Understanding the Prevalence of Diabetes and Associated Health Care Expenditures by Nativity and Years in the U.S.: Testing Age as a Moderator

Purpose: This study compared 1) a “healthy immigrant effect” between younger (18-64) and older (65+) age groups focusing on the prevalence of diabetes, and 2) the total health care expenditures associated with diabetes by age group and nativity. In addition, this study examined the factors related to having diabetes among foreign-born individuals. By doing so, this study aimed to increase the understanding of heterogeneity within the immigrant population in terms of age, length of residence in the U.S., and regions of origin.

Methods: The 2003-2007 pooled Medical Expenditure Panel Survey (MEPS) linked to the National Health Interview Survey (NHIS) was analyzed (N=147,545). Bivariate statistics and logistic regression models were used to compare the prevalence of diabetes by immigration status (U.S.-born; foreign-born with 0-10 years in the U.S.; foreign-born with 11+ years in the U.S.). Separate models were tested by age group (moderator; 18-64 vs. 65+) to examine whether the relationship between immigration status and the likelihood of having diabetes differs in the younger and older age groups. The prevalence of diabetes was calculated based on the question asking whether an individual was told by a doctor or other health care professionals he/she had diabetes. To examine heterogeneity within the immigrant population, additional models were tested among foreign-born individuals with regions of origin and legal status variables added. Finally, separate OLS regression models by age group were tested for the total medical expenditures among those with diabetes while controlling for self-rated health and ten other health conditions. The STATA survey procedures were used to account for the complex sampling design.

Results: The findings supported a healthy immigrant effect in relation to diabetes only among younger adults. While 2.3% of recent immigrants with less than 10 years in the U.S. suffered from diabetes, the percentage was 6.8% among longer-term immigrants, and 5.8% among the U.S.-born. In the older adult group, recent immigrants had the highest prevalence of diabetes (29.4%) compared to longer-term immigrants (18.6%) and the U.S.-born (18.5%). However, these bivariate differences disappeared in the older age group after controlling for SES and demographic factors. Within the immigrant population, immigrants from South/Central America, Russia, and Middle East were more likely to have diabetes across the age groups compared to their counterparts from European countries. Among younger immigrants, immigrants from Asia and Africa were also more vulnerable than younger immigrants from Europe. In terms of the total annual health care spending, among those who were diagnosed with diabetes, the U.S.-born had higher health care expenditures than their recent immigrant counterparts (p=0.02) only in the younger age group after controlling for SES/demographic characteristics, health, and health insurance status.

Implications: Prevention and treatment of diabetes requires communication between an individual and a health care professional because diabetes is closely related to the overall life-style of an individual. This study supports targeted health educational interventions depending on their age, nativity, length of stay, and regions of origin, considering different prevalence rates and health service utilization across different sub-groups.