Methods: A randomized control trial was conducted among 120 Hispanic patients who were eligible for colorectal cancer screening at the Texas Tech Northeast Family Practice Clinic in El Paso, Texas. All participants provided an initial interest rating for colorectal cancer screening and completed a demographic questionnaire, a knowledge questionnaire, an acculturation measure, and two patient autonomy measures. Participants read an American Cancer Society pamphlet and supplemental information on colorectal cancer screening. Participants were then randomized into one of two experimental arms. In the control condition, participants provided an interest rating about colorectal cancer screening after reading the materials. In the experimental condition, participants used a decision aid based on the Analytic Hierarchy Process prior to providing the colorectal cancer screening interest rating. Lastly, all participants completed the decisional conflict scale and the post-knowledge questionnaire.

Results: Hierarchical multiple regressions were performed to predict interest scores, decisional conflict scores and post-knowledge scores. After controlling for participant sex, age, location of the experiment (home versus clinic), insurance, education, initial preference in screening, acculturation and patient autonomy, an indicator variable representing experimental condition and the experimental condition by study location interaction accounted for an additional 5.2% of the variability in interest scores. Individuals who used the decision aid had statistically higher interest scores and the decision aid was more effective if the study was performed in the clinic. The decision aid and the decision aid by intervention arm interaction had no statistical effect on decisional conflict or post-knowledge scores. Participants positively evaluated the decision aid with respect its ease of understanding, their ability to learn useful information as a result of using the decision aid and whether they would use the decision aid in future decisions.

Discussion: Among U.S Hispanics, colorectal cancer is the second leading cause of cancer-related mortality. Investigating the utility of decision aids with Hispanic populations meets the goals of the Healthy People 2010 objectives, which call for the communication of health information to increase health literacy in traditionally under-served populations. These findings demonstrate that a decision aid promotes interest in screening and has favorable evaluations.