Cancer screening compliance in an underserved population

A112

Introduction: Cancer screening has been shown to be effective in prevention and early detection of certain cancers. However, one of the barriers to the effectiveness of screening programs is low compliance with the recommended frequency of screening. This study was undertaken to investigate factors that affect compliance with cancer screening recommendations in a historically underserved population.
 Methods: 7362 individuals age 40 and over screened for breast, cervical, colon or prostate cancer in Northwestern Louisiana since 1999 were used for study. For this study, screening compliance is defined as average return to screening clinic between 300 to 540 days since initial screening. Factors studied include age, sex, race, marital status, smoking history, personal history of cancer, family history of cancer, chronic disease status, income, physical activity and alcohol ingestion history. Chronic disease status is defined as self-reported ever had any following diseases: heart disease, heart attack, stroke, diabetes, and blood clot and polyps of colon. All p-values < 0.05 are considered as statistically significant.
 Results:
 The overall compliance rate was 37.6%. Mean age of participants was 54.7 years with 64.3% of the population female, 55.8% African-American, 46.6% married, and 25.3% current smokers. Univariate analyses showed a statistical difference in compliance as follows with the % compliance in parentheses: income < $20,000 (41.2%) versus income ≥ $20,000 (53.0%); non-married (36.0%) versus married (42.3%); non-smokers (39.8%) versus smokers (31.2%); alcohol ingestion (44.6%) versus no alcohol ingestion (38.5%); limited physical activity (39.0%) versus greater physical activity (52.1%); no chronic disease (45.5%) versus any chronic disease (42.0%); and no family history of cancer (36.5 %) versus positive family history of cancer (41.2%). Age group, sex, personal history of cancer, and racial differences were not statistically significant. With multivariate logistic regression, it was found that race, sex, income, physical activity, smoking status and family history of cancer were highly significant predictors of compliance. The association of predicted probabilities and observed responses is 61.7%.
 Conclusion: It appears that among African-Americans those with a family history of cancer, smoking history, income > $20,000 per year, physically active and female were more likely to comply with screening recommendations suggesting that special attention be given to the other subpopulations.