Relation between burden of disease and randomised evidence in sub-Saharan Africa: survey of research

Editor- When setting priorities for research, it is important to use
some form of rational decision making other than the scientific excellence
of projects alone. Recently, there have been calls within medical
journals to set research priorities according to burden of disease
rankings (1). The paper by Isaakidis and colleagues is the latest example
of this genre (2).

Certainly there is no consensus amongst health economists on this
issue. There is little evidence that burden of disease rankings constitute
the best way to allocate scarce clinical research funds, and it is even
arguable whether they should be used at all to set priorities for research
(3).

First, there is no agreement on which measure should be used. While
research funding bodies, such as the MRC in the UK, ask for measures of
burden to justify the importance of a research question, a number of
measures could be used (4). This lack of consensus can lead advocates to
choose the measure of burden that best suits their particular
constituency4. Isaakidis and colleagues use DALYs but do not justify why
this particular measure is superior to others in the context of priority
setting for research (2).

Second, it is arguable whether burden of disease rankings should be
used at all to set priorities for research. The value of conducting
clinical research ultimately lies in improving the information (reducing
the uncertainty) surrounding a decision between alternative interventions
(5). By focusing on diseases rather than on interventions, burden of
disease approaches fail to measure the value added by research. For
example, a disease could have a high burden but there may be little
question about which treatment to adopt because there may be no
alternative technologies for this disease. There would therefore be little
value in conducting further research.

At best, burden of disease rankings can suggest broad areas of
concern, but they cannot help inform specific priority setting in clinical
research. Care should be taken not to simply assume that rankings of
burden automatically translate into research priorities. Otherwise, it is
the populations that desperately need improved interventions that will
ultimately suffer from misdirected research efforts.

Rachael L. Fleurence, MRC DPhil student in health economics

David J. Torgerson, reader

Department of Health Sciences
University of York,
Heslington, York YO10 5DD

Correspondence to rls108@york.ac.uk

Reference List

1. Michaud CM, Murray CJ, Bloom BR. Burden of disease--implications
for future research. JAMA 2001;285:535-9.

2. Isaakidis P, Swingler GH, Pienaar E, Volmink J, Ioannidis JPA.
Relation between burden of disease and randomised evidence in sub-Saharan
Africa: survey of research. BMJ 2002;324:702.

3. Williams A. Calculating the global burden of disease: time for a
strategic reappraisal? Health Econ. 1999;8:1-8.

4. Gross CP, Anderson GF, Powe NR. The relation between funding by
the National Institutes of Health and the burden of disease. N.Engl.J.Med.
1999;340:1881-7.

5. Claxton K, Sculpher M, Drummond MF. A rational framework for
decision making by the national institute for clinical excellence. The
Lancet 2002;In press.