Racial and ethnic differences in adult asthma prevalence, problems, and medical care

Abstract

Objectives. We document the extent to which racial and ethnic differences exist in asthma prevalence among US adults, and among asthmatic adults, we document racial differences in asthma problems and asthma-related medical care. We also explore potential explanations for racial and ethnic disparities in each outcome.

Design. Using data on a 24-state sample of white, black, Hispanic, Asian, and Native American adults from the 2004 Behavioral Risk Factor Surveillance System, we examine race/ethnic differences in asthma prevalence, problems (had an asthma attack last year; asthma symptoms; sleep difficulties; activities limited because of asthma), and asthma-related medical care (number of routine doctor visits; medication use; urgent doctor visits; visited an emergency room for asthma-related care last year). We then explore whether any observed disparities are due to differences in demographic characteristics, socioeconomic status, health behavior, and environmental conditions across racial/ethnic groups.

Results. Asthma prevalence is lowest among Asian and Hispanic adults, and highest among black and Native American adults. Considerable racial/ethnic differences in asthma-related problems and medical care are also present, with Asians doing as well or better than whites, while blacks, Hispanics, and especially Native Americans report more asthma-related problems and medical care use. For some groups (i.e., Asians and Hispanics), we were mostly unable to explain away observed differences with white adults with adjustment for potential explanatory mechanisms, while for other groups (i.e., blacks and Native Americans) adjusting for socioeconomic status and air quality accounted for much of the observed disparity with whites.

Keywords:

• asthma
• medical care
• race/ethnicity

Notes

1. Includes Alaska, Arizona, California, Georgia, Idaho, Iowa, Kentucky, Louisiana, Maryland, Michigan, Minnesota, Mississippi, Montana, New Hampshire, New Jersey, New York, Ohio, Oklahoma, Pennsylvania, Texas, Utah, Vermont, Virginia, and Wisconsin

2. Missing values on independent measures were imputed on the basis of demographic and socioeconomic predictors, using the impute command in STATA (see StataCorp 2003, for more information). For imputed measures, the majority had a very low rate of missing values (under 5%) although income was substantially higher (12.9%). Given the high rate of missing values that were imputed for income, all models that include this measure also include a control that flags imputed cases (not shown).