Background: Despite the popularity and widespread practice of complementary/alternative medicine (CAM), researchers may face problems accruing patients to randomized clinical trials, considered the gold standard of biomedical research. Strict exclusion criteria and barriers to participation may limit accrual. Inadequate numbers of subjects decrease the ability of studies to detect an effect that exists and generalize their findings. This article describes the recruitment experience of a CAM trial, details reasons for non-participation, and contrasts participants and non-participants on demographic, clinical, and treatment-related variables.
Methods: Women who were Houston area residents and spoke English, had primary breast cancer (excluding Stage IV), and were 1 to 30 months posttreatment with no steroids, tamoxifen, substance abuse, psychiatric or heart disease, or immune deficiency were eligible. The enrollment process involved three contacts (i.e. introductory letter and brochure, telephone calls, and reminder post cards). Potential participants were told that the study would require blood samples (30 cc) to assess immune function; psychosocial measures to assess emotional well-being, quality-of-life, social support, and coping strategies; and possible assignment to six weekly support or imagery sessions. Factors influencing recruitment and reasons for non-participation were assessed by stratified analysis and multivariate logistic regression.
Results: Of 158 eligible participants, 30% (N = 47) consented to participate. Primary reasons for non-participation included work/childcare (33.3%), transportation/travel (30.6%), and lack of interest (24.3%). Participants were more likely to be 40-54 years of age versus younger or older, divorced/separated, and able to pay some/all medical expenses. Divorced or separated women appeared to be more likely to participate, regardless of financial status.
Conclusions: Researchers must assess the impact of exclusion criteria on accrual and recognize the special needs of their target population. Although age, marital status, and pay status were the strongest predictors of participation, these factors are not amenable to intervention. Based on this study, researchers might boost accrual by providing interventions available during the day and evening to accommodate working women, child care services, transportation, or reimbursement for travel costs.