Migration, Loss-To-Follow-Up and Population Surveillance in South Africa

Michael J. White, Brown University
Mark Collinson, University of the Witwatersrand
Kojo Antobam, University of the Witwatersrand

In this paper we analyze migration in one well-known long-running surveillance program, the Agincourt HDSS in South Africa. Migration is a key demographic event for populations monitored with surveillance systems, yet we argue that it is underappreciated in current HDSS practice. We provide (1) analysis of secondary data predicting migration, and hence LTFU; and (2) results from a randomized study of migrant contact and follow-up, testing the efficacy of in-person versus face-to-face interviews in the South African setting. Our results are designed to help researchers and administrators connected to HDSS understand migration dynamics and its potential impact on population and health studies.