Beyond the numbers

Although India has a high number of HIV-infected persons, the relatively low rate of HIV transmission leaves enough room to plan and fund effective strategies aimed at preventing a wildfire spread of the AIDS epidemic in the country.

JAYA SHREEDHAR

STATISTICS on few other diseases have generated as much confusion and controversy as those related to Acquired Immune Deficiency Syndrome (AIDS). In fact, as yet there is no reliable method to determine the number of people infected with the Human Immunodeficiency Virus (HIV), which causes AIDS. While there have been a few authentic epidemiological studies to extrapolate data on the extent of the AIDS epidemic in Asia, statistical mythologies about the magnitude of the epidemic continue to proliferate and these are endorsed by different schools of thought.

Why are figures so important? All prevention and control efforts are based on epidemiology, the science that elucidates the interplay of factors that determine the frequency and distribution of a disease. Policies, strategies and allocation of resources are linked to the epidemiologist's findings. Inadequate or questionable epidemiological studies result in lack of clarity and focus. Conflicting reports appear in the media, and these cause either panic or complacence - not a happy situation either way.

As the reality probably lies somewhere in between, the epidemiologist faces a daunting task. A major deterrent to free and unfettered data collection on AIDS and HIV transmission is the fact that studies on lifestyles and sexual behaviour patterns are not easy to formulate or execute. Human, technical and financial resources for such assignments are often cited as being inadequate. Such studies being donor-driven and donor-dependent, they often make jerky and uncertain progress. The final outcome would inevitably reflect these underlying realities.

THE joint World Health Organisation (WHO)-United Nations Program-me on HIV-AIDS (UNAIDS) Report on the Global HIV-AIDS Epidemic (June 1998) has estimated that there are about four million HIV-infected persons in India and the rate of HIV transmission in the adult population in the country is below 1 per cent. Even this low percentage makes India the country with the highest number of HIV-infected people in the world.

N. BALAJI
A recent study has estimated that there are 4,50,000 HIV-infected people among Tamil Nadu's adult population of 25 million.

Researchers who make estimations and projections for HIV-AIDS in India follow their own logic based on individual experiences and interpretations. Some of them apply the African experience to India in making statistical projections. This approach fails to take into account the difference between the African and Indian settings with regard to social and sexual behaviour, lifestyles, living standards, beliefs and customs. Projections based on experiences in the United States and Europe have similar defects in the Indian context.

Projections also vary depending on who makes them. Bio-statisticians and epidemiologists employ different norms while making estimations. Bureaucrats and politicians have their own ways of understanding and presenting epidemiological data before Parliament, the public and the media.

Projections for the entire country are made on estimations of the spread of HIV infection that are based on pilot studies conducted in pocket-sized locales. This approach has brought into question the fundamental ethics underlying epidemiological studies and the validity of consequent reportage. For instance, the prevalence of HIV among people exhibiting high-risk sexual behaviour is sometimes extrapolated to be representative of the entire populace.

Overwhelmed by the enormity of the challenge, the National AIDS Control Organisation (NACO) has worked hard to gauge the number of HIV-infected people in the country. Its attempts consist primarily of sero-sentinel surveillance through unlinked, anonymous HIV screenings in different subpopulation groups, such as pregnant women, patients attending sexually transmitted disease (STD) clinics, truck drivers and commercial sex workers. There is no denying that this is one of NACO's valuable data collection procedures, but it merely assesses to what extent and how fast HIV infection has spread within the groups studied. NACO has been compelled to extrapolate these figures and apply them to the whole population of the country, perhaps because of the absence of a better alternative or credible data. However, it does seem far-fetched to present the rate of prevalence of HIV infection within a subpopulation group such as truck drivers as that of the larger population.

Source: UNAIDS-WHO Report on the Global HIV-AIDS epidemic, June 1998.

NACO believes that HIV prevalence among pregnant women can be taken as being the most indicative of HIV prevalence in the general population. A common assumption is that pregnant women do not indulge in high-risk sexual behaviour and are probably infected by their spouses. Even within the small numbers of pregnant women who have been screened, NACO estimates the prevalence of HIV infection to be around 1.22 per cent as of end-June 1998.

Some researchers view this as a warning as it indicates that there are large numbers of HIV-infected women in the general population. Others say that the samples reflect only a small percentage of pregnant women, those who have access to the maternal and child health (MCH) services of government hospitals. There exist little data on the prevalence of HIV infection among the vast sections of the poor and disadvantaged pregnant women who do not avail themselves of MCH services. Nor do the data adequately reflect the infection rate among affluent women who utilise the services of private nursing homes.

Besides, the sero-sentinel surveillance data exclude people with HIV-related symptoms or full-blown AIDS, who are too ill to come to hospitals having the sentinel sites. They also do not convey the impact of the fair number of effective HIV-STD prevention programmes that have been under way in the last few years. While making projections, researchers should take into consideration the possible impact of intervention programmes on HIV transmission.

At no point has there been a system in place to collect information and technical data from medical institutions on HIV infection at the grassroots, leave alone analysis and interpretation of available data. The procedure of HIV sero-sentinel surveillance is taken up every six months and is limited to the collection of 400 blood samples from particular groups.

