Bringing science and development together through news and analysis

  • BioMed Analysis: Honesty is the best policy

Shares

Maternal health advocates may feel financially threatened by falling death tolls but the data must not be hidden, says Priya Shetty.

When The Lancet broke the news this month that maternal health is improving across the world, public health experts should have been celebrating. After all, maternal health often languishes at the bottom of global health priorities.

But the journal also revealed that some women's health campaigners had put heavy pressure on it to delay publication of the data because they believed it might threaten their funding and undermine progress on global commitments to maternal health.

These fears are understandable, but misguided.

It is far more likely that the encouraging statistics will help the cause of maternal health advocates. Funding for healthcare today is more business-like than ever before — those seeking money must not only demonstrate need, but also have to show that the investment is yielding results, and the latest maternal health data does just that.

Rather, the pressure not to publish highlights the murky border between science and advocacy that plagues not only maternal health but other key areas such as HIV/AIDS.

Epidemiologist Elizabeth Pisani, currently a visiting fellow at DePaul University in Chicago, says that death has become a valuable commodity for people seeking funding to support their cause. "The causes of death have become a big political football. If you can say something is a cause of death then it is a big source of finance as well," she says.

Double counting deaths

As the amount of health statistics grows, the sources of those numbers seem to be becoming less independent. Nongovernmental organisations — whose funding is often directly related to how their cause is perceived by policymakers and the publicare increasingly becoming widely used sources of online news and data.

It is here, perhaps, that science could get buried under the weight of advocacy because these organisations will inevitably be biased in the way they present data.

For example, one death might be counted in several groups. HIV is a key example: a person who dies infected with both HIV and tuberculosis will count towards mortality measures of both diseases. If the person was pregnant or a new mother, she will also be counted in maternal mortality estimates.

Pisani spent many years working at the Joint UN Programme on HIV/AIDS (UNAIDS), which raises awareness and funding for HIV/AIDS.While the organisation never inflated the numbers, she says: "What we did do was cast the numbers in their worst possible light: every death of someone with HIV was chalked up as an HIV death even though they may have died of unrelated causes."

The vast sums of money put towards HIV funding — which has been disproportionate compared with diseases such as TB or malaria — has led some to feel that UNAIDS sometimes exaggerates HIV estimates to maximise funding.

But Peter Piot, the former executive director of UNAIDS, told SciDev.Net that he was "not aware of any attempts to hide estimates, nor to inflate figures. I have not seen that any of these accusations that UNAIDS has inflated HIV estimates are factually based."

Data quality challenge

Both advocates and scientists have the additional problem of ensuring that the data they obtain in resource-poor settings are robust.

Putting the principles of scientific rigour into practice can be difficult. Data on health problems in rural areas or conflict-ridden regions can be particularly patchy, either because of a shortage of trained professionals or difficulties in accessing remote populations.

Scientists and advocates must take care to communicate the uncertainty about their numbers. This is not easy, even for trained researchers, says Pisani. Commenting on The Lancet paper on maternal mortality, which is full of "incredibly complicated" statistics, she says: "Even those of us in the field have a difficult time parsing out how acceptable one assumption is over another."

When data on India's HIV epidemic finally emerged in 2006 it was revealed that 2–3 million people were living with the virus — a substantially lower figure than previous estimates by UNAIDS. The organisation immediately — and, as Pisani says, courageously — corrected its estimates and made the new figures public.

Good news about reducing disease does not have to mean bad news for finance. Instead, says Piot, funding for HIV/AIDS in India actually went up after the lower estimates were released. These funds had been allocated according to a long-term strategic plan, and the new data did not suggest that the HIV epidemic was of no concern anymore.

Advocating the truth

Policymakers are increasingly bombarded with data on diseases, but they must sift out trustworthy information with which to frame policies. Online publishing means that nongovernmental organisations and small publications can produce scientific-looking articles with ease. But health data that informs major decisions must still be subjected to intense scrutiny.

Journals such as The Lancet, with their numerous editorial checks and balances, have a key role to play in ensuring accuracy and trustworthiness through rigorous peer review, which often includes additional review by a statistician.

Advocates must realise that they can only win global support for their cause with an honest portrayal of the real picture. Anything less would be doing a major disservice to the millions of people whose health they are fighting for.

 

Journalist Priya Shetty specialises in developing world issues including health, climate change and human rights. She has worked as a news editor at New Scientist, assistant editor at The Lancet, and commissioning editor at SciDev.Net.

 

Republish
We encourage you to republish this article online and in print, it’s free under our creative commons attribution license, but please follow some simple guidelines:
  1. You have to credit our authors.
  2. You have to credit SciDev.Net — where possible include our logo with a link back to the original article.
  3. You can simply run the first few lines of the article and then add: “Read the full article on SciDev.Net” containing a link back to the original article.
  4. If you want to also take images published in this story you will need to confirm with the original source if you're licensed to use them.
  5. The easiest way to get the article on your site is to embed the code below.
For more information view our media page and republishing guidelines.