PhD: Developing a needs-based resource allocation model for healthcare expenditure in Bangladesh

This PhD addressed the counterfactual question of what would have been the allocation to each district had the needs of the population been accounted for. Two alternative approaches were considered. The first used a simple capitation formula in which weights for the adjustment of the current allocation are generated directly based on the relative values of proxies for needs. The second approach predicted adjustment weights from the estimation of a standard econometric model of needs, controlling for a range of determinants including individual, household and district characteristics. Important predictors of current allocation were found to be the number of hospital beds and health workers rather than need factors. Important predictors of needs include demographic and socio-economic characteristics. The findings suggest that a needs-based allocation can be developed for Bangladesh. This research provides an alternative approach to generating weights showing systematic relationships between the need adjustment factors. The robustness of the methods used will be sensitive to the quality of the data and the assumptions of the models.

Outcome and Translation

As these approaches are based on sound economic analysis and are open to independent assessment, they will help to inform policy debate and can reduce the influence of politically motivated allocations. A gradual process of implementation and regular review of the methods used would be a way forward. Future areas of research may include: re-analysing data at smaller-area level and use of different components of allocations.

PhD Student: Zahid Quayyum

Supervisors: Dami Olajide (HERU), T Ensor, D Newlands (Economics, University of Aberdeen Business School (UABS)) and N Campbell (Division of Applied Health Sciences (DAHS), University of Aberdeen)


Quayyum, Z. (2012) ‘Developing a needs-based resource allocation model for health care expenditure in Bangladesh’, PhD Thesis, University of Aberdeen.