Evaluation and Comparison of Theoretical Models’ Abilities to Explain and Predict Colorectal Cancer Screening Behaviors

BACKGROUND: Colorectal cancer (CRC) is the fourth most common and second most deadly cancer in the United States. However, it is highly preventable and treatable if detected at the precancerous or local stage of development. There exists multiple screening methods each with varying sensitivity, required effort, and recommended frequency of use. Complete adherence to screening guidelines by the recommended, at-risk population would halve the current mortality rate. Unfortunately, screening adherence remains the lowest of all screened cancers with a median state screening adherence rate of about 65%. To understand what individual-level factors influence an individual’s decision to be screened, health behavior theory is used. However, few studies have evaluated the performance of entire behavioral theories in their ability to explain CRC screening intentions and behaviors. METHOD: Health Belief Model, Theory of Reasoned Action, Theory of Planned Behavior, and Attribution Theory were evaluated within the context of colorectal cancer screening using an online national sample (N=403) of at-risk individuals age 50 and older. Confirmatory factor analyses were performed for each evaluated construct of the theory. Structural equation models were created using the estimated constructs for each theory. Each theory was evaluated for the following screening use: colonoscopy, sigmoidoscopy, fecal occult blood test (FOBT), and general screening use. Fit statistics were estimated for each model. Models with acceptable fit were examined for significant pathways within the model as well as consistency of the model with the behavioral theory. RESULTS: All models displayed adequate fit statistics. While not all pathways were significant in each model, no estimate was the inverse in directionality to that hypothesized. This provides support that each theory lends some explanatory power and none of the theories evaluated detract from understanding CRC screening intentions and behaviors. Comparison of the models illustrates advantages to each theory and suggests potential integration of theories. CONCLUSION: The constructs of the Health Belief Model, Theory of Planned Behavior, and Attribution Theory all provide adequate explanations of individual-level CRC screening behavior influences. Although, further review and refinement of the theories is warranted and recommended