Associations between breast cancer risk factors and religiousness in American women in a national health survey.

Breast cancer is a leading cause of death in American women. Data are lacking from representative samples of total populations on the association of risk factors for breast cancer and religiousness. The sixth cycle of the National Survey of Family Growth (NSFG VI) included 3,766 women aged 30-44 years with complete data on self-reported religiousness, and selected breast cancer risk factors. Of women in the analysis, 1,008 reported having four or more breast cancer risk factors. Women who never attended services were over seven times more likely to report having four or more risk factors than those who attended more than weekly (P < 0.0001). After adjusting for age, race, Hispanic ethnicity, nativity, education and marital status by logistic regression, women who never attended services were still over six times more likely to report having four or more risk factors (P < 0.0001). The combination of frequent attendance at religious services, very high importance of religion in daily life, and self-identification as a Protestant evangelical was particularly protective. Multiple dimensions of religiousness are independently associated with multiple breast cancer risk factors.

Hispanic ethnicity, race and blood donation in the United States.

The aim of this study was to assess the hypothesis that blood donation rates vary with Hispanic ethnicity (family origin in Spanish-speaking countries) in addition to race in the United States. Lower blood donation rates have been reported among African Americans (AAs) compared with non-Hispanic European Americans (EAs). Adequate published reports on donation rates are not available for Hispanic Americans (HAs). Using data from a 2002 national survey, which included 4923 men and 7600 women aged 15-44 years with complete data, we tested the hypothesis using weighted bivariate and multivariate statistics. Among men aged 25-44 years, the percentage [95% confidence limits (95% CL)] with a history of blood donation since 1985 was similar at ages 25-34 years (46%, 42-49) and 35-44 years (41%, 37-45). It was highest in non-Hispanic EA (49%, 45-52%), intermediate in AA (35%, 30-40%) and lowest in HA (30%, 25-36%) (P < 0.001). Other variables significantly (P < 0.01) associated with history of blood donation in bivariate analyses were nativity (United States/other), education (<12/>or=12 years), poverty (<200%/>or=200% poverty limit) and married (yes/no). Variables that are not significantly associated were age, metropolitan residence (yes/no), receipt of public assistance (yes/no), current labour-force participation (yes/no) and religion raised. Compared with non-Hispanic EA, the adjusted odds ratios were essentially the same for Hispanics 0.66 (95% CL 0.47-0.92) and AAs 0.64 (95% CL 0.49-0.84). Only 34% of women had donated blood, but the association with race/ethnicity was similar. Similar patterns were also seen at ages 15-24 years. HAs and AAs have similar low blood donation rates compared with non-Hispanic EAs. The difference is not explained by sociodemographic variables.