By Carol C. Korenbrot (Click here to view the entire P&R issue)
Race and ethnic data play an important role in national, state and some local health policies. A number of public databases document differences in health status and services among different racial and ethnic groups, and many of the differences indicate a need to prioritize policies for different ethnic groups:
* African-Americans have shorter life expectancies than white Americans.
* The excess mortality of black males has become progressively greater for almost all major chronic diseases during the last 40 years.
* Homicide is the leading cause of depth for African-American women ages 15-34.
* African American women ages 25-55 are three to four times more likely to die of heart disease, stroke and complications of pregnancy than white women.
* African American and Hispanic women have higher rates of diabetes, hypertension and cardiovascular disease than white women.
* African American and Hispanic women (one-fifth of all women) account for three-fourths of all US women who are reported to have AIDS.
* Whites have more contacts with physicians, and most are by telephone or office visit; blacks have had more physician contacts in a hospital, clinic or emergency room.
* More Hispanics (27%) and blacks (19%) have no health insurance than whites (12%).
At the same time, these government data reveal dangers of the double-edged sword of race-categorized health data that can both help and hurt racial and ethnic minorities. Data that document higher rates of poor health status or higher use of public health services in a particular group can be used to stigmatize all members of the group. In addition, confounding social, demographic, economic or political factors that contribute to disease are rarely factored out of published ethnic differences in health. Either of two problems can develop as a result:
* Health policies are changed with no impact on health status, because far more sweeping policy changes in housing, employment or general assistance are needed; or
* Health policies are not changed because the health differences are ascribed to the underclass characteristics that explain the differences and are considered beyond the scope of health care policy.
The definition and use of racial categories in health data were largely unquestioned until recently because of the tacitly accepted role of genetically transmitted physical characteristics in defining racial groups (such as skin colors), as well as the role of genetically transmitted physical characteristics in health status differences (such as skin cancers). When race is used in health data, there is a tendency to assume that a genetic reason may explain differences. But often there is no known potential genetic explanation for differences that are documented. Until recently, there has been little research dedicated to how well or how poorly the racial categories of health data used for public policy actually meet scientific criteria of mutually exhaustive and exclusive groupings of people by any definable characteristics. A recent multiracial Public Health Service workshop concluded, “Current concepts of race and ethnicity in public health surveillance data lack clarity, precision and consensus.” It is important to note, however, that the group recommended not to abolish racial/ethnic categories but to: “Establish definitions for race and ethnicity tailored for specific purposes in public health….”
In our paper on health data for the PRRAC Federal Data Reconnaissance Project, we documented the usefulness of racial group-specific data in monitoring the progress of the country towards equity in health status and health services. It would be a shame to allow the recent burgeoning progress in understanding of ethnic differences in health to be threatened by political attempts to reduce ethnic categorization. Clearly, more categories are necessary, not fewer. At the same time, continued action is needed to avoid biased attitudes, conscious and subconscious, not just in the categorization of racial groups, but in the interpretation and use of the data.
