P-B24 Factors associated with attrition in a prospective cohort study in Nigeria. (January 2016)

• Record Type: Journal Article
• Authors: Olaniyan, Olayinka; Dareng, Eileen; Adebamowo, Sally; Offiong, Richard; Dakum, Patrick; Adebamowo, Clement
• Is Part Of: Journal of acquired immune deficiency syndromes. Volume 71(2016)Supplement 1
• Publication Date: 2016-01
• Subjects: AIDS (Disease) -- Periodicals; Acquired Immunodeficiency Syndrome -- Periodicals; AIDS (Disease); Periodicals; 616.9792005
• DOI: 10.1097/01.qai.0000479596.35968.71
• Languages: English
• ISSNs: 1525-4135

Abstract

Abstract : Prospective studies with a high proportion of attrition can suffer from selection bias. It is important to assess factors associated with attrition and reasons for study drop out, as results may help in study design. We designed a prospective cohort study to evaluate epidemiological, host and HPV-viral factors associated with cervical pre-cancer. We recruited 1020 women in Nigeria and followed them over a median period of 14.4 months with scheduled clinic visits at 6 and 12 months. Women with at least 2 visits during the study period were considered responders. We conducted exit phone interviews for non-responders and compared demographic, lifestyle, reproductive and sexual characteristics of responders and non-responders using logistic regression and explored the reasons for attrition. Of the 1020 women enrolled, 717 (70%) returned for at least one follow up visit. Of the sociodemographic characteristics evaluated (age, marital status, length of time at residence, educational level, religion and socioeconomic status), only age was significantly associated with attrition, with older women being less likely to drop out than younger women (OR: 0.96, 95% CI: 0.94 to 0.98, P < 0.001). Of the lifestyle risk factors evaluated (smoking, alcohol consumption, exercise frequency, presence of other chronic ailments, personal perception of health status and HIV status), HIV infection was statistically associated with attrition, with HIV positive women being less likely to drop out than HIV negative women (OR: 0.46, 95% CI: 0.34 to 0.62, P < 0.001). The main reasons for study drop out were inability to be reached (39%, 118/303) appointment no show (34%, 103/303), ineligibility during study period (16%, 47/303) and voluntary withdrawal (11%, 33/303). Inability to reach participants was the commonest reason for study drop out. Future prospective cohort study designs in a developing country like Nigeria, need to account for this in sample size considerations and plan to reduce this by collecting adequate tracking information at recruitment.