hazard assumption. The model estimated the probability of developing disability in the subsequent 2 years given an event-free status (i.e., no disability) and a person’s risk factor profile. If a person’s disability status was known at the beginning and end of an interval (e.g., 1998–2000, 2000–2002, 2002–2004), this record was included in the analysis. Time differences in the racial/ethnic hazard ratios were tested by interaction terms. We estimated the discrete hazard model with SAS Proc GenMod (SAS Institute Inc, Cary, NC) to fit a generalized linear model with a complementary log–log link. We estimated variance using balanced repeated replication, a form of bootstrapping.33,34 Results are reported as hazard ratios; an associated 95% CI that excludes the number 1 indicates a significant predictor for developing disability.

We restricted analyses to 1998 HRS respondents. Compared with respondents, non-respondents (4.59%) tended to be disproportionately African American or Hispanic. Using standard sampling methodology, we adjusted for potential bias related to missing interview information and nonresponse by treating respondents with complete data as an additional sampling stage, thereby obtaining adjusted sampling weights.33

The 10 524 members of the 1998 HRS cohort 65 or older were 12.48% African American, 3.99% Hispanic interviewed in Spanish, and 2.74% Hispanic interviewed in English. Compared with Whites, the prevalence of disability in activities of daily living in this cohort was significantly greater among African Americans (prevalence = 18.01%; difference from prevalence among Whites = 7.36 percentage points; 95% CI = 4.66%, 10.06%) and Hispanics interviewed in Spanish (prevalence = 23.40%; difference = 12.75 percentage points; 95% CI = 8.40%, 17.10%). Prevalence of disability was similar for Hispanics interviewed in English (prevalence = 10.78%; difference = 0.13 percentage points; 95% CI = 3.38%, 3.63%) and Whites (prevalence = 10.65%).

Six-year population mortality was greatest among African Americans (31.14%), whereas Hispanics interviewed in Spanish (24.69%), Hispanics interviewed in English (22.96%), and Whites (26.70%) had similar rates. Among survivors, rates of follow-up nonresponse were 4.97%, 7.08%, and 8.78%, respectively, for the 2000, 2002, and 2004 follow-up interviews. All but 2.36% of African Americans, 1.90% of Hispanics interviewed in Spanish, 3.13% of Hispanics interviewed in English, and 1.96% of Whites participated in 1 or more follow-up interviews.

The development of disability over 6 years was assessed for 8161 people who were free of disability at baseline (903 African Americans, 292 Hispanics interviewed in Spanish, 216 Hispanics interviewed in English, and 6750 Whites; Table 1). This population of Medicare-aged persons represented 2.0 million African Americans and 1.3 million Hispanics at risk of developing disability. This cohort without baseline disability was primarily female (57.70%) and included African Americans (7.5%), Hispanics interviewed in Spanish (2.54%), Hispanics interviewed in English (2.14%), and Whites (87.81%) when weighted to the US population. More than 83.71% of this cohort reported 1 or more co-morbid health conditions.

Over the subsequent 6 years, disability developed in 21.21% of those who were free of disability at the baseline interview. The development of disability was significantly more frequent among African Americans (30.41%) and Hispanics interviewed in Spanish (32.67%) than among Whites (20.13%), whereas Hispanics interviewed in English (19.98%) had disability rates comparable to those of Whites. Table 2 shows disability rates over time by racial/ethnic groups. A test for differences over time in the racial/ethnic hazard ratios (HRs) showed no significant trends, indicating that the observed hazard rates were constant over time. Overall, development of disability was 61% higher among African Americans (HR = 1.61) and 76% higher among Hispanics interviewed in Spanish (HR = 1.76) compared with Whites. The disability rate for Hispanics interviewed in English was similar to that of Whites (HR = 0.99).

