Factors contributing to cancer screening in African Americans

Abstract

6514 Background: To address disparities in cancer mortality, the factors affecting participation in screening need to be identified. For African American cancer survivors in particular, little is known about continued participation in screening. The purpose of this study was to identify variables that influence compliance with cancer screening in African Americans. Methods: For this descriptive, case-control study of African Americans (500 cancer survivors, 512 non-cancer controls), the survivor group included breast (n=214), prostate (n=197), and colon cancer (n=89) survivors from 16 CALGB institutions throughout the country, who were essentially cancer free and had completed primary treatment at least 3 years earlier. Non-cancer controls were selected via random digit dialing and were matched as a group to the survivors, so the two groups were similar in ethnicity, gender, education, and city of residence. Multiple regression modeling was used to evaluate the contribution of 14 variables to cancer screening participation. Results: An unadjusted comparison showed that the cancer survivors participated more frequently in cancer screening tests than the non-cancer controls (p< .001). For both groups participation in screening was explained by six variables: female gender, greater trust in their physician, college education, having health insurance, city of residence, and more comorbidities. All contributed to more frequent screening. The strongest predictors were gender and trust in physician (p< .0001), which together explained 11% of the variance. Culturally specific variables thought to influence screening were not significant: beliefs in cultural myths, felt racial discrimination, and spirituality. Age, anxiety about cancer, general anxiety, depression, social support, and quality of life also were not significantly related to screening. No interactions were significant. Conclusions: Trust in the patient's own physician was one of the strongest variables associated with screening for both African American cancer survivors and those who never had cancer. Since trust can be directly influenced in clinical practice, these findings suggest future work to facilitate the development of trust, particularly for African American men. No significant financial relationships to disclose.