Should the GBD risk factor rankings be used to guide policy?

2060 www.thelancet.com Vol 380 December 15/22/29, 2012 The Global Burden of Disease Study 2010 (GBD 2010) estimates provide important new insights. Alongside estimates of health burden attributable to 291 diseases and injuries, Stephen Lim and colleagues estimate the health burden associated with 67 risk factors, organised into a hierarchy of clusters. So as to distinguish real changes in global burden and risk factors from changes in methods, they not only estimated the burden and risk factor ranking for 2010, but also recalculated estimates for 1990. This study represents the work of several expert working groups, who led systematic reviews of the health eff ects and prevalence of each risk factor. In such a complex and ambitious exercise, trade-off s between rigour and policy relevance are inevitable. Judgment calls have to be made when data are not reliable or consistent, and these will sometimes be conten tious. In the long term, the work’s value will depend on whether the fi ndings are internally consistent, complete, and supported by scientifi c consensus. Although many of the rankings of disease burden and risk factors are internally consistent, discrepancies exist because of the incompleteness of risk factors analysed. For example, diarrhoea and HIV/AIDS are leading causes of global disability-adjusted life years (DALYs), but their associated risk factors do not feature strongly (fi gure). For diarrhoea, in 2010 the associated risk factors of sanitation and unsafe water only ranked 26 and 33, respectively, and estimates for poor hygiene were not included. For HIV, unsafe sex was not included as a risk factor, by contrast with the previous Global Burden of Disease analysis (GBD 2006). More generally, the 1990–2010 comparison of risk factors suggests that alcohol, tobacco smoking, and several dietary factors have moved up the rankings, whereas others, such as being underweight, suboptimal breastfeeding, poor sanitation, vitamin A defi ciency, zinc defi ciency, and unsafe water, have decreased in importance. These changes portray real demographic Should the GBD risk factor rankings be used to guide policy?