Predicting the Length of Stay among Healthcare Workers in Underserved Communities: A Quantitative Retrospective Cohort Study

Download Article

DOI: 10.21522/TIJPH.2013.05.04.Art005

Authors : Sangiwe Moyo


Background: While prior studies have identified a number of demographic factors related to general health practitioners’ decision to stay in public health practice, recruitment agencies have no validated methods to predict how long these health workers will commit to their placement. We aim to use machine learning methods to predict health professional’s length of practice in the rural public healthcare sector.

Methods: Recruitment and retention data from Africa Health Placements (n=13 698 with 1 838 completers) was used to development machine learning models to predict health workers’ length of practice. A cross-validation technique was used to validate the models, to evaluate which model performs better, based on their respective aggregated error rate of prediction. Length of stay was categorised into 4 groups (less than 1 year, less than 2 years, less than 3 years, and more than 3 years). Three machine learning models were trained and used 10-fold cross validation techniques to attain evaluative statistics.

Results: The three models attain almost identical results, with negligible difference in accuracy. The ‘best’-performing model (Multinomial logistic classifier) achieved a 47.34% [SD 1.63] while the decision tree model achieved an almost comparable 45.82% [SD 1.69]. The three models achieved the average AUC of approximately 0.66 suggesting sufficient predictive signal at the four categorical variables selected.

Conclusions: Machine learning models give us an effective tool to predict the recruited health workers’ length of practice. These models can be adapted beyond the scope of demographic information such as information about placement location and income. This modelling will also, allow strategic planning and optimization of public health care recruitment.

Key message

Human resource planning in healthcare can employ machine learning to effectively predict length of stay of recruited health workers who are stationed rural areas.


[1].     Alhassan RK, Spieker N, van Ostenberg P, Ogink A, Nketiah-Amponsah E, de Wit TFR. Association between health worker motivation and healthcare quality efforts in Ghana. Hum Resour Health. 2013; 11(1):37. doi: 10.1186/1478-4491-11-37.

[2].     Agyepong IA, Anafi P, Asiamah E, et al. Health worker (internal customer) satisfaction and motivation in the public sector in Ghana. Hum Resour Heal. 2012; 11(247). doi: 10.1186/1472-698X-12-25.

[3].     Buchan J, Couper ID, Tangcharoensathien V, et al. Early implementation of WHO recommendations for the retention of health workers in remote and rural areas. Bull World Health Organ. 2013; 91(11):834-840. doi:10.2471/BLT.13.119008. NDoH. National Health Insurance; 2017.

[4].     Bangdiwala IS, Fonn S, Okoye O, Tollman S. Workforce Resources for Health in Developing Countries. Public Heal Rev. 2010; 32(1):296-318.

[5].     Cometto G, Tulenko K, Muula AS, Krech R. Health Workforce Brain Drain: From Denouncing the Challenge to Solving the Problem. PLoS Med. 2013; 10(9). doi:10.1371/journal.pmed.1001514.

[6].     Dovlo D. The Brain Drain and Retention of Health Professionals in Africa. A case study Prep a Reg Train Conf Improv Tert Educ sub-Saharan Africa Things that Work. 2003:23–25.

[7].     Delobelle P, Rawlinson JL, Ntuli S, Malatsi I, Decock R, Depoorter AM. Job satisfaction and turnover intent of primary healthcare nurses in rural South Africa: A questionnaire survey. J Adv Nurs. 2011;67(2):371-383. doi:10.1111/j.1365-2648.2010.05496.x.

[8].     George G, Gow J, Bachoo S. Understanding the factors influencing health-worker employment decisions in South Africa. Hum Resour Health. 2013; 11(1):15. doi: 10.1186/1478-4491-11-15.

[9].     Hand DJ. A Simple Generalisation of the Area under the ROC Curve for Multiple Class Classification Problems. Mach Learn. 2001:171-186.

[10].  Habte, D., Dussault, G., Dovlo D. Challenges confronting the health workforce in Sub-Saharan Africa. World Hosp Heal Serv. 2004; 40(2):23-26.

[11].  Hatcher AM, Onah M, Kornik S, Peacocke J, Reid S. Placement, support, and retention of health professionals: national, cross-sectional findings from medical and dental community service officers in South Africa. Hum Resour Health. 2014; 12(1):12:14. Doi: 10.1186/1478-4491-12-14.

[12].  Kok MC, Dieleman M, Taegtmeyer M, et al. Which intervention design factors influence performance of community health workers in low- and middle-income countries? A systematic review. Health Policy Plan. 2014; 30(9):1207-1227. doi:10.1093/heapol/czu126.

[13].  Landis JR, Koch GG. An Application of Hierarchical Kappa-type Statistics in the Assessment of Majority Agreement among Multiple Observers. Biometrics. 1977; 33(2):363. doi: 10.2307/2529786.

[14].  Labonté R, Sanders D, Mathole T, et al. Health worker migration from South Africa: causes, consequences and policy responses. Hum Resour Health. 2015; 13(1):92. doi: 10.1186/s12960-015-0093-4.

[15].  Rosenthal EL, Brownstein JN, Rush CH, et al. Community health workers: part of the solution. Health Aff (Millwood). 2010; 29(7):1338-1342. doi:10.1377/hlthaff.2010.0081.

[16].  Steinmetz S, Vries DH de, Tijdens KG. Should I stay or should I go? The impact of working time and wages on retention in the health workforce. Hum Resour Health. 2014; 12(1):23. doi: 10.1186/1478-4491-12-23.

[17].  Sieleunou I. Health worker migration and universal health care in Sub-Saharan Africa. Pan Afr Med J. 2011; 10: 55.

[18].  Viscomi M, Larkins S, Sen Gupta T. Recruitment and retention of general practitioners in rural Canada and Australia: a review of the literature. Can J Rural Med. 2013; 18(1):13-24.