Keeping track: Patients at a New Delhi AIDS centre

a fresh controversy has broken out over India’s HIV figures. The latest estimates of the number of people living with HIV in India released recently by the health ministry have evoked criticism from doctors and non-government organisations. Some in these circles have described the conclusions drawn from the estimates as “ridiculous”. Senior officials last week claimed that the estimates had been misinterpreted.

The nation’s health authorities themselves may be blamed for seeding the confusion and facilitating the misinterpretation of HIV surveillance data in India. In a statement released two weeks ago, the health ministry said that India had an estimated 28,000 additional infections during the year 2004, compared to 520,000 additional infections in the previous year. Last year, the National AIDS Control Organisation (NACO) had claimed that new infections had dropped from about 610,000 during 2002 to 520,000 in 2003. A mere 28,000 new infections during 2004 would have represented a dramatic decline in the number of new cases of HIV in India. But doctors and NGOs working with HIV-infected people find this abrupt drop hard to believe.

The truth is that the actual number of new HIV infections each year will forever remain hidden unless India begins to keep track of the people who die from AIDS each year. Estimates for new infections will remain widely off the mark without reliable data about AIDS mortality, because even as people die from AIDS, newly infected victims join the reservoir of people living with HIV.

It is surprising that nearly two decades after HIV surfaced in India, health authorities still have no idea of how many people die of AIDS in the country each year. Despite concerns expressed every now and then by public health experts, there has been little attempt to remedy the situation. The gap in AIDS mortality data has been explained as a problem in the death registration system. Doctors do not have to record primary causes of death.

Patients with AIDS tend to die from infections such as tuberculosis or diarrhoea, but most deaths are recorded as cardiac arrests. Death certification systems need to be changed. Doctors should list the primary and secondary causes of death. A patient who died from TB may also have been infected with HIV. In this case, the primary cause is AIDS, and TB is just the secondary cause.

A meeting of experts from UNAIDS, the health ministry, and NACO last week discussed various aspects of HIV surveillance, including strategies to improve AIDS mortality data. Special efforts to find out how many people are dying from AIDS in India each year are long overdue. Until the ministry has that information, its estimates for new infections should always be taken with a load of salt.