Burden of illness estimates for priority setting: a debate revisited.

This paper returns to the debate in this journal about a decade ago on the value of cost of illness (COI) and burden of illness (BOI) estimates in priority setting. Concern is expressed that there has been a resurgence of interest in calculating and using BOI estimates in such priority setting. It is especially concerning that this interest seems to have support from both the World Bank and the World Health Organisation (WHO) (although perhaps less so recently from the latter). It is argued that in terms of priorities for health services, BOI calculations are irrelevant except possibly in the context of some (less than ideal) concept of need in support of equity. If the need basis for equity is set in terms of 'capacity to benefit', then BOI calculations become even less relevant. There is an argument for some research funding being prioritised in terms of BOI but only when it is genuinely the case that there is total ignorance, beyond the size of the problem, about a particular policy or disease area. Such a level of ignorance will happen very seldom and then some fairly approximate estimates of BOI will suffice. It is better to concentrate in priority setting on estimating the costs and benefits of marginal changes than devoting scarce analytical resources to superfluous estimates of BOI.