Risk factor distribution among sociodemographically diverse African American adults

Because African Americans tend to have lower socioeconomic status (SES) tha n whites and numerous health indicators are related To SES variables, it is important when examining between-group differences in health indices to ac count for SES differences. This study examined the effects of income and ed ucation on several biologic and behavioral risk factors in a sample of soci odemographically diverse African American adults. Approximately 1,000 Afric an American adults (aged 18-87) were recruited from 14 churches with predom inantly black membership to participate in a nutrition education interventi on. Demographics, height, weight, blood pressure, self-reported cigarette a nd alcohol use, self-reported diet by food frequency questionnaire, serum c arotenoids, serum total cholesterol, and nutrition knowledge were assessed. The association of these risk factors were examined by four levels of educ ation and income. For men, body mass index, blood pressure, total cholester ol, daily intake of fruits and vegetables, serum carotenoids, heavy alcohol Else, or exercise were not associated significantly with income or educati on using analysis of variance (ANOVA). Past month alcohol use and nutrition knowledge were associated positively with education, bur nor income. For w omen, body mass index and smoking were associated inversely with income, bu t not with education. Blood pressure, total cholesterol, intake of fruits a nd vegetables, heavy alcohol use, and exercise were not associated with eit her income or education using ANOVA. Serum carotenoids, any 30-day alcohol use, and nutrition knowledge were associated positively with both income an d education. Results using linear regression generally were similar for men and women, although a few more variables were associated significantly wit h SES compared to ANOVA analyses. Several health indicators that have been associated with socioeconomic variables in whites were not associated or on ly weakly associated in this diverse sample of African Americans. One inter pretation of these findings is that SES factors may function differently am ong blacks and whites.