Health Disparities in the US Population: A Comparison of Latino, Asian, and White American Older Adults

Background/Purpose: Health disparities and factors contributing to health conditions have received considerable attention, but findings to date are inconsistent (Adams, 2010; Manary et al., 2013).  Moreover, studies on racial/ethnic disparities in health habits and health outcomes using large-scale representative samples are limited in number.  The purpose of the present study is to examine racial/ethnic differences in direct and indirect effects of selected predictors toward chronic health conditions as an outcome variable among Latino, Asian and White American older adults.    

Methodology: This study examined the 2013-2014 California Health Interview Survey (CHIS) combined data, which were collected with a multi-stage strata sampling design.  A subsample of Latino (n=919), Asian (n= 1,198), and White American older adults (n=12,476) aged 65 and older (total N=14,593) was examined with an overall mean age of 73.1 years old (sd=7.00).  Predictors of chronic health conditions included three latent variables of SES, psychological distress (Cronbach α=0.923), and health habits (smoking, drinking & dietary habit) in addition to an observed variable of health coverage.  The latent outcome variable is chronic health conditions measured with diabetes, hypertension, heart disease and overweight.  In order to find racial/ethnic differences in effects of these four predictors, a Structural Equation Modeling (SEM) with multi-group tests was conducted. 

Results: Latino and Asian older adults were significantly less educated (χ²=2730, df=6, p=.000), more impoverished (χ²=760, df=8, p=.000), far more foreign-born (χ²=5816, df=4, p=.000), more likely to stay married (χ²=90, df=6, p=.000), and less favorably rated their general health (χ²=843, df=8, p=.000) as compared to White American counterparts.  It is worthwhile to note that there were significant differences in chronic health conditions among the three groups of older adults.  Significantly more Latino and Asian older adults were diagnosed with diabetes (χ²=254; df=2; p=.000) and hypertension (χ²=32; df=2; p < .01), while more White Americans were diagnosed with heart disease (χ²=26; df=2; p=.000), and far more Latino and White Americans had overweight (χ²=343; df=2; p=.000) compared to Asian older adults. The hypothesized SEM model tested for each racial group fits well to the Latino, Asian and White subgroup data (CFI=.969, RMSEA=.032, Δχ²/Δdf=1.95; CFI=.965, RMSEA=.029, Δχ²/Δdf=1.987; and CFI=.887, RMSEA=.048, Δχ²/Δdf=29.56 respectively).  Most notably, health habits significantly predicted chronic health conditions for White older adults (β = -0.692; p = 0.000), while health habits did not have a significant direct effect on chronic health conditions for Latino and Asian counterparts. On the other hand, SES had a direct significant effect on chronic health conditions only for Latino older adults (β = -0.482; p = 0.048).

Conclusions and Implications: Findings from this study suggest racial/ethnic differences in the effects of four predictors on chronic health conditions in Latino, Asian and White American older adults.  This study presents important implications for ethnic sensitive social work practice as well as public health literacy/education programs.