Church rosters: is this a viable mechanism for effectively recruiting African Americans for a community-based survey?

OBJECTIVES: The purpose of this report is to describe the process, results, and implications in the phone recruitment of African Americans through church rosters for a survey of diet-and blood pressure-related awareness and hypertension prevalence.
DESIGN: The survey was conducted using a non-probability sample of churches and a random selection of participants from church rosters. Recruitment strategies included frequent contact with pastors and church representatives, presentations, standard and tailored recruitment approaches, and bi-annual progress reports. Church representatives provided the rosters and assisted in arranging interviews, which were conducted at church or the participants' homes.
RESULTS: Of 742 randomly selected, 315 (42.4%) were ineligible because of an unavailable or unreachable number, a move, discontinued church membership, death, or other reasons. Of the 344 eligible, 45.8% participated, 30.2% refused, 4.4% agreed to participate but did not, and 19.6% were incompletes (called less than three times before recruitment was terminated). Among participants, 70.4% were female, 58.2% had completed college, and the age range was 19-91 years. The survey's sample size goal of 196 was met.
CONCLUSIONS: In this study population, over 45% who were eligible participated. Rapport established with church representatives and congregations was critical to the sampling process. Using church rosters can be a low-cost, effective recruitment tool. However, key factors to consider when recruiting African Americans in this manner include: trust, study eligibility criteria, roster accuracy, and time, and generalizability.