Do volunteer subjects bias clinical trials?

Abstract

The availability of volunteers for research on antidepressant medications raises questions about similarities between symptomatic volunteers and clinic patients. Volunteers accepted for a depression study and patients with major depressive disorder have been shown to have some similarities with regard to depressive symptoms and responses to treatment. However, it is unknown whether the population from which volunteers are drawn differs from the patient population in other ways that affect the generalizability of research findings. Respondents to newspaper advertisements for symptomatic volunteers and new clinic patients at a university-based, outpatient psychiatric facility received the Beck Depression Inventory, the Millon Behavioral Health Inventory and a questionnaire developed for the study. The first 30 respondents in each group with a Beck Depression Inventory score of at least 10 constituted the sample. No significant differences (p less than 0.05) between the groups were found on demographics, coping styles with help givers, perceptions of psychosocial stress, or overall depression severity. However, volunteers were sadder, more discouraged, and less interested in others than clinic patients. Symptomatic volunteers also more frequently reported financial concerns as an important factor in the decision about seeking help. However, they reported the desire to find effective treatment as the primary factor in the decision. Clinic patients, on the other hand, cited being referred as primary in the seeking help decision, with the desire for effective treatment being secondary. Results of this study suggest that volunteers for drug research are similar in important ways to persons who seek treatment for similar symptomatology. They also suggest that perception of need for treatment may be affected by the strength of specific depressive symptoms.