A randomized trial of incentives to improve response rates to a mailed women's health questionnaire

Background: Mailed questionnaires can be a convenient method for collecting data on women's health, although poor response rates are a concern. Methods: As part of a survey of women's health conducted in Maryland in 2001, a randomized trial was performed to assess the effects of two incentives ($1.00 or a lottery ticket) as well as precontact with an introductory postcard on response rates. Questionnaires were mailed to 3000 women aged 40-60 who were randomized to one of six incentive/precontact groups: lottery/postcard, money/postcard, postcard only, lottery only, money only, or no incentive/no postcard. Results: The overall response rate was 37.6%. Each incentive/precontact group yielded a higher response rate than the no incentive/no postcard group, although only the response rates for the lottery/postcard group (41.3%) and the money only group (40.0%) were significantly higher than that of the no incentive/no postcard group (33.1%). Money was the only factor that had a significant independent effect on likelihood of response (hazards ratio [HR] compared to no incentive = 1.22, 95% confidence interval [CI] = 1.03, 1.43). Response rates were lower in minority ZIP codes, although the effects of the incentives were generally greater than in the nonminority ZIP codes. Conclusions: These results indicate that response rates to mailed women's health questionnaires may be improved with modest incentives, particularly cash incentives.