How to measure a country's HIV burden

The figure for HIV prevalence in India for 2004 looks encouraging -- an increase of only 28,000. But how has this figure been arrived at? And how accurate is it?

Every year the National AIDS Control Organisation (NACO) releases figures on India 's HIV burden. These figures are based on data from annual sentinel surveillance -- blood samples are taken from designated ‘sentinel' groups in every state and union territory in the country. The sentinel groups are pregnant women at government hospitals, clients visiting sexually transmitted diseases (STD) clinics, and groups with high risk behaviour such as female sex workers, men having sex with men, and injecting drug users. Most surveillance sites are in urban areas although a certain number of rural sentinel sites are also sampled to get an idea of HIV prevalence in rural areas. A specified number of samples is collected from each sentinel site (400 from the sites of pregnant women and 250 from others) and tested for HIV. Incidentally, India has the largest HIV sentinel surveillance programme in the world. In 2004, it was carried out at more than 650 sites throughout the country.

Sentinel surveillance is used along with other measures to look at trends in HIV prevalence. Information from various sources is triangulated -- surveillance data, number of AIDS cases reported, number of AIDS deaths reported, age-specific mortality, blood bank data, and size of population of groups with high risk behaviour. The behavioural surveillance survey is an excellent way of monitoring risk behaviour. The first national behavioural surveillance survey was conducted in 2001. A second round is being planned.

Sentinel surveillance is a good tool for noting trends in HIV prevalence, and changes over the years. It is a costly and labour-intensive exercise (over Rs 2-3 crore is spent every year), is well carried out and generally well supervised. External quality assurance in the testing of samples is also done.

The problem arises when sentinel surveillance data is used to estimate a country's HIV burden. Sentinel surveillance is not designed for this.

Estimating the HIV burden in a low prevalence area

For one, HIV is not very prevalent among the general population. The chances that a sample survey will be accurate depend partly on the prevalence of the condition. The lower the prevalence, the higher the minimum sample needed. Also, sampling biases are worsened when the condition has a low prevalence.

And there are various biases in the existing sampling process. For example, practically all sentinel sites are in government hospitals, whereas the majority of people use private services. We don't know the HIV prevalence among those who attend private hospitals. Estimates of the overall HIV burden are mainly based on prevalence among pregnant women attending government hospitals. This excludes those who go to private hospitals for antenatal care -- and those who don't receive any healthcare at all.

Further, the samples are of pregnant women and various groups with risk behaviour. They offer no direct information on other women or on men outside these groups. Finally, samples taken from STD clinics are intrinsically biased -- they are taken from people with symptoms of a sexually transmitted disease who attend government STD clinics for treatment.

In addition, if the condition is unevenly distributed in the population any sample taken will not be representative of this population. Representative samples are necessary in order to make projections or estimates, or else the results will be unreliable. To illustrate, each state provides 400 samples each for the annual surveillance, from several antenatal clinics. Just two or three positive samples among them could skew the overall results. In Uttar Pradesh, in 2003, at least eight of the 17 antenatal clinics did not have a single positive sample.

By contrast, in South Africa , the antenatal prevalence in 2003 was 28%. If they had used the same system as ours they would have had 112 positive samples out of 400 samples at a single site.

The only way to get an accurate picture of the HIV burden is through a ‘head count', which obviously is not possible. So one has to be satisfied with the limitations of using sentinel surveillance data.

Assumptions behind the NACO algorithm

When NACO mentions an increase of 28,000 in the last year, it is referring to an estimated number based on a calculation. When such estimates are made, the policy is to look at trends over time rather than the value in a particular year, in assessing the rate of growth of the epidemic. This is because since these figures are projected estimates based on sample surveys, sampling errors affect the final estimate.


If one looks at the rate of increase of HIV burden, in 2002 there was an increase of 6.1 lakh over 2001. In 2003 there was an increase of 5.26 lakh over 2002. In 2004 there is an increase of only 28,000. It is quite possible this represents a true slowing of the epidemic due to certain preventive efforts paying off. But one can be really certain only if a similar slowing is seen in subsequent years as well. We don't have to wait very long to find out. The preliminary results of this year's surveillance round should be available by the end of November.

