Race- and ethnicity-specific characteristics of participants lost to follow-up in a telephone cohort

• Published in: American journal of epidemiology Vol. 140; no. 2; p. 161
• Main Authors: Psaty, B M, Cheadle, A, Koepsell, T D, Diehr, P, Wickizer, T, Curry, S, VonKorff, M, Perrin, E B, Pearson, D C, Wagner, E H
• Format: Journal Article
• Language: English
• Published: United States 15-07-1994
• Subjects: Adolescent; Adult; Analysis of Variance; Cohort Studies; Continental Population Groups; Epidemiologic Methods; Ethnic Groups; Female; Follow-Up Studies; Health Behavior - ethnology; Health Surveys; Humans; Interviews as Topic; Male; Middle Aged; Odds Ratio; Patient Dropouts; Socioeconomic Factors; Telephone
• ISSN: 0002-9262
• DOI: 10.1093/oxfordjournals.aje.a117226

The purpose of this study was to describe race- and ethnicity-specific characteristics of subjects lost to follow-up. For a study of community-based health interventions, adult subjects from 11 US communities were initially recruited by random digit dialing and interviewed by telephone in 1988; 2 years later, they were recontacted, and the same survey was administered a second time. Associations with loss to follow-up were assessed in separate models for whites, African Americans, Hispanic Americans, and Asian Americans. After 2 years, 40.8% of the 5,851 participants were lost to follow-up; cohort attrition was highest among African Americans (51.3%) and lowest among whites (37.5%). Age, aspects of employment, education, marital status, and income were significant independent predictors of loss to follow-up for one or more of the four racial and ethnic groups. Characteristics of subjects lost to follow-up in this telephone cohort differed among various racial and ethnic groups. After adjustment for demographic, socioeconomic, and health status variables, the important behavioral predictors of loss to follow-up were current smoking for whites (p < 0.05), having a high fat diet for African Americans (p < 0.10), consuming one or more alcoholic drinks per day for Hispanic Americans (p < 0.10), and high levels of physical activity for Asian Americans (p < 0.05).