There is a need to better understand the effectiveness of HIV-prevention programs. Cluster randomized designs have major limitations to evaluate such complex large-scale combination programs. To close the prevention evaluation gap, alternative evaluation designs are needed, but also better articulation of the program impact pathways and proper documentation of program implementation. Building a plausible case using mixed methods and modeling can provide a valid alternative to probability evidence. HIV prevention policies should not be limited to evidences from randomized designs only.
aDepartment of Public Health, Institute of Tropical Medicine, Antwerp, Belgium
bInspection and Evaluation Division, UN, OIOS, New York
cPayson Center for International Development, Tulane University, New Orleans
dIndependent Evaluation Group, World Bank, Washington, District of Columbia, USA.
Correspondence to Marie Laga, Department of Public Health, Institute of Tropical Medicine, Nationalestraat 155, 2000 Antwerp, Belgium. Tel: +32 3 24763 16; fax: +32 3 24765 32; e-mail: firstname.lastname@example.org
Received 13 October, 2011
Revised 5 January, 2012
Accepted 27 January, 2012