THE STUDY CONDUCTED by Ellen and colleagues1 in this issue reports what we have come to expect of studies and race and sexually transmitted disease in the United States: that African-American adolescents are at greater risk for sexually transmitted diseases (STDs) than adolescents who are white, Hispanic, or members of other racial/ethnic groups. Once again, with almost elegant simplicity, race succeeds in explaining a significant portion of the variation in individual health status, and once again, results confirm that African-Americans appear to be at a particular disadvantage.
The authors hypothesize that these differences in the rates of reporting a history of an STD may arise because African-American adolescents are more likely to have sexual contact with "high risk" adults who are members of so-called "core groups" among whom higher rates of STDs are prevalent. African-Americans are also more likely, the authors suggest, to live in areas where such core group members reside and are, therefore, more likely to be exposed to infection.
The fact that the rate of a reported STD history decreased with the level of educational attainment of parent suggests strongly that race and social class may be markers for the respondent's neighborhood. The sons and daughters of African-American parents with less than a high school education are likely to live in poorer neighborhoods with high rates of risk behaviors associated with STD infection.2,3 Those with better educated parents are, by contrast, less likely to live in such areas.
The relationship between race, poverty, and geography is particularly strong in the United States. Poverty is increasingly becoming concentrated in communities of color-most notably among communities of African-Americans-and with this concentration comes a host of social and public health problems. Massey and colleagues, for example, write:
The geographic concentration of poverty is alarming: The intense clustering of poor people in neighborhoods leads to a concentration of other deleterious social and economic circumstances associated with poverty. The fact that black poverty became more concentrated during the 1970s and 1980s implies a simultaneous concentration of crime, violence, welfare dependency, family disruption, and educational failure. These harsh trends have produced an increasingly harsh and extremely disadvantaged social environment for African-Americans that has undermined their broader well-being in society.2
Ellen and colleagues call for future studies that "directly examine the behaviors and prevalence of STDs among sex partners of adolescents." Ethnographies that focus on the social and sexual partner networks of African-American adolescents seem particularly appropriate for achieving these objectives. We would do well to remember that in examining the dynamics of STD transmission, we are examining discrete, intimate behaviors that are contained in what Potterat4 has described as "socio-geographic space." This "space" marks the locations where sexual partners meet, socialize, have sex, are exposed to an STD (or not), and engage in other social activities. Significantly, these spaces are finite and can, in some instances, be easily enumerated.
An example of the clustering of STD cases in place and space comes from the classic study of gonorrhea in Colorado Springs5 where cases exhibited residential concentration (51% of the cases were found in just 4 census tracts) and frequented a comparatively small number of nighttime leisure activity locales (6 out of a possible 300). Significantly, although African-Americans were overrepresented among gonorrhea cases in this study, not every African-American in Colorado Springs was at risk for an STD. In the language of the logician, African-American race was neither a necessary nor sufficient condition for participation in the "sexual lots" among whom gonorrhea infection was confined in this community.
The elegance of that study serves to raise important questions about the utility of studies that focus on the use of race as a predictor of risk behavior, particularly antisocial behavior. At this point in the 1990s, the results of such studies are much more likely to stigmatize the populations studied than to help them confront the problems under study.
The social consequences of racial comparisons in medical publications may depend more on the perceptions that adverse comparisons reinforce in society than the accuracy of the data, the intent of the authors, or the correctness of their interpretation. Does the impression that nonwhites are more violent, disease prone, and mentally deficient influence the perception landlords, real estate agents, and police officers have about African-Americans? Does the medical literature reinforce these prejudices? (p. 278)6
The question is troubling and should force us to rethink our notions about the conduct of future studies of STDs called for by Ellen and colleagues. For example, standard survey techniques for creating a random, racially stratified sample will do an excellent job of representing the general population but will be a poor tool for studying the dynamics of STD propagation within locally discrete risk networks. By contrast, snowball sampling techniques in which members of risk networks identify their contacts and describe their interactions with them appear to be much more promising for studying STDs.7 In such research, the race of a network member is far less critical for our understanding of how a pathogen is transmitted than information about where, with whom, and under what conditions did a sexual encounter occur.
The authors are correct in asserting that race/ethnicity is an important starting point for identifying such networks, but at some point STD researchers must do more than equate risky sexual behavior with race. The interventions that they enjoin us to develop require a much more in-depth understanding of how adolescents move in and out of social and sexual networks. At the risk of speaking the obvious, STD prevention campaigns targeted at some black adolescents-hopefully those who are at greatest risk-appear far more cost-effective than those targeting all black adolescents. As one of the most stigmatized groups in 20th century America, these youngsters deserve no less.
1. Ellen JM, Aral SO, Madger LS. Do differences in sexual behaviors account for the racial/ethnic differences in adolescents' self-reported history of a sexually transmitted disease? Sex Transm Dis 1998;25:125-129.
2. Massey DS, Gross AB, Shibuya K. Migration, segregation, and the geographic concentration of poverty. Am Soc Rev 1994; 59:425-445.
3. Lillie-Blanton M, Anthony JC, Schuster CR. Probing the meaning of racial/ethnic group comparisons in crack cocaine smoking. JAMA 1993; 269:992-997.
4. Potterat JJ. "Socio-geographic space" and sexually transmissible diseases in the 1990s. Today's Life Sci 1992; 5:16-31.
5. Potterat JJ, Rothenberg RB, Woodhouse DE, Muth JB, Pratts CI, Fogle JS. Gonorrhea as a social disease. Sex Transm Dis 1985; 12:25-32.
6. Osborne NG, Feit MD. The use of race in medical research. JAMA 1992; 267:275-279.
7. Rothenberg R, Narramore J. Commentary: The relevance of social network concepts to sexually transmitted disease control. Sex Transm Dis 1996; 23:24-29.