Table 3 shows HRs for developing disability over 6 years for minority groups versus Whites (reference group), initially adjusted for demographic characteristics and then further adjusted for health and socioeconomic factors potentially amenable to public health or public policy interventions. The disability hazard rates adjusted for demographics were significantly higher among African Americans (adjusted HR = 1.57) and Hispanics interviewed in Spanish (adjusted HR=1.70) than among Whites, but they were not significantly elevated for Hispanics interviewed in English (adjusted HR = 1.03). Further adjustment for health factors related to chronic conditions, functional limitations, and health behaviors reduced the excess hazard 68% among African Americans (to 1.18). Notably, for both African Americans and Hispanics interviewed in English, health behaviors alone explained 28% to 51% of the excess hazard, whereas differences in chronic diseases had little impact on excess disability rates.

As expected, functional limitations alone, which may be viewed as precursors of disability, also explained a substantial portion of excess risk among African Americans and Hispanics interviewed in Spanish. Analyses in which we controlled for socioeconomic factors further showed reduced excess risk among African Americans by another 7% (to 1.14) and among Hispanics interviewed in Spanish by another 17% (to 1.20), even after differences because of health and demographic factors were accounted for. Recognizing that the impact of socioeconomic factors may overlap with that of health factors, we performed further analyses in which we controlled for socioeconomic factors and demographics (without the influence of health factors); these showed that socioeconomic factors alone explained 65% of excess risk for African Americans and 100% for Hispanics interviewed in Spanish. This finding reflects the presence of higher economic disparities among minorities compared with Whites, as indicated in Table 1. However, economic disparities and types of insurance held are likely to partially reflect the health of individuals.

The relative impact of demographic, health, and socioeconomic factors on the risk for developing disability in the hazard model that included all predictor variables (data not shown) indicated that the only significant demographic factor was older age, which approximately doubled the hazard rate with each additional decade of life (for age 74–84 years, HR = 1.92; 85 years or older, HR = 3.69 relative to reference age [65–74]). Chronic conditions that significantly increased the risk of developing disability were a history of stroke (HR = 1.71), diabetes (HR = 1.31), and arthritis (HR = 1.22). Functional limitations in instrumental activities of daily living (HR = 3.21) and physical limitations (HR = 1.45) were strong predictors of developing disability. Significant health behaviors in the full model included weight loss (HR = 2.17), lack of regular vigorous physical activity (HR = 1.90), weight gain (HR = 1.51), and being underweight (HR = 1.46); the use of alcohol was protective (HR = 0.68). Socioeconomic factors were not significant predictors of developing disability after we controlled for other risk factors in this Medicare-aged population, except for holding private health insurance, which was associated with significantly lower risk for developing disability (HR = 0.85).

This national study provides evidence of racial/ethnic differences in the prevalence and development of disability among Medicare-aged US adults on the basis of 1998–2004 HRS data, but the relationships are complex. The prevalence of disability in activities of daily living was substantially greater among African Americans (18.01%) and Hispanics interviewed in Spanish (23.40%) than among Whites (10.65%). Prevalence of disability among Hispanics who preferred an English interview (10.78%) was similar to that among Whites. Among persons free of disability, almost one third of older African Americans (30.41%) and Hispanics interviewed in Spanish (32.67%) developed disability over 6 years, with lower rates among Hispanics interviewed in English (19.98%) and Whites (20.13%).

These findings show that African American versus White differences in prevalence of disability and the development of disability persist into the new century, consonant with earlier national data.11,12,35,36 Importantly, our study adds insight into the Hispanic experience of disability. Previous cross-sectional national studies reported differences between Hispanics and Whites in prevalence of disability.37,38 We found that older Hispanics interviewed in Spanish had higher prevalence of disability and were at greater risk to develop disability than were Whites. By contrast, Hispanics interviewed in English had rates of disability similar to those of Whites, as well as similar risk for developing disability. There were striking differences between Hispanics interviewed in Spanish and those interviewed in English in both their rates of disability and predisposing factors, indicating that these 2 Hispanic groups need to be considered separately. Compared with Hispanics interviewed in English, Hispanics interviewed in Spanish were 50% more likely to report disability and functional limitations; had substantially fewer assets in terms of education, income, and wealth; and were more likely to depend on Medicaid.

To guide a public health response to promote equity in health outcomes, we investigated the extent to which differences in the development of disability were mediated by demographic, health, and socioeconomic factors. In the cohort free of disability at baseline, Whites were disproportionately older (aged > 75 years) and disability developed more frequently among African Americans and Hispanics interviewed in Spanish. It is therefore not surprising that demographic differences did little to attenuate differences in the development of disability.

