Trafficking, sex work, and HIV: efforts to resolve conflicts.

94 www.thelancet.com Vol 385 January 10, 2015 and Evaluation responded to criticisms and improved its practices. The new GBD 2013 estimates show the vast expansion of collaborators, acquiring primary information even from countries that traditionally provide little data, such as China, Mexico, Turkey, and Russia. The description of methods is more detailed and transparent, and the instruments of analysis— many of which are now available online—are becoming increasingly sophisticated. Particular care was taken to address the key weaknesses, correcting questionable estimates from GBD 2010. Unfortunately, the GBD 2013 estimates fall short of allowing full independent replication of all results. A comparison of the GBD 2013 estimates to those of WHO and its affi liates suggests that we should expect the grand convergence between two sets of estimates to begin this year, with the global causes of child deaths being a prime example. The remaining diff erences are, in fact, useful because they point to the most important data gaps or the most controversial sources of data used. They will help focus subsequent debate on an increasingly specifi c set of questions. Therefore, the competition between WHO and the GBD has benefi ted the entire global health community, leading to converging estimates of the global causes of death that everyone can trust. The GBD initiative has emerged as a well-organised and rapidly growing collaboration that is now seriously challenging WHO’s role in generating global health estimates. WHO maintains its position for several key strengths, for which it benefi ts from collaboration with affi liated groups of external academics. However, it will need to rethink its own role and massively scale up its capacity to generate global health estimates to remain competitive. WHO’s indecision over investment in global health metrics, or over the role it should have in the long term, will help the GBD collaboration to gain widespread support for its estimates. Idleness by WHO might even lead to a new monopoly in global health metrics, with the centre of activity moving from Geneva to Seattle. Such a scenario might again simplify global health politics but, in the absence of healthy competition, science could be the poorer.