HIV NUMBERS: TRUE LIES: INDIA'S NEW AIDS GOOGLY

Halving the number of infected persons is a statistical trick. the prevalence may not have come down; the new and old estimates cannot be compared.
Mihir Srivastava reports:

Data from endemic, high prevalence states has been extrapolated to states like UP and Bihar to arrive at the figure

THE NEWS that India does not have the world's biggest HIV-positive caseload, and is less affected than South Africa and Nigeria, has produced its share of sighs of relief, but experts and activists are questioning the easy conclusions being drawn from the new data. While launching the third phase of the National AIDS Control Programme on July 6, 2007, the Minister for Health and Family Welfare, Dr Anbumani Ramadoss announced a dramatic fall in the estimated HIV+ population in India: 2 million to 3.1 million people infected with a prevalence level of about 0.36 percent, against the previous estimate of 5.7 million cases with a prevalence rate of 0.9 percent.

The new statistical methodology adopted this time to estimate HIV+ prevalence in India is said to have made all the difference. The new estimate is based on the National Family Health Survey (NFHS), which is a countrywide household survey covering about 2,00,000 people aged between 15 and 54, and was conducted through face-to-face interviews across India between December 2005 and August 2006. Also, the data from sentinel surveillance (prenatal clinics, sexually transmitted infection clinics and public hospitals) has been expanded to 1,122 sites from the earlier 703.

Ramadoss tried to score a point by saying: "We have always been faulted for underestimating the seriousness of the epidemic. That was a disturbing allegation as we were, and continue to be, very committed to containing and reversing the HIV AIDS epidemic." He, however, warned that the two sets of figures cannot be compared to show that there has been a steep decline. "These figures are not comparable. By using the same methodology for the past years that we used this year, there is only a marginal reduction in the prevalence," he clarified.

The implication is clear. There is no significant downfall in the HIV+ population. It is just a new estimate using a new method that is dubbed to be more reliable. "We should look at these figures with a pinch of salt," says Ashok K. Rau of the Freedom Foundation, a self-help group working with HIV+ people. "Different sets of people will arrive at different figures using the same data. It all depends on how it is treated statistically, on what assumptions are made. It is no surprise that the same data can be used to arrive at a higher figure than the last time. I will call it no more than another guesstimate."

Ashok Alexzander , the director of Bill and Melinda Gates Foundation's HIV/AIDS prevention initiative in India, Avahan, agrees that the fresh estimate indicates no trend. "The change in figure does not at all show a fall in the number of HIV+ cases. It's a new estimate with NFHS data, which is supposed to be more reliable. But it gives a clearer picture with southern states showing alarming figures with some hot spots in the northern states."

YOUR GUESS MY GUESS THEIR GUESS

TOTAL AFFECTED 5.7 million (Old estimate) 2.47 million (New estimate)

PREVALENCE RATE 0.9 % (Old estimate) 0.3 % (New estimate)


FUNDING IMPACT

The two figures are not comparable as different methodologies have been followed to arrive at them. The government is still to provide details of the new methodology used this time. For this reason the figures do not represent a decline, they are merely a new estimate that the government claims to be more reliable

NGOs are skeptical about the new estimate. They want more information on the new methodology adopted and demand aggregate data that has always being kept under wraps. They are apprehensive that this sudden fall in estimated numbers of HIV+ people will adversely affect the funding of AIDS mitigation efforts.

Anand Grover of the Lawyers' Collective says there might have been an over estimation of HIV+ population in India in the past, but questions the labelling of the new exercise as a more reliable estimate. "They say the earlier estimates were not so reliable, this one is more reliable. But nobody tells what was wrong with the earlier methodology. What changes have been made to arrive at this new figure? The primary data is never shared with NGOs, it is with agencies like NACO, UNAIDS or Bill and Melinda Gates Foundation," he says.

Rau points to the fact that data collected from endemic, high prevalence states has been extrapolated to other states to arrive at the figure. "It does not take into account states like UP and Bihar, which have huge populations and not much is known about them," he says.

There is still a paucity of reliable data on high-risk groups like sex workers and drug addicts using syringes. Says Vivek Diwan, lawyer and activist on HIV/AIDS issues, "There is little information on what is happening to vulnerable groups. This is crucial. It is important to provide these vulnerable groups with counselling. The level of literacy is shocking in these groups." Agrees Alexzander, "We are carrying out studies on high-risk groups that would provide vital information to design our mitigation response."

THE CHIEF of the National AIDS Control Organisation (NACO), Sujatha Rao, said the figure "has come down" and not merely been "re-estimated to be less", despite her minister Ramadoss saying that the two estimates were not comparable. "Even as the general prevalence is reported to have come down, the trends of HIV infection continue to remain the same. It's more in rural areas and amongst women," Rao said in a speech. Peter Ghys, in-charge of epidemic and impact monitoring at UNAIDS, seconded Sujatha Rao's claim. "We are today a lot more confident that what is being presented to you is closer to the true prevalence as it exists in the population," Ghys said.

There are concerns that the revised estimate increases chances of large pharmaceutical firms securing patents for essential AIDS drugs. "If more than one percent of population was HIV+ an epidemic would have been declared. This was very much on the cards with 0.9 percent of the population estimated to be HIV+. If the figure wasn't revised downwards it would have hampered the granting of patents on AIDS drugs," said a senior functionary in the Ministry of Health and Family Welfare.

NGOs say the new estimate will adversely affect the funding of Aids mitigation projects. "The reduced numbers will definitely mean reduced funding for AIDS-related projects that till now have been a major money spinner for NGOs," says Grover. "More numbers means more money," agrees Chandran. Alexander, however, doesn't see a decline ahead in support from international donor agencies. "I do not think so because the numbers are still very high. In fact, there is a greater need for funding in India. In per capita terms, India is getting less than countries like China and Cambodia," he said. Meanwhile, HIV+ patients remain, at best, confused.