RESUMEN
OBJECTIVE: We studied attrition (i.e., participant dropout) 3 years after the baseline interview in a population-based Spanish cohort. STUDY DESIGN AND SETTING: The Neurological Disorders in Central Spain study is a longitudinal, population-based survey of chronic neurologic conditions in persons aged > or = 65 years. Multivariate logistic regression analysis was used to examine baseline differences between participants who took part in the 3-year follow-up evaluation vs. those who did not. RESULTS: Of the 5,278 participants evaluated at baseline (1994-1995), 1,482 (28.1%) were lost to the follow-up (1997-1998). Factors that were associated with dropout due to refusal were as follows: living in a rural area and being single. Contact failure was associated with older age and being unmarried and widowed. Unreachable participants were less likely to take three or more medications or live in a professional neighborhood. Predictors of dropout due to death were male gender, older age, having impaired activities of daily living, taking more medications, poor self-perceived health, dementia, and smoking. CONCLUSIONS: These findings on participants who are at high risk of dropout may be useful in the planning phase of future prospective studies. Of the possible reasons for attrition, refusal is the most important because it is amenable to change.