NACO uses an algorithm to calculate the HIV burden using sentinel surveillance data. The algorithm uses the following information: HIV prevalence among pregnant women, prevalence among STD clinic attendees, percentage of men and women, between the ages of 15 and 49, in urban and rural areas, ratio of HIV prevalence among men and women, ratio of HIV prevalence in the urban population to that of the rural population, etc. This requires using certain assumptions that could result in large margins of error in the final result. From time to time NACO also makes changes in the algorithm taking new evidence into account to help make a better estimate.

These official calculations by NACO are made by a select team of experts including people from the World Health Organisation (WHO) and UNAIDS.

Only NACO does surveillance in India on a national level. All the other estimates being discussed by various individuals and organisations today are projections based on NACO's sentinel surveillance data. Using different assumptions and algorithms result in different estimates.

Actually it matters little if we have more HIV cases than South Africa . Comparisons should be based on proportions -- per capita or a fraction of the population -- not absolute numbers. South Africa has an adult HIV prevalence of about 21.5% and an estimated HIV burden of 53 lakh. In India , 0.91% of adults have HIV, which is a relatively low figure; our estimated burden is 51.3 lakh. India 's huge population means that even an increase of 0.1% in prevalence will add 5 lakh more infected people. So this controversy over whether or not we have overtaken South Africa , which has an HIV burden of 53 lakh, is pointless.

In India , which has an overall low HIV prevalence and a non-uniform spread, when we make projections with only 400 samples from each site there are bound to be uncertainties in the final estimate. No one can be sure about the margin of error. But since 1998, when sentinel surveillance was first done on a national level, a number of states have registered only marginal increases in HIV prevalence. Many large states such as Uttar Pradesh, Madhya Pradesh, Bihar , etc, have only 0%-0.25% prevalence levels. This indicates that HIV is not spreading as rapidly as we once thought it would in our country.

In any case, why be sensitive about HIV estimates? A large estimate does not necessarily mean that the AIDS control programme is failing.

The picture today

I was able to take a look at a considerable portion of the surveillance data for 2004. There is surprisingly little increase in prevalence. Many states have remained at the same levels for years. Among the high prevalence states, Andhra Pradesh has worsened. Karnataka and Maharashtra also have major problems. Tamil Nadu, which has been hovering at a prevalence level of 0.7% among pregnant women for the last three years, should no longer be considered a high prevalence state.

How do we reconcile this with the opinions of civil society organisations that claim that the country's HIV burden is much higher than that claimed by NACO?

Civil society organisations generally see only a small part of the whole picture. They tend to see people who are symptomatic or have AIDS. The number of symptomatic people and people with AIDS is certainly increasing, since those infected years ago are now developing symptoms/AIDS. There is no doubt that doctors and civil society organisations are seeing more people who need care. But a spate of AIDS cases does not mean an absolute increase in the number of people infected. All it means is that the epidemic is becoming more visible now since the proportion of symptomatic patients has increased.

There is another factor as well. AIDS is now a big industry. Civil society organisations (and states too) become uncomfortable when surveillance and other data show that the HIV epidemic is not increasing as rapidly as was expected. This means that expected levels of funding will not materialise. So it is in their best interests to propagate a high HIV level myth.

The quality of surveillance is improving every year. Also, the algorithm for making estimates has slowly been modified based on improved information. The algorithm used for 2003 was better than the one used in 1998. Both are available on NACO's website.

With an effective anti-retroviral therapy (ART) programme the number of people living with HIV will increase. Therefore the burden will go up instead of coming down. But it would be inaccurate to say that the ART programme has resulted in reducing HIV prevalence. The ART programme was started only on April 1, 2004 , while the annual sentinel surveillance got over by September 30, 2004 . By this time only a few hundred people in the country had received ART and so the programme cannot be said to have had any effect on the HIV burden as estimated for 2004.

InfoChange News & Features