Researchers find ‘return-of-service’ agreements are not effective in retaining skilled healthcare workers in four Southern African countries, but identify factors for better outcomes.

Millions of people around the world are struggling to access healthcare, and it’s predicted to get worse. Global shortages of doctors, nurses and midwives is expected to reach 10 million by 2030. For resource-limited nations, who must compete with high-income countries to train and retain skilled health professionals, this is a significant challenge to public health.

To keep skilled workers in country, policymakers in Africa and elsewhere have invested heavily in ‘return-of-service’ agreements. Individuals receive government funded health worker training, in exchange for a commitment to serve in an underserved community for a specified period of time.

For the first time, a UNSW-led retrospective cohort study has evaluated the effectiveness of return-of-service agreements in South Africa, Botswana, Eswatini and Lesotho.

Despite being used extensively, the effectiveness (where effectiveness relates to the retention of health workers) of these costly schemes in low- and middle-income countries has been until now unknown. It's vital that these schemes are successful in resource-limited nations.

“Interviews we’ve led with key policymakers running these schemes in South Africa, Botswana, Eswatini and Lesotho, suggested that the agreements have been hampered by poor planning and information systems, high defaulter rates, and a lack of reviews to understand the impact of these policies on society’s health and wellbeing,” says Dr Sikhumbuzo A. Mabunda, a researcher at the UNSW School of Population Health and The George Institute for Global Health, which is affiliated with UNSW.

The research team assessed the percentage of beneficiaries (individuals who received government-funded training) who fulfilled their contractual obligations, and those who remained in service beyond their contractual agreements.

The findings were published in BMJ Global Health Journal this week.

They reviewed data from the national governments of Eswatini, Lesotho and Botswana and South African state governments. They looked at a range of beneficiaries (doctors, pharmacists, dentists, physiotherapists, audiologists, speech therapists, occupational therapists and specialist doctors) supported between the years 2000 and 2010. They also requested information on beneficiaries’ work records up to March 2023.

The research was funded by UNSW, the National Heart Foundation and the National Health and Medical Research Council.

Defaulting on contracts is common

Despite the challenges of modest data systems, the research team were able to obtain data for 5,616 of an expected 14,000 beneficiaries. The high proportion of missing data could suggest that many beneficiaries are not being tracked.

The results did not find that return-of-service agreement schemes are effective in retaining healthcare professionals.

“Of the individuals with data available, only 19% had full information on their work history. Of those with full information, 67% defaulted their contracts and only 22% served their contracts,” says lead author Dr Mabunda.

Local internships have best outcomes

However, the research team identified a few factors associated with better outcomes.

“Beneficiaries who undertook their practical internship within the funding province were far more likely to be retained in service. We found that governments needed to fund six beneficiaries to have one beneficiary fulfilling their contractual agreements if they undertook internship outside their province,” says senior author of the study Associate Professor Rohina Joshi, a researcher at the School of Population Health and an affiliate of The George Institute for Global Health India.

“Unfortunately, only 32% of the 861 individuals who were supposed to do such an internship, undertook it within their province,” adds study co-author and researcher at The George Institute for Global Health in Sydney, Dr Blake Angell.

Current policy allows beneficiaries to undertake internship training outside their funding province. This study is the first to find that this practice is associated with higher defaulter rates.

Implications for policy makers

Dr Mabunda notes that the schemes have seen sustained investment over many decades and seemingly enjoy good political support, and that it is important to make the most of that goodwill. In the short term, Dr Mabunda believes that the governments should look to strengthen their technical capacities for managing these schemes, and strengthen information systems to capture the beneficiary life cycle. The schemes can be made more resilient through linking to research and evidence, implementing regular reviews, assessing impacts and retaining data.

Importantly, the study authors recommend that long term, internships should be undertaken within the funding jurisdiction, and governments must build internal capacity for training graduates they need within their own jurisdictions.

Variants of these schemes have been used in countries of all income levels with limited evidence that they are an effective mechanism to build a high-quality health workforce.

“Insights into their operation such as those we have generated through this study offer governments the chance to reflect on these programs and consider ways to improve their success in building an effective health workforce,” Dr Angell adds.

“Countries like Australia, who implement similar agreements, have a lot to learn from the experiences of these four countries to help improve the availability of health workers in regional areas.”