Can we then ever have an authentic picture of the situation? In 1987, the Global Program on AIDS developed Epimodel, a computer software programme to assess past and present rates of incidence of AIDS. The utility of Epimodel for individual countries was limited by the lack of essential data inputs, such as the starting point of the epidemic, the rate of its spread, statistics related to the progression from HIV infection to AIDS to death, from the countries concerned.

In 1994-95, as the requisite data were unavailable in individual countries, cases of HIV infection were simply totalled for each region and 10 individual regional models were constructed and used to prepare estimates of global HIV and AIDS incidence. At present, however, the sources of information on HIV, AIDS and STDs have vastly increased worldwide and their levels of reliability have improved. This has helped obtain a macro-view of the pandemic, one that may be relatively closer to reality. Epidemiologists say that large-scale research studies with good quality control and a satisfactory sampling framework are central to the process. An initiative has already begun in this direction.

The only Indian study to assess community prevalence of STD including HIV infection, was completed in 1997-98 in Tamil Nadu. Funded by the United States Agency for International Development (USAID), this large-scale effort was undertaken through the AIDS Prevention and Control Project (APAC) administered by the Voluntary Health Services, Chennai, in collaboration with the Post-Graduate Institute for Biomedical Sciences (Chennai), the Christian Medical College Hospital (Vellore), the Meenakshi Mission Hospital and Research Centre or MMHRC (Madurai) and the MGM or Mahatma Gandhi Medical College (Mumbai).

The third of only three such community prevalence studies of STDs-HIV in the world, the Tamil Nadu study was preceded by one in Rakai district of Uganda and another in the rural Mwanza region of Tanzania. The were intended to unearth quality data on the prevalence of common sexually transmitted infections in the community, their risk factors and the association between symptoms, STD syndromes and disease.

Three districts in Tamil Nadu were selected at random for sampling. The study estimates that out of the population of 2.5 crores in the age group of 15 to 45 years in the State, 4.5 lakh people are HIV-infected, about 24.25 lakh people have STD, and about 13.25 lakh people are infected with the Hepatitis-B virus.

APAC Project Director P. Krishnamurthy says that qualifying the data is critical for interpretations and conclusions. Although called a community prevalence study, it is important to understand that the information obtained from it applies only to a particular age group in the State and at a particular point in time.

Additionally, interpreting data also requires care. For example, there have been instances in which similar studies came up with conflicting results. The Mwanza study in Tanzania proved that diagnosing and treating STDs in the community could be correlated to a decrease in new HIV infections. The conclusion was that STD control measures played an important role in preventing HIV transmission. The Rakai study in Uganda, however, observed no such impact on HIV incidence although there was a stringent STD control programme, which in turn implied that there might be no correlation between STD control and HIV transmission.

The apparent conflict in the two sets of results baffled experts - until Dr. James Chin, renowned epidemiologist, explained it. Dr. Chin said that HIV prevalence in Rakai had exceeded the critical threshold of 5 per cent and as such was spreading on its own steam, so to say. In the circumstances, STD control measures would have little or no impact. In Mwanza, the HIV epidemic was at an early stage and STD control measures could exert a discernible effect on HIV transmission.

Such studies hold important lessons for India, where HIV prevalence among the general adult population at present is estimated to be less than 1 per cent, well below the critical threshold. The policy implication would be to fund well-planned STD control programmes generously and enable their vigorous implementation. Such a measure could exert a major effect on checking HIV transmission.

Source: UNAIDS-WHO Report on the Global HIV-AIDS epidemic, June 1998.

There are probably tremendous variations in the community prevalence of HIV across different States in India and the figure may be high or low depending on several factors. For instance, if HIV gained entry into an area relatively recently, it would be an early epidemic and would reflect a low prevalence. In another area there may be speedy HIV transmission owing to a great deal of high-risk sexual behaviour, wherein even an early epidemic may reflect a higher prevalence. The AIDS epidemic in India is in fact a patchwork of smaller epidemics of varying age and intensity, and these are in a state of constant flux.

Globally, the majority of HIV infections fall under the viral group HIV 1. A second group, HIV 2, which was originally thought to be confined to parts of Africa, has been increasingly detected in other parts of the world, including India. Several subtypes of HIV 1 have already been detected (Frontline, September 8, 1995). According to a report in the latest issue of Science, a study conducted by a research team led by Dr. Francois Simon of Bichat Hospital in Paris has identified a new strain of HIV, named Ybf30, which may not be detectable by the tests that are currently in use.

Such strains, which lie outside the range of known subtypes, have been found to surface occasionally. HIV blood tests will only look for subtypes that are known and outliers may therefore go undetected, contaminate blood supply systems and silently spread the infection in the community. This has always been a cause for concern. Such developments in the path of the evolution of the epidemic pose a challenge to the efforts of the epidemiologist to gather a complete picture of the HIV epidemic.

The controversy over the exact numbers of HIV-infected persons in India is likely to continue. As the mid-1998 WHO-UNAIDS report on the global HIV-AIDS epidemic acknowledges, the current estimates do not pretend to be an exact count of infections. They are constantly being revised, upwards and downwards, as countries improve their surveillance systems and collect more information.

The HIV epidemic, however, is here and spreading. It will continue to impact society and the economy. The degree of the impact may be debatable. The quest for precise numbers can only be answered by guesstimates, ranging from the realistic to the absurd. So colossal a human tragedy cannot, however, be confined to its numerical limits.

Table of Contents

Home | The Hindu | Business Line | Sportstar