Infectious disease, development, and climate change: a scenario analysis

We study the effects of development and climate change on infectious diseases in Sub-Saharan Africa. Infant mortality and infectious disease are closely related, but there are better data for the former. In an international cross-section, per capita income, literacy, and absolute poverty significantly affect infant mortality. We use scenarios of these three determinants and of climate change to project the future incidence of malaria, assuming it to change proportionally to infant mortality. Malaria deaths will first increase, because of population growth and climate change, but then fall, because of development. This pattern is robust to the choice of scenario, parameters, and starting conditions; and it holds for diarrhoea, schistosomiasis, and dengue fever as well. However, the timing and level of the mortality peak is very sensitive to assumptions. Climate change is important in the medium term, but dominated in the long term by development. As climate can only be changed with a substantial delay, development is the preferred strategy to reduce infectious diseases even if they are exacerbated by climate change. Development can, in particular, support the needed strengthening of disease control programs in the short run and thereby increase the capacity to cope with projected increases in infectious diseases over the medium to long term. This conclusion must, however, be viewed with caution, ∗Corresponding Author: R. Tol, ESRI, Whitaker Square, Sir John Rogerson’s Quay, Dublin 2, Ireland. email: [email protected]. Three anonymous referees had very helpful comments. This paper would not have been possible without the series of Climate Change Impacts and Integrated Assessment meetings at Snowmass, masterly organized by Susan Sweeney and John Weyant of the Energy Modelling Forum. Financial support by the European Commission DG Research (ENSEMBLES), the Hamburg University Innovation Fund, and the Princeton Environmental Institute is gratefully acknowledged. All errors and opinions are ours. 688 Richard S.J. Tol et al. because development, even of the sort envisioned in the underlying socio-economic scenarios, is by no means certain.