Randomized trial of a computer-tailored intervention for patients with depression.

Purpose. Examine the efficacy of a computer-tailored intervention (CTI) based on the transtheoretical model (TTM) for reducing depression, an increasingly important component of health promotion programs. Design. Pretest-posttest randomized trial. Setting. Participants were recruited and treated at home after being identified in two primary care clinics in Eastern Massachusetts and Chicago, Illinois. Subjects. A total 350 adults experiencing at least mild symptoms of depression but not involved in or planning to seek treatment for depression. Intervention. A print manual and three CTI reports tailored to stage of change for using effective methods to prevent or reduce depression, other TTM variables, level of depression, and behavior. Measures. Pre-post changes and reliable and clinically significant change on the Beck Depression scale II and pre-post changes on the 20-item Medical Outcomes Study Short Form survey-based measure of physical functioning at 9 months' follow-up. Analysis. t-tests and chi(2) tests. Complete-case analysis and two intention-to-treat analyses-assumption of no change and multiple imputation (MI)-are reported. Exploratory analyses examined whether the effects of the intervention on depression were moderated by five subject characteristics: baseline level of depression, baseline level of physical functioning, baseline stage of change for preventing or managing depression, age, and education. Results. Complete-case and intention-to-treat analyses showed that the intervention group experienced significantly greater improvements in depression (d = .220-.355); results for physical functioning were weaker (d = .150-.309) and did not reach statistical significance in the MI analysis. The effects of the intervention on reliable and clinically significant change in depression were largest among participants who were experiencing moderate depression (d = .363-.519) or severe depression (d = .603-.718) or who were in the precontemplation or contemplation stage (d = .573-.856) at baseline. (Am J Health Promot 2011;26[2]:77-89.)