Article Text
Abstract
Introduction Longitudinal cohort studies provide unique insights but preventing drop-out and missing values is challenging. The aim of this study is to describe how to recruit and maximise retention of 400 Chlamydia trachomatis (CT) positive women in FemCure, an ongoing longitudinal multicenter cohort study aiming to recruit (extra)genital CT positive women.
Methods Recruitment took place at 3 STI-clinics (South Limburg, Amsterdam and Rotterdam). Of the invited women 29% joined. They collected vaginal, anorectal and (nurse-taken) oral swabs and completed online questionnaires pre-treatment, 1, 2, 4, 6, 8, 10 and 12 weeks post-treatment. To minimise loss to follow up (ltfu) various reminders were sent, small incentives were given and no show at the 2nd visit were replaced. Logistic regression analyses were used to assess predictors for ltfu (i.e. age, education level, steady partner, ethnicity, previous CT diagnosis, recruitment clinic, and anatomic site of infection). The results include the first 7 months of recruitment.
Results By 11-dec-16 143 participants completed fu. Of them, 81% had complete data, 17% (n=24) was ltfu, of which 18 before the 2nd visit. Nine cases had missing data. Of all questionnaires and swabs (n=1144), 12% was completed after a reminder. Multivariate logistic regression showed that education and age were associated with ltfu. Low education was related to more ltfu than higher education (OR=2.8, 95% CI 1.1–6.8). Those aged 23=> were related to more ltfu than those <23 (OR=2.5, 95% CI 1.0–6.4). No other variables were statistically significantly associated, this may change with ongoing recruitment. At the last visit 86% reported to participate again in a similar study (reasons given include helping others/friendliness of the staff) while the others mention time constraints as a barrier.
Conclusion These preliminary results show a retention rate of 81% without missing data. Approaches described, especially combined with committed healthcare providers can be used to motivate participants to complete fu in future longitudinal multicenter studies.