Why Don’t People Take their Medicine? Experimental Evidence from Kenya

Abstract

The standard public finance approach to health care in less developed countries involves identifying and heavily subsidizing priority health interventions that generate treatment externalities (e.g., vaccines). A popular recent approach instead argues that health education, community mobilization, and cost-recovery are also necessary for successful programs. We focus on the case of intestinal worms – which affect one in four people worldwide – and provide evidence from a series of randomized evaluations that subsidies are more effective than other factors in influencing deworming drug take-up among schoolchildren in Kenya. An effort to educate children on worm prevention was ineffective at changing behavior. Those who were randomly exposed to more information about the deworming drugs through their social networks were significantly less likely to take them. A verbal commitment “mobilization” intervention from social psychology led to significantly lower treatment rates. In contrast, drug take-up was highly sensitive to cost: the introduction of a small fee led to a massive eighty percent reduction in treatment rates (relative to free treatment). Cost-sharing appears particularly inappropriate for medical conditions characterized by large treatment externalities – like worms and other infectious diseases. The results suggest that, to promote new health technologies in less developed countries, scarce public resources should be focused on subsidies. * Contact: Edward Miguel (emiguel@econ.berkeley.edu) or Michael Kremer (mkremer@fas.harvard.edu). The authors thank ICS Africa and the Kenya Ministry of Health, Division of Vector Borne Diseases for their cooperation in all stages, and would especially like to acknowledge the contributions of Elizabeth Beasley, Laban Benaya, Simon Brooker, Pascaline Dupas, Alfred Luoba, Sylvie Moulin, Robert Namunyu, Polycarp Waswa, Peter Wafula, and the entire PSDP field staff and data group, without whom the project would not have been possible. Gratitude is extended to the teachers, school children, and households of Busia for participating in the study. Melissa Gonzalez-Brenes and Tina Green have provided excellent research assistance. We thank George Akerlof, Guido Imbens, Botond Koszegi, David Laibson, Kaivan Munshi, Mark Rosenzweig, and Chris Udry for helpful conversations. We are grateful for financial support from the World Bank, the NIH Fogarty International Center (R01 TW05612-02), and U.C. Berkeley Center for Health Research. All errors are our own.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 11 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.