Two factors, however, emerged as strong mediators for disability. The first factor is related to differences in health. Our analyses (Table 3) indicated that among African Americans and Hispanics interviewed in Spanish, health factors related to chronic conditions, functional limitations, and health behaviors all contributed to disparities in the development of disability. Notably, health behaviors had a stronger influence on excess disability in these groups than on differences in chronic disease. Specifically, increasing physical activity and maintaining weight (i.e., avoiding either weight gain or weight loss) were important for reducing disability and promoting health equity in older adults. This finding is relevant to potential public health action because health behaviors are amenable to intervention.

The second factor is socioeconomic status. Limited resources in terms of education and finances, plus a large dependence on Medicaid coverage, contributed to disability disparities for both African Americans and Hispanics interviewed in Spanish. However, the only significant socioeconomic factor for developing disability after demographic and health differences were accounted for was having private health insurance. Taken together, these findings may indicate that some minorities not only cannot afford private insurance but cannot afford, or do not access, medical care made available through Medicare or Medicaid.

In addition, private insurance may be an indicator of the quality of health care received. Individuals in lower-tier health plans commonly have fewer choices with regard to health services, which can compromise their quality of care.3941 Compounding these factors for Hispanics interviewed in Spanish are language barriers.4245 While reasons for these system problems are complex,46 older members of minority groups with limited economic resources may have less effective interface with the health care system than Whites, which is manifest in the disparate proportions of disability attributable to socioeconomic factors.

Notably, the risk of developing disability among Hispanics interviewed in English was similar to that among Whites. We used preferred language of interview as a measure of acculturation.20 Although some cultural features are not captured by this simple measure, it differentiates 2 distinct populations.20,47 Hispanics interviewed in Spanish have low socioeconomic status in terms of education and assets, consistent with lower acculturation.48 Language barriers may limit opportunities for integration with another culture and reduce social acceptance.20 In our study, the fact that, after we controlled for other confounding risk factors, the risk of developing disability among Hispanics interviewed in Spanish was higher than among Whites, but the risk was similar for Hispanics interviewed in English and Whites, may indicate the influence of other unmeasured environmental factors contributing to disability among the Hispanic population interviewed in Spanish. Such factors may include poorer living conditions and segregation.21 Life course disadvantages stemming from limited educational and occupational choices and social stress related to poverty may contribute to the higher rates of disability in the Hispanic group interviewed in Spanish.22,4952

According to a “Hispanic paradox” reported in the literature, the US Hispanic population has lower mortality than do Whites despite wide differences in socioeconomic status.5355 One possible explanation for this paradox is that older Mexican Americans may return to their country of origin when they are old and disabled.19 If this explanation holds, then disability rates among Hispanics interviewed in Spanish are higher than we detected and the disparities are underestimated.

Limitations

Several limitations common to secondary databases may affect our findings. This study included self-reported information from consenting adults, which would have likely excluded persons with dementing illness and other cognitive impairments. However, this omission is unlikely to have substantially influenced our results, because such people would have frequently had prevalent (baseline) disability, making them ineligible for incident disability cohort analyses. Our study lacked measures of disease severity. However, self-reported physical limitations and limitations in instrumental activities of daily living, the likely consequences of disease, provide surrogates for measures of disease severity. Our methodology of dividing Hispanics into 2 groups based on interview language is a gross measure of acculturation or assimilation, but lack of proficiency in the host language alone explains over 70% of variance in acculturation scores.56 Because such lack of proficiency is often linked to low education, Hispanics interviewed in Spanish may have difficulty understanding questions even when administered in Spanish. There are probably also important cultural differences aggregated into African American and White subgroups. Therefore, in this study, analyses were adjusted for covariates of race/ethnicity (e.g., education, income, wealth).

