IJE Advance Access originally published online on May 6, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
International Journal of Epidemiology, Volume 33, Number 3, pp. 493-498
IJE vol.33 no.3 © International Epidemiological Association 2004; all rights reserved.
Article |
Reducing attrition in panel studies in developing countries
Department of Epidemiology and Population Health, London School of Hygiene and Tropical Medicine, 49–51 Bedford Square, London, WC1B 3DP. Tel: 020 7299 4631, Fax: 020 7299 4637. E-mail: Zelee.Hill{at}lshtm.ac.uk
Background Panel studies offer repeated observations of individuals over time, but the mobility of populations in the developing world often causes attrition in panel studies. Such attrition can cause bias if it is selective but can be reduced by tracking respondents. Tracking in developing countries can be costly and difficult as populations are often highly mobile, infrastructure is poor, structures frequently change, and formal address systems or population records rarely exist.
Method In this paper, the attrition and tracking experiences of panel studies in developing countries are reviewed and recommendations made for ensuring effective tracking.
Comments Tracking can reduce attrition by up to 45% and is feasible if procedures are locally appropriate, well planned, involve the community, collect as much locating data as possible, and have explicit criteria, and if tracking is done at regular intervals, and interviewers are well trained, supervised, and motivated.
Conclusion Attrition is an important issue in panel studies, whilst tracking can be costly it can reduce attrition if effective procedures are used.
Accepted 23 October 2003