Distinguished scientists, partners in health development, sister agencies within the United Nations system, staff, ladies and gentlemen,
Thank you for giving us your time and your expertise. This is an important meeting, and a challenging one.
You are here to look at different approaches to global health estimation and advise on ways to improve current estimation practices. This is what we are dealing with: estimates.
We all know the fundamental problem. Some 85 countries, representing 65% of the world’s population, do not have reliable systems for civil registration and vital statistics. Deaths are not recorded. Causes of death are not investigated.
In the countries where the disease burden is greatest, the data needed to monitor the health situation and assess trends simply do not exist.
The ideal solution, of course, is to strengthen the capacity of countries to generate, analyze, and use their own health statistics. This must be the urgent overarching priority, and it can be done.
In the meantime, WHO, the many experts and expert groups we consult, and academic institutions are forced to rely heavily on statistical modeling to fill the gaps. Analytical and estimation tools have become increasingly sophisticated, but uncertainties, to varying degrees, are inherent in all these methods.
In a situation that is far from ideal, we have yet another challenge. It arises from the fact that estimates are being produced by UN agencies and their expert groups, and by academic institutions.
In some cases, the estimates are similar. In other cases, the estimates differ substantially. The existence of multiple and significantly different estimates for the same indicator can cause confusion.
Let me be frank. We need to do a better job of examining the reasons for these differences and explaining them to the general public as well as to policy-makers.
The worldwide monitoring of the health situation, trends, and determinants is a core function of WHO. We take this function very seriously.
We will need to continue to rely on estimates to obtain comparable country data and provide global and regional trends, because the data gaps cannot be solved quickly.
We must combine this global estimation work with development of country capacity for estimates, such as the UNICEF-led work on child mortality estimates, and UNAIDS-led efforts for HIV/AIDS estimates. It can be done, but it takes substantial and sustained investment over a considerable time.
Ladies and gentlemen,
I convened this technical meeting to review current work on global health estimates and to gather your views on ways to improve these estimates. We are looking for your agreement on some basic standards for ensuring the transparency of data and the methods used to generate them.
We need to establish high-level principles on how to bring together estimates from different sources in a seamless, comparable and comprehensive way. Such principles can help resolve contradictions, which are causing confusion in Member States and fuelling unhealthy confrontation. As many of you know, I have been in the eye of the storm.
An overarching advisory process might be a consensus way to move forward.
You may see other problems and other areas where data and methods can be improved.
We must never forget why we go to the great trouble of producing global health estimates. We do so to assist countries and their development partners.
We do so to signal problems, pinpoint unmet needs, guide the selection of priority interventions and priority population groups, shape the allocation of resources, adjust strategies in line with results, and help individual countries benchmark their progress against that of their peers.
In short, we do so to improve the health outcomes of people. And we do our utmost to get the methods and the results right.
I thank you for your wise counsel as we undertake this work together.