Abstract
Background: Provider recommendation of colorectal cancer (CRC) screening is an important predictor of screening in African Americans. Little is known about influences of screening in African Americans without provider recommendations.
Objectives: The objectives of this study were to test, using the biopsychosocial (BPS) model, the relative predictive strength for screening of variables grouped into biological, psychological and behavioral, and social system factors and to compare CRC screening predictors in African Americans with and without provider recommendations.
Methods: Secondary analysis of the 2002 Maryland Cancer Survey data was done using (a) hierarchical logistic regression to examine the relative influence of factors on screening and (b) simultaneous logistic regression to examine predictors of screening in individuals with and without provider recommendations to screen. Factors included biological (age and gender), psychological and behavioral (mammogram, prostate specific antigen, body mass index, activity level, fruit and vegetable consumption, alcohol, smoking, and cancer perceptions), and social system (education, employment, insurance, and healthcare provider access).
Results: The social system factor influenced CRC screening in the overall sample (n = 492). In African Americans with provider recommendations (n = 337), the very active were 2.43 (95% confidence interval [CI] = 1.11-5.28) times more likely to screen than were less active. Insured were 3.25 (95% CI = 1.14-9.31) times more likely to screen than were uninsured. In African Americans without provider recommendations (n = 155), 65- to 69-year-olds were 9.99 (95% CI = 2.31-43.32) times more likely to screen than were those 50- to 54-year-olds without screening.
Discussion: The BPS model confirms social system factor strength in influencing CRC screening in African Americans. Identifying other social system variables that enhance healthcare provider access is critical to increase provider visits, which will generate recommendations and subsequent CRC screening.