Train to retain: the role of specialty training in stemming Malawi's medical brain drain.

Background: There is a growing awareness of the need to retain health workers in low-and middleincome countries. This is particularly the case in Malawi, with few doctors and historically high emigration. Previous retention efforts have focused on salary supplementation and expansion of undergraduate training. There has been little focus on training new doctors to become specialists, despite evidence of its value to Malawian junior doctors. In light of the considerable investment into medical education and retention, this thesis investigates the role of specialty training in stemming Malawi's medical brain drain. Methods: A tracing study was carried out to locate all Malawian doctors who graduated between 2006 and 2012. Literature reviews and qualitative interviews informed the design of a discrete choice experiment exploring junior doctors' preferences for different types of training posts. Nearly all eligible doctors in Malawi participated in the survey. The results were incorporated into a Markov model of the Malawi medical workforce as part of a cost-effectiveness analysis of expanded provision of specialty training. Results: The odds of leaving the public sector and Malawi rose with time after graduation, with most of those outside Malawi in specialty training. Junior doctors had strong preferences for different types of specialty training, with subgroups showing distinct preferences. Doctors would require substantial compensation to undertake training only in Malawi or in less popular specialties. Despite this, expanding training within Malawi was the most cost-effective means to retain doctors in the long-term, although more costly than current government spending. Conclusions: Almost all Malawian junior doctors desire to specialise, but not all specialty training is valued equally. Expansion of specialty training in Malawi, however, would lead to higher returns on investments in medical education. More cost-effectiveness modelling and a “whole-career” perspective to policy interventions would strengthen health workforce policy in low-resource settings.