Global IDEA

We applaud the recent informal assessment of the potential health and economic benefits that might flow from an accelerated program of research to combat global infectious diseases.1 The policies adopted for the organization and prioritization of health research should themselves, wherever possible, be evidence-based.2 However, many conceptual and empirical obstacles face those attempting retrospective analyses of economic benefits from specific programs of health research; in particular, it is necessary to take full account of the cost of the contributing research and of its application, and to then assess the value of the incremental health and other benefits that follow.3

The proposal from the Global Infectious Disease Evidence and Analyses (Global IDEA) network1 refers to work from the United States on the economic benefits of health research4 that is rightly receiving considerable attention. In assessing returns in the context of the global debate about infectious diseases, however, one obvious problem is the value placed on the health gain. The US study valued the life of a US citizen at about $3 million but even if that is the appropriate value to use in the US context, it is improbable that such a figure would be applied rationally by decision-makers in other countries.

The Global IDEA Scientific Advisory Committee also argues that if $2 billion is spent over 10 years for research on new tools that lead to a 5% increase in lives saved, this could, using figures from the important report from the Commission on Macroeconomics and Health,5 result in annual returns of about $9 billion. We suggest that, in estimating the real return, it is essential to allow for the (possibly very substantial) costs of applying any new tools or technologies that result from the research.3

More work is needed to refine the methods for analyzing the payback from investments in health research. Initiatives such as the program proposed by Global IDEA might then be supported with firmer evidence of their possible benefit.