A framework for optimal health worker allocation in under-resourced regions

Abstract

The effectiveness of health systems is dependent on the availability of health workers and their distribution across a healthcare system. Decision support systems for health workforce planning can play an important role in achieving an effective and equitable allocation of workers. However, existing methodologies in resource-constrained health systems rely on expert panels to determine the relationship between facility performance and staff levels rather than empirical evidence, require large amount of facility-specific data collection, and frequently fail to account for geographic, social, and economic differences between health facilities. We propose a framework for health worker allocation that overcomes some of these limitations. By integrating multiple sources of publicly available data with key facility-specific measures, statistical modeling can be used to estimate the relationship between staff allocations and facility performance. The resulting model can then be used in an optimization framework to explore how changes in policy scenarios and demographics can affect optimal staffing allocation. We explore this framework in a case study of South African health facilities, demonstrating the effectiveness of even this limited application of our framework, despite the challenges posed, and discuss the implications for future policy decisions and data collection.