Conclusions

Although rates of disability in the United States have been decreasing over the past 2 decades, they remain high among older adults, whose health care is largely covered by public insurance through Medicare. These national data document higher rates of disability among older African Americans and Hispanics interviewed in Spanish compared with Whites, but Hispanics interviewed in English had risks of disability similar to those of Whites. Differences in socioeconomic status between groups were great. The finding that holding private insurance as a supplement to Medicare explained reduced disability may suggest that the lack of private insurance in the years prior to Medicare eligibility is especially problematic given the lack of wealth, income resources, and public insurance available to minority populations prior to age 65. At the public health level, prevention or intervention programs, specific to language and cultural groups, designed to increase physical activity and promote weight maintenance may prove to be efficient strategies not only for reducing rates of disability in activities of daily living but also for lowering racial/ethnic disparities in disability.

Table
TABLE 1— Baseline Characteristics of People 65 Years or Older Without Disability at Baseline (n = 8161), by Race/Ethnicity: The Health and Retirement Study, 1998
TABLE 1— Baseline Characteristics of People 65 Years or Older Without Disability at Baseline (n = 8161), by Race/Ethnicity: The Health and Retirement Study, 1998
DemographicsAfrican Americans (n = 903), %Hispanics Interviewed in Spanish (n = 292), %Hispanics Interviewed in English (n = 216), %Whites (n = 6750), %
Women62.17*55.5961.6357.29
Age, y
    65–7465.12a**60.6669.94a*56.91
    75–8427.3630.8124.4336.29
    ≥ 857.518.535.636.80
Unmarried63.08**44.8151.18*40.04
Living alone36.03*20.47**30.5130.85
Chronic conditions
    Arthritis60.19**45.34*53.2253.19
    Cancer10.01**5.67**7.26**13.33
    Diabetes19.87**18.4719.27*11.61
    Heart disease20.2411.81**16.39*22.15
    Hypertension61.62**45.0545.0546.48
    Psychiatric problems6.5710.347.018.09
    Pulmonary disease5.10**3.52**2.82**8.22
    Stroke6.621.79**4.675.60
    Poor vision/legally blind8.85**10.28**8.34*4.35
Functional limitations
    Physical function38.16**43.76*31.0326.36
    IADL disability10.67**15.13*8.716.26
Health behaviors
    Current smoker14.16*13.0610.0310.55
    Current alcohol use15.03**15.52**28.0432.29
    Regular vigorous physical activity33.27**38.8736.54*45.23
Weight
    Obeseb27.32**18.7824.88*15.39
    Underweightc4.05**6.583.396.52
    Gain > 10 lbd16.20**16.97*13.9911.20
    Loss > 10 lbd21.81**15.9613.0014.87
Socioeconomic status
    Low education (<12 y)59.21**82.56**52.17**25.80
    Low incomee57.70**80.28**47.94**23.92
    Low net wealthe52.44**66.42**41.51**15.74
Health insurance
    Medicare only50.30a**41.65a**59.10a**55.39
    Any private insurance26.976.8823.6139.77
    Medicaid/CHAMPS19.6447.4015.714.28
    None or missing3.104.061.590.56

Note. IADL = instrumental activity of daily living; CHAMPS = Community Health Automated Medicaid Processing System; HRS = Health and Retirement Study. In the statistical test, minority group was compared with White reference group using the χ2 test.

aχ2 test over multiple risk factor categories.

bObesity was defined as a having a body mass index (weight in kilograms divided by height in meters squared) of 30 kg/m2 or more, calculated from self-reported height and weight.

cUnderweight was defined as a having a body mass index of less than 20 kg/m2, calculated from self-reported height and weight.

dMeasured over the past 2 years.

eRefers to that of those in the lowest HRS population-weighted quartile.

* P < .05; **P < .01 (unadjusted).

