Follow-up of a large cohort of Black women.

High retention rates in follow-up studies reduce the potential for biased data due to selective losses. The Black Women's Health Study began in 1995 when 64,500 participants aged 21-69 years enrolled by completing postal health questionnaires. Follow-up is carried out biennially. On the basis of data collected between enrollment and completion of the first follow-up, the authors assessed the usefulness of various follow-up methods and compared the characteristics of respondents, nonrespondents, and women lost to follow-up because of an unknown address. The 1997 questionnaire was completed by 82.8% of the participants. The study population was highly mobile: 56.5% moved at least once, and 1.5% moved at least four times. Moving was associated with younger age: A total of 71.7% of participants aged 21-29 years moved at least once compared with 43.2% of women aged 50-69. The most successful and cost-effective method for eliciting completed questionnaires from participants was sending multiple waves of questionnaires. Telephone calls to nonrespondents were successful but were highly labor intensive. Demographic and health characteristics of the women were similar regardless of which mailing was completed, except that early respondents had higher levels of education. Respondents were more highly educated and older than were nonrespondents and lost subjects but were quite similar in all other characteristics. These data suggest that follow-up of a mobile population of African-American women can be successful.