A new idea: find out what works

Michela Wrong | 18 Dec 2006
New Statesman

A link between Malaria and HIV has made ignorance too costly to tolerate

When I lived in Africa, watching expatriate friends go down with malaria became a regular event. Sometimes it was a slow slide - aching joints and 24-hour headaches leading to soaring fevers - sometimes a dramatic collapse. One friend was found crouched, semi-delirious, in her hallway, wearing an interesting combination of boxer shorts (she felt unbearably hot) and fleece mittens (she felt unbearably cold).

I never caught it, and couldn't help wondering if my smug good health had something to do with my dutiful pill-popping. "Did you, er, take the medication?" I would occasionally ask. "No point," they'd croak mournfully from their hospital beds. "Everyone knows it doesn't work."

A failure to link cause and effect seems to be a characteristic of the world of malaria. For decades, the organisations combating this disease, which kills more children and pregnant women in Africa than Aids, have indulged in a logical disconnect similar to that of my whey-faced friends. Which is why the December White House Summit on Malaria, part of a $1.2bn US presidential initiative to halve malaria deaths in 15 African countries by 2010, is interesting. For the first time, organisations winning Usaid contracts will be obliged to measure the effectiveness of their interventions.

Strange as it may seem, this marks a decisive break with the past. You might think any non-governmental organisation or private contractor working with an African government to combat malaria would be obsessed with the impact its programmes were having on prevalence. In fact, a programme's success is almost always measured by donor cheques received, numbers of health workers taken on, numbers of mosquito nets distributed and quantities of drugs delivered - what goes in, rather than what comes out, such as fewer cases and fewer deaths. If this sounds a clear case of getting your priorities back to front, there is some justification.

Sophisticated statistical analysis, which comes easily to rich western societies, poses an enormous challenge in run-down African countries. The donations, mosquito nets and treatments bought get measured because they are easy to quantify. A lot easier, in any case, than measuring prevalence, which involves taking hundreds of individual blood samples before an intervention, examining each sample for signs of the malarial parasite, and then returning to the area the following year to repeat the exercise. With the exception of wealthy South Africa, no country on the continent has gone through this fidgety, time-consuming exercise.

But there is something lazier and more insidious at work here, too. As William Easterly highlighted in his recent book The White Man's Burden, an absence of feedback, a disinclination to be held to account - failings that would not be tolerated in any successful company or government department - are the defining characteristics of today's aid industry.

Some of this can be traced to the nebulous utopianism at the sector's very root. Any organisation proclaiming "Make Poverty History" or "Working for a Fairer World" as its target is promising such sweeping, sometimes mutually contradictory things, that only a fool would take it to task for failing to deliver. When there are no benchmarks for what constitutes success, there will always be a danger of sexy token gestures triumphing over the quietly humdrum.

Agencies that run the health programmes in Africa know on whom their survival ultimately depends, and it is not the African malaria victim. It is the concerned western voter who wears plastic wristbands, supports several charities and wants to see her government increasing aid to the developing world. With this constituency, announcements of large donations and ambitious interventions push the "Something Must Be Done" button, while the scientific minutiae of prevalence surveys hold little meaning.

It doesn't take a genius to register the dangers of this skewed approach. If a health programme becomes the equivalent of throwing spaghetti at a wall, but never bothering to check which strands have stuck, taxpayers' money is bound to be wasted. Should nets be provided free or sold? What chemical to spray, how often, and where? Does teaching peasants why they should value a net and exactly how to use it matter more, in terms of delivering results, than the equipment?

"It's astonishing how little useful measurement goes on," says Roger Bate of the American Enterprise Institute, a conservative think-tank based in Washington. "Instead, the emphasis is always on how much aid is donated and spent. If the movie business worked on the same basis, we'd rate films exclusively on the cost of their special effects."*

The approach adopted at the White House summit breaks with this pattern. Organisations and countries will in future be obliged to conduct systematic prevalence studies or lose their Usaid contracts. This will not only yield valuable information about what works best; it will establish a model of greater accountability that other bodies, such as the Department for International Development, will come under pressure to adopt.

With the latest research indicating that HIV and malaria feed into each other in Africa, each rendering the other's impact more damaging, ignorance has become too costly to tolerate.

*"Usaid's health challenge: improving US foreign assistance" by Roger Bate, Journal of the Royal Society of Medicine, Volume 100. (Available here: http://www.rsmpress.co.uk/USAID.pdf)