Table
TABLE 2— Cumulative Risk of Developing Disability Over 6 Years of Follow-Up Among People 65 Years or Older (n = 8161) Without Disability at Baseline: The Health and Retirement Study, 1998–2004
TABLE 2— Cumulative Risk of Developing Disability Over 6 Years of Follow-Up Among People 65 Years or Older (n = 8161) Without Disability at Baseline: The Health and Retirement Study, 1998–2004
  Risk of Disability by Years of Follow-Up, %
Racial/Ethnic GroupNo.2 Years4 Years6 Years
African Americans90311.8121.9430.41
Hispanics interviewed in Spanish29214.2624.0732.67
Hispanics interviewed in English2168.1414.8119.98
Whites67507.1414.2220.13
Table
TABLE 3— Racial/Ethnic Hazard Ratios (HRs) for Developing Disability in Activities of Daily Living Among People 65 Years or Older (n = 8161) Without Disability at Baseline: The Health and Retirement Study, 1998–2004
TABLE 3— Racial/Ethnic Hazard Ratios (HRs) for Developing Disability in Activities of Daily Living Among People 65 Years or Older (n = 8161) Without Disability at Baseline: The Health and Retirement Study, 1998–2004
 Racial/Ethnic Group, Adjusted HR (95% CI)
Adjustment FactorsaAfrican Americans (n = 903)Hispanics Interviewed in Spanish (n = 292)Hispanics Interviewed in English (n = 216)Whites (n = 6750) (Reference)
Unadjusted1.61 (1.30, 1.92)*1.76 (1.44, 2.08)*0.99 (0.63, 1.36)1.00
Demographic factors1.57 (1.27, 1.87)*1.70 (1.36, 2.04)*1.03 (0.62, 1.43)1.00
Demographics + health factors (chronic disease, functional limitations, health behaviors)1.18 (0.96, 1.40)1.32 (1.11, 1.53)*0.96 (0.60, 1.32)1.00
Demographics + chronic diseases1.51 (1.23, 1.80)*1.75 (1.38, 2.11)*0.99 (0.57, 1.40)1.00
Demographics + functional limitations1.36 (1.13, 1.59)*1.30 (1.09, 1.52)*0.99 (0.62, 1.36)1.00
Demographics + health factors1.28 (1.05, 1.52)*1.50 (1.21, 1.80)*0.95 (0.57, 1.34)1.00
Demographics + socioeconomic factors1.20 (0.97, 1.42)0.99 (0.77, 1.22)0.80 (0.45, 1.14)1.00
Demographics + health factors + socioeconomic factors1.14 (0.92, 1.36)1.20 (0.95, 1.45)0.92 (0.57, 1.28)1.00

Note. CI = confidence interval.

aAdjustment factors included demographic factors (race/ethnicity, age, gender, marital status, living arrangement), health factors (chronic diseases: arthritis, cancer, diabetes, heart disease, hypertension, psychiatric problems, pulmonary disease, stroke, vision problem; functional limitations: physical limitations, limitations in instrumental activities of daily living; health behaviors: smoking, alcohol use, exercise, weight problems), and socioeconomic factors (education, income, wealth, health insurance).

* P < .05.

This study was supported in part by funding from the National Institute for Arthritis and Musculoskeletal Diseases (grant P60-AR48098) and the National Center for Medical Rehabilitation Research (grant R01-HD45412).

We gratefully acknowledge formative comments from David Baker.

Human Participant Protection This study received exemption from human subjects review by the Northwestern University institutional review board for these analyses of public data.

