Medical expenditures among immigrant and nonimmigrant groups in the United States: findings from the Medical Expenditures Panel Survey (2000-2008)

Abstract

Objective: The objective of the study was to examine time trends and differences in medical expenditures between noncitizens, foreign-born, and US-born citizens.

Methods: We used multi-year Medical Expenditures Panel Survey (2000-2008) data on noninstitutionalized adults in the United States (N=190,965). Source specific and total medical expenditures were analyzed using regression models, bootstrap prediction techniques, and linear and nonlinear decomposition methods to evaluate the relationship between immigration status and expenditures, controlling for confounding effects.

Results: We found that the average health expenditures between 2000 and 2008 for noncitizens immigrants ($1836) were substantially lower compared with both foreign-born ($3737) and US-born citizens ($4478). Differences were maintained after controlling for confounding effects. Decomposition techniques showed that the main determinants of these differences were the availability of a usual source of health care, insurance, and ethnicity/race.

Conclusions: Lower health care expenditures among immigrants result from disparate access to health care. The dissipation of demographic advantages among immigrants could prospectively produce higher pressures on the US health care system as immigrants age and levels of chronic conditions rise. Barring a shift in policy, the brunt of the effects could be borne by an already overextended public health care system.

Figure 1

Trends in average, share^d, and share realtive to population size^e total expenditures among adults 18 years and older in the United States by immigrant status. Results are from the Medical Expenditures Panel Survey (2000–2008). ^a Adult (18 Years and Over) U.S. Population ^b Adult (18 Years and Over) U.S. Population ^c Top Decile of Adult Spenders ^d Share of overall expenditures indicates total group healthcare expenditures as a percentage of overall U.S. healthcare expenditures among adults 18 years and over. ^e Share relative to population size indicates ratio of group healthcare expenditures as a percentage of overall expenditures to group size as a percentage of the U.S. adult (18 years and over) population.

Figure 2

Trends in source specific (i.e. out-of-pocket, private insurance and public insurance) average proportions of total expenditures^a among adults 18 years and older in the United States by immigrant status. Results are from the Medical Expenditures Panel Survey (2000–2008). ^a Indicates ratio of healthcare expenditures paid by source relative to overall healthcare expenditures.

Figure 3. Fairlie and Oaxaca-Binder decompositon of differences in expected probability and log of expenditures between adult (18 years and older) immigrant groups in the United States. Results are from the Medical Expenditures Panel Survey (2000–2008)

† Comparison group * Reference group Note 1: “NC” indicates Non-citizens, “FB” indicates Foreign-born citizens and “USB” indicates U.S.-born citizens. Note 2: “USC” indicates a Usual Source of Care. Note 3: Bars included in the negative quadrant of the graph represent factors that lower the level of spending among comparison groups. Bars included in the positive quadrant of the graph represent factors that increase the level of spending among the comparison groups. The width of each included bar represents the percentage of between group difference in outcome explained by the factor or, more specifically, the expected change in group outcome difference if both groups had similar factor characteristics. For example, USC is an unfavorable factor for the NC group leading to a lower expected probability of spending among group members compared to the USB group. It accounts for 28.6% of the difference in the estimated probabilities of spending between the USB and NC groups. Stated differently, having similar USC characteristics would narrow the difference in the estimated probabilities between the USB and NC groups by 28.6%.