Factors affecting sample selection in a randomized trial of balance enhancement: The FICSIT study

OBJECTIVE: To determine demographic, functional, and health-related factors that may have influenced the selection of older adults for a randomized trial of balance enhancement. DESIGN: Comparison of participants with nonparticipants at various stages of the recruiting process. SETTING: Northeastern suburban community. PARTICIPANTS: Registered voters aged 75 and older (n = 7191). MEASUREMENTS: Demographic, health-related, functional, balance, gait, and falling characteristics. RESULTS: The overall participation rate in the randomized trial was 1.5%. Compared with nonparticipants, participants were significantly more likely to be male, married, living with others, living in a house, highly educated, healthy, and physically active. CONCLUSION: Recruiting older subjects by mail to studies of rigorous interventions can produce significant selection biases that may limit the population to which results can be generalized.