References

1. Manton KG, Gu X. Changes in the prevalence of chronic disability in the United States black and non-Black population above age 65 from 1982 to 1999. Proc Natl Acad Sci U S A. 2001;98:6354–6359. Crossref, MedlineGoogle Scholar
2. Fried LP, Guralnik JM. Disability in older adults: evidence regarding significance, etiology, and risk. J Am Geriatr Soc. 1997;45(1):92–100. Crossref, MedlineGoogle Scholar
3. Guralnik JM, Fried LP, Salive ME. Disability as a public health outcome in the aging population. Annu Rev Public Health. 1996;17:25–46. Crossref, MedlineGoogle Scholar
4. Cutler DM. Disability and the future of Medicare. N Engl J Med. 2003;349(11):1084–1085. Crossref, MedlineGoogle Scholar
5. Lubitz J, Cai L, Kramarow E, Lentzner H. Health, life expectancy, and health care spending among the elderly. N Engl J Med. 2003;349(11):1048–1055. Crossref, MedlineGoogle Scholar
6. Congressional Budget Office. Projections of expenditures for long-term care services for the elderly. March 1999. Available at: http://www.cbo.gov/show-doc.cfm?index=1123&sequence=0. Accessed April 13, 2006. Google Scholar
7. US Census Bureau. Language Use and English-Speaking Ability: 2000. Washington, DC: US Department of Commerce; October 2003. Census 2000 Brief C2KBR-29. Google Scholar
8. Trends in aging—United States and worldwide. MMWR Morb Mortal Wkly Rep. 2003;52(6): 101–104, 106. MedlineGoogle Scholar
9. Institute of Medicine. Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. Washington, DC: National Academies Press; 2002. Google Scholar
10. Kelley-Moore JA, Ferraro KF. The Black/White disability gap: persistent inequality in later life? J Gerontol B Psychol Sci Soc Sci. 2004;59(1):S34–S43. Crossref, MedlineGoogle Scholar
11. Mendes de Leon CF, Barnes LL, Bienias JL, Skarupski KA, Evans DA. Racial disparities in disability: recent evidence from self-reported and performance-based disability measures in a population-based study of older adults. J Gerontol B Psychol Sci Soc Sci. 2005;60(5):S263–S271. Crossref, MedlineGoogle Scholar
12. Schoeni RF, Martin LG, Andreski PM, Freedman VA. Persistent and growing socioeconomic disparities in disability among the elderly: 1982–2002. Am J Public Health. 2005;95(11):2065–2070. LinkGoogle Scholar
13. Cutler DM. Declining disability among the elderly. Health Aff (Millwood). 2001;20:11–27. Crossref, MedlineGoogle Scholar
14. Freedman VA, Martin LG. Understanding trends in functional limitations among older Americans. Am J Public Health. 1998;88(10):1457–1462. LinkGoogle Scholar
15. Mendes de Leon CF, Beckett LA, Fillenbaum GG, et al. Black–White differences in risk of becoming disabled and recovering from disability in old age: a longitudinal analysis of two EPESE populations. Am J Epidemiol. 1997;145(6):488–497. Crossref, MedlineGoogle Scholar
16. Sudano JJ, Baker DW. Explaining US racial/ethnic disparities in health declines and mortality in late middle age: the roles of socioeconomic status, health behaviors, and health insurance. Soc Sci Med. 2006; 62(4):909–922. Crossref, MedlineGoogle Scholar
17. Abraido-Lanza AF, Chao MT, Florez KR. Do healthy behaviors decline with greater acculturation? Implications for the Latino mortality paradox. Soc Sci Med. 2005;61(6):1243–1255. Crossref, MedlineGoogle Scholar
18. Angel JL, Buckley CJ, Sakamoto A. Duration or disadvantage? Exploring nativity, ethnicity, and health in midlife. J Gerontol B Psychol Sci Soc Sci. 2001; 56(5):S275–S284. Crossref, MedlineGoogle Scholar
19. Palloni A, Arias E. Paradox lost: explaining the Hispanic adult mortality advantage. Demography. 2004;41(3):385–415. Crossref, MedlineGoogle Scholar
20. Arcia E, Skinner M, Bailey D, Correa V. Models of acculturation and health behaviors among Latino immigrants to the US. Soc Sci Med. 2001;53(1):41–53. Crossref, MedlineGoogle Scholar
21. Williams DR, Collins C. Racial residential segregation: a fundamental cause of racial disparities in health. Public Health Rep. 2001;116(5):404–416. Crossref, MedlineGoogle Scholar
22. Lynch JW, Kaplan GA, Shema SJ. Cumulative impact of sustained economic hardship on physical, cognitive, psychological, and social functioning. N Engl J Med. 1997;337(26):1889–1895. Crossref, MedlineGoogle Scholar
23. Sundquist J, Winkleby MA. Cardiovascular risk factors in Mexican American adults: a transcultural analysis of NHANES III, 1988–1994. Am J Public Health. 1999;89(5):723–730. LinkGoogle Scholar
24. Heeringa S, Connor J. Technical Description of the Health and Retirement Study Sample Design. HRS Documentation. Ann Arbor, Mich: Population Studies Center; 1995. HRS documentation report DR-002. Google Scholar
25. Heeringa S. Technical Description of the Asset and Health Dynamics Among the Oldest Old (AHEAD) Study Sample Design. Ann Arbor, Mich: Population Studies Center; 1995. HRS Documentation Report DR-003. Google Scholar
26. McKay RB, Breslow MJ, and Tarnai J. Translating Survey Questionnaires: Lessons Learned. New Directions for Program Evaluation. John Wiley & Sons, New York: 1996. Google Scholar
27. Institute of Medicine. Disability in America: Toward a National Agenda for Prevention. Washington, DC: National Academy of Sciences; 1991. Google Scholar
28. Centers for Disease Control and Prevention. State-specific prevalence of disability among adults—11 states and the District of Columbia, 1998. MMWR Morb Mortal Wkly Rep. 2000;49(31):711–714. MedlineGoogle Scholar
