Some of your criticisms of child-mortality figures from the Millennium Villages project in Africa are unjustified (Nature 485, 147; 2012).
You question the methods we used to generate mortality estimates, which were based on mothers' recall of the numbers of births and deaths of their children (P. M. Pronyk et al. Lancet http://dx.doi.org/10.1016/s0140-6736(12)60207-4; 2012). But this is standard practice, and was approved by the paper's reviewers. To avoid any potential systematic bias, we used the same technique and reporting period for the Millennium and comparison villages.
No “alarm bells” sounded over the mortality rate in the comparison villages. These were closely matched to the Millennium sites, chosen because they were poor, often remote, hunger hotspots. We therefore had no expectation that mortality rates at any of these sites would track national trends.
You wrongly suggest that the costs of the project were not properly reported. Our paper presents costs by site, sector and funding source — providing much more information than in most comparable studies. Detailing intervention costs and spending by partners is a core purpose of the project, executed with scrupulous care.
Our results on the decline in child mortality in the Millennium Villages are statistically significant over time and, in relation to the comparison villages, follow protocol. As you point out, the comparisons made with national trends were not tested for statistical significance, which is why they have been withdrawn (P. Pronyk Lancet http://dx.doi.org/10.1016/S0140-6736(12)60824-1; 2012).
Since then, the project has been piloting techniques for assessing changes in the under-5 mortality rate in the Millennium Villages using trained community-health workers to collect detailed, accurate birth and death data. Leading experts on public health will be invited to form an independent committee to review and help strengthen the collection and processing of these data.