Sumita Kale: AIDS - down but not out

India's first AIDS patient was diagnosed exactly 20 years ago in Chennai. The National AIDS Control Organisation (NACO) currently estimates that 5.21 million Indians are afflicted with this disease. Yet uncertainty prevails about the exact number of cases in the country, with the World Bank noting in 2004 that it was "unfortunate that these estimates are unreliable in India today". NACO estimates for 2004 showed a sharp drop of more than 90 per cent in the number of fresh cases and their latest data for 2005 confirms the low growth trend.

Many NGOs involved with AIDS believe that the government is in a state of denial over the actual extent of this disease. Since NACO estimates are based on samples covering essentially Antenatal and sexually transmitted diseases clinics, patients who are outside the system would automatically be excluded, especially in backward areas. While NACO does adjust for under-reporting and has been increasing the number of surveillance sites, it retaliated to the growing criticism by convening a meeting with technical experts in June 2005. The conclusion: "the estimation process...follows the internationally recommended and recognised methodology. This process gives estimates and makes no efforts to count the exact number of infected persons, but gives the figure that can be used to demonstrate trend from year to year." It was felt that increasing accuracy would not justify the cost nor the efforts, as the figures were unlikely to significantly impact policy and planning strategies.

For the government then, the exact number of cases is not as crucial as the trend in growth. Turning to independent sources to corroborate/ contradict the trend shown in official data, a study published in Lancet (April 2006) yields some support for the falling trend in new cases. Using the prevalence in young women aged 15-24 years as an indirect proxy for the incidence of new infections, Kumar et al found the prevalence in southern states had registered a significant decline over the period 2000 to 2004. A modest reduction was also seen in these states in men aged 20-29 years. Northern states, with a fifth of the prevalence present in the south, however, showed no significant increase or decrease in the period of study.

The regional dimension has implications for future growth trends. Though all high-prevalence states "" Tamil Nadu, Manipur, Nagaland, Maharashtra and so on "" have been much more committed to raising awareness, the movement of workers from states like Bihar, Chattisgarh and Rajasthan, where prevalence and awareness of the disease are low, has potential to create problems. Again, low-prevalence states cannot afford to be complacent as they too harbour districts identified as high-prevalence.

NACO estimates also reveal an increasing trend of HIV infection among injected drug users in Delhi, Manipur, Nagaland, Tripura and Assam. The situation in the north-east could be more critical than official data shows (the Lancet study does not cover these states). Here, the impact of the opium triangle and poor access to general health services make resolution of the problem even harder.

The policy focus at present is on targeting high-prevalence districts and reinforcing awareness with behavioural and attitudinal change. Though Tamil Nadu, where prevalence has dropped significantly, has shown the way, unless all states show similar commitment, the glimmer of hope in the official data could quite as easily be termed just a mirage.

The writer is advisor, Indicus Analytics