29. Centers for Disease Control and Prevention. Prevalence of disabilities and associated health conditions—United States, 1991–1992. MMWR Morb Mortal Wkly Rep. 1994;43(40):730–731, 737–739. MedlineGoogle Scholar
30. Smith JP. Wealth inequality among older Americans. J Gerontol B Psychol Sci Soc Sci. 1997;52:74–81. Crossref, MedlineGoogle Scholar
31. Health and Retirement Study. Sampling Weights Revised for Tracker 2.0 and Beyond. Ann Arbor: University of Michigan, Survey Research Center; 2002. Google Scholar
32. Prentice R, Gloeckler L. Regression analysis of group survival data with applications to breast cancer. Biometrics. 1978;347:57–67. CrossrefGoogle Scholar
33. Korn EL, Graubard BI. Analysis of Health Surveys. New York, NY: John Wiley and Sons Inc; 1999. Google Scholar
34. Korn EL, Graubard BI. Epidemiologic studies utilizing surveys: accounting for the sampling design. Am J Public Health. 1991;81(9):1166–1173. LinkGoogle Scholar
35. Reynolds SL, Silverstein M. Observing the onset of disability in older adults. Soc Sci Med. 2003;57(10): 1875–1889. Crossref, MedlineGoogle Scholar
36. Kington RS, Smith JP. Socioeconomic status and racial and ethnic differences in functional status associated with chronic diseases. Am J Public Health. 1997; 87(5):805–810. LinkGoogle Scholar
37. Ostchega Y, Harris TB, Hirsch R, Parsons VL, Kington R. The prevalence of functional limitations and disability in older persons in the US: data from the National Health and Nutrition Examination Survey III. J Am Geriatr Soc. 2000;48(9):1132–1135. Crossref, MedlineGoogle Scholar
38. Carrasquillo O, Lantigua RA, Shea S. Differences in functional status of Hispanic versus non-Hispanic white elders: data from the Medical Expenditure Panel Survey. J Aging Health. 2000;12(3):342–361. Crossref, MedlineGoogle Scholar
39. Institute of Medicine. Unequal Treatment: What Health Care System Administrators Need to Know About Racial and Ethnic Disparities in Healthcare. Washington, DC: National Academies Press; 2003. Available at: http://www.iom.edu/Object.File/Master/14/973/DisparitiesAdmin8pg.pdf. Accessed August 23, 2007. Google Scholar
40. Schneider EC, Zaslavsky AM, Epstein AM. Racial disparities in the quality of care for enrollees in Medicare managed care. JAMA. 2002;287(10): 1288–1294. Crossref, MedlineGoogle Scholar
41. Petersen LA, Wright SM, Peterson ED, Daley J. Impact of race on cardiac care and outcomes in veterans with acute myocardial infarction. Med Care. 2002; 40(suppl 1):I86–I96. MedlineGoogle Scholar
42. Garbers S, Chiasson MA. Inadequate functional health literacy in Spanish as a barrier to cervical cancer screening among immigrant Latinas in New York City. Prev Chronic Dis. 2004;1(4):A07. MedlineGoogle Scholar
43. Bard MR, Goettler CE, Schenarts PJ, et al. Language barrier leads to the unnecessary intubation of trauma patients. Am Surg. 2004;70(9):783–786. MedlineGoogle Scholar
44. Sarver J, Baker DW. Effect of language barriers on follow-up appointments after an emergency department visit. J Gen Intern Med. 2000;15(4):256–264. Crossref, MedlineGoogle Scholar
45. Ashton CM, Haidet P, Paterniti DA, et al. Racial and ethnic disparities in the use of health services: bias, preferences, or poor communication? J Gen Intern Med. 2003;18(2):146–152. Crossref, MedlineGoogle Scholar
46. Thomas SB, Fine MJ, Ibrahim SA. Health disparities: the importance of culture and health communication. Am J Public Health. 2004;94(12):2050. LinkGoogle Scholar
47. Fiscella K, Franks P, Doescher MP, Saver BG. Disparities in health care by race, ethnicity, and language among the insured: findings from a national sample. Med Care. 2002;40(1):52–59. Crossref, MedlineGoogle Scholar
48. Franzini L, Fernandez-Esquer ME. Socioeconomic, cultural, and personal influences on health outcomes in low income Mexican-origin individuals in Texas. Soc Sci Med. 2004;59(8):1629–1646. Crossref, MedlineGoogle Scholar
49. Nazroo JY. The structuring of ethnic inequalities in health: economic position, racial discrimination, and racism. Am J Public Health. 2003;93(2):277–284. LinkGoogle Scholar
50. Williams DR. Racial/ethnic variations in women’s health: the social embeddedness of health. Am J Public Health. 2002;92(4):588–597. LinkGoogle Scholar
51. Williams DR. The health of men: structured inequalities and opportunities. Am J Public Health. 2003; 93(5):724–731. LinkGoogle Scholar
52. Moritz DJ, Kasl SV, Berkman LF. Cognitive functioning and the incidence of limitations in activities of daily living in an elderly community sample. Am J Epidemiol. 1995;141(1):41–49. Crossref, MedlineGoogle Scholar
53. Markides KS, Eschbach K. Aging, migration, and mortality: current status of research on the Hispanic paradox. J Gerontol B Psychol Sci Soc Sci. 2005;60(spec no. 2):68–75. Crossref, MedlineGoogle Scholar
54. Patel KV, Eschbach K, Ray LA, Markides KS. Evaluation of mortality data for older Mexican Americans: implications for the Hispanic paradox. Am J Epidemiol. 2004;159(7):707–715. Crossref, MedlineGoogle Scholar
55. Smith DP, Bradshaw BS. Rethinking the Hispanic paradox: death rates and life expectancy for US non-Hispanic White and Hispanic populations. Am J Public Health. 2006;96(9):1686–1692. LinkGoogle Scholar
56. Ver Ploeg M, Perrin E, National Research Council Panel on DHHS Collection of Race and Ethnicity Data. Eliminating Health Disparities: Measurement and Data Needs. Washington, DC: National Academies Press; 2004. Google Scholar

