A key feature of the Global Fund is that it says it is "results-based" - that it can measure the impact that its grants have, and it can prove to its donors that their money is well-spent.
Recently, PEPFAR (the $15 billion US bilateral AIDS program) has had a difficult time coping with some confusions that its published numbers have caused. The experience provides a sobering lesson for the Fund.
On 1 February, GFO published an article pointing out that PEPFAR stated in January that it is "supporting" the treatment of 32,839 people in Botswana - which is exactly equal to the total number of people who were on antiretroviral treatment in Botswana from all forms of support, including those whose treatment was paid for by themselves or their employers. Shortly afterwards, PEPFAR reduced that Botswana number to 20,000.
Then on 1 July, the Washington Post ran a front page story on the same subject. The article quoted Botswana's deputy permanent secretary for health services as saying he couldn't identify a single person in Botswana who was receiving treatment as a direct result of PEPFAR financing.
Then last week, a public meeting was arranged in WashingtonDC so that PEPFAR could explain more about how its treatment numbers are computed.
The problem relates to two things: terminology, and how numbers are measured.
First, terminology. When launching PEPFAR, President Bush said that the program "will treat at least 2 million people with life-extending drugs." And until February 2005, PEPFAR said on multiple occasions that it would "provide" treatment to 2 million people worldwide. However, after February, PEPFAR changed its terminology and said only that it would "support" treatment for 2 million people.
On 13 June 2005, Ambassador Randall Tobias, US Global AIDS Coordinator, said "The goal set forth a year ago was to support treatment for more than 200,000 people in these 15 countries by June of this year; and so we didn't just exceed the goal by 35,000, we did it three months earlier than the goal. In my opinion, that is striking."
This is not really correct. The original goal was to "provide" treatment to that many people, not to "support" treatment to that many people.
Second, measurement. PEPFAR says that its definition of "support" of treatment includes contributions to a variety of activities up to and including curriculum development. But no cut-off levels are provided - PEPFAR does not state whether a $5 contribution to curriculum development would count as supporting all those receiving treatment in that country.
Mark Dybul, Deputy US Global AIDS Coordinator, speaking at last week's meeting in Washington DC, said that the US had always intended its grants to support (rather than directly provide) treatment, and that it has changed its terminology - not its actions - because what it said earlier was being mis-interpreted. And Ambassador Tobias and the Botswana Minister of Health wrote a letter to the Washington Post saying "The appropriate role for the U.S. plan was to support the [Botswana] national laboratory and training and quality assurance systems... Focusing on alleged squabbles about who should take credit for progress on AIDS in Botswana badly misses the mark."
Somehow, the Global Fund has thus far escaped the spotlight on this issue. It said last month that "220,000 people are now on AIDS treatment through programs supported by the Global Fund," but despite its normally impressive transparency, it has never provided any country-by-country breakdown or any explanation of how its treatment numbers are computed. When asked, the Fund simply informed GFO that it is working with PEPFAR and others to develop a common approach to how treatment numbers are calculated, and this will be used when results are published at the end of 2005.
There are two possible paths forward. The first option is for PEPFAR and the Global Fund to state something along the following lines: "We are both working hard in multiple countries on programs designed to support treatment, as are the governments of those countries, employers, and others. PEPFAR has spent $X on treatment-related work, the Fund has spent $Y, and other players have spent $Z. The collective result of everyone's work is that N people are on treatment."
The other option is for PEPFAR and the Fund to agree on and publish precise criteria for measuring how many people's treatment each of them is supporting, with a distinction being made between people for whom the agency in question has covered the great majority of the direct treatment costs (that is, people of whom the agency can say "We are treating these people"), and people for whom the agency is providing a significant and defined share of the direct or indirect treatment costs.
Without one or other of these options being pursued, the media and the public will continue to mis-understand Global Fund and PEPFAR statements about credit that each deserves for numbers of people put on treatment.
[Bernard Rivers (email@example.com) is Executive Director of Aidspan and Editor of its GFO.]