Related

No related items

TOOLS

SHARE

ARTICLE CITATION

Dorothy D. Dunlop, PhD, Jing Song, MS, Larry M. Manheim, PhD, Martha L. Daviglus, MD, PhD, and Rowland W. Chang, MD, MPHDorothy D. Dunlop is with the Institute for Healthcare Studies, the Multidisciplinary Clinical Research Center in Rheumatology, and the Department of Medicine, Feinberg School of Medicine, Northwestern University, Chicago, Ill. Jing Song is with the Department of Medicine, Feinberg School of Medicine, Northwestern University, Chicago. Larry M. Manheim is with the Institute for Healthcare Studies, the Multidisciplinary Clinical Research Center in Rheumatology, and the Department of Physical Medicine and Rehabilitation, Feinberg School of Medicine, Northwestern University, Chicago. Martha L. Daviglus is with the Department of Preventive Medicine, Feinberg School of Medicine, Northwestern University, Chicago. Rowland W. Chang is with the Multidisciplinary Clinical Research Center in Rheumatology, the Department of Medicine, the Department of Physical Medicine and Rehabilitation, Feinberg School of Medicine, Northwestern University, Chicago, and the Arthritis Center, Rehabilitation Institute of Chicago, Chicago. “Racial/Ethnic Differences in the Development of Disability Among Older Adults”, American Journal of Public Health 97, no. 12 (December 1, 2007): pp. 2209-2215.

https://doi.org/10.2105/AJPH.2006.106047

PMID: 17971548