Global Alert and Response (GAR)

WHO Report on Global Surveillance of Epidemic-prone Infectious Diseases - Human Immunodeficiency Virus and Acquired Immune Deficiency Syndrome (HIV/AIDS)

Description of the data

The worldwide spread of HIV and the development of AIDS are being closely monitored worldwide. HIV surveillance is carried out to assess the seriousness of the situation, to monitor the rate of HIV spread (its incidence and prevalence), to increase awareness of the medical, social, economic, political impact of the disease and to promote effective planning and policy in relation to HIV/AIDS.

For most purposes precise data are not needed, as long as the general trends and the range or order of magnitude of the existing infection can be measured. However, for some purposes, such as measuring the impact of specific interventions, or for testing the efficacy of vaccines and treatments, precise data must be obtained.

Both data on the reported number of clinical AIDS cases and on seroprevalence of HIV infection are being collected. In assessing the seriousness of the AIDS pandemic, the level of HIV infection in a population is more informative than the number of people who have already progressed to AIDS. HIV infection is usually measured by sentinel seroprevalence studies, that is, the regular testing of selected groups of people for the presence of antibodies to HIV.

AIDS case surveillance

WHO has requested countries to submit regular reports on cases of AIDS since 1981 (Table 9.1). Updates on aggregated information by sex, age and presumed mode of transmission are gathered annually. The reported AIDS figures give a useful general overview but cannot be assumed to give an accurate or strictly comparable picture of the epidemic in different countries. While giving a general idea of the increase of AIDS in a population, the figures do not reflect the actual prevalence of AIDS disease so much as the accuracy of detection, diagnosis and reporting of the disease syndrome.

Nearly all countries have AIDS case-reporting systems in place, but the quality of the AIDS case reports varies significantly. The considerable variation in the percentage of AIDS cases that are reported to WHO from different countries reflects differences in the quality and extent of available services and testing facilities and the extent to which the population has access to and uses the facilities. Other main reasons for the variation in reporting between countries are the AIDS case definitions used, whether HIV testing is required or not for reporting, the availability and use of guidelines for diagnosing and reporting AIDS, the availability of HIV testing, the willingness and capacity to collect, compile and analyse the information and the regularity and completeness of reporting to WHO.

In spite of the limitations, information from reported AIDS cases is used in different ways. These include alerting countries to the presence of HIV in new areas or population groups, to assess the disease burden and AIDS-associated morbidity, to raise awareness and commitment, to provide information on the sociodemographic characteristics of the groups most affected, including sex ratios, age groups and main modes of transmission, and, in some situations, to estimate HIV prevalence and incidence through back-calculation.

Fig. 9.1 Reported cases of AIDS in industrialized countries (3)

Fig. 9.1 Reported cases of AIDS in industrialized countries (3)

HIV sentinel surveillance

The main epidemiological tool used to monitor trends in HIV infection prevalence in population groups is HIV sentinel surveillance. This is HIV screening of selected groups in the population, including those who are easily accessible, such as pregnant women (whose blood is routinely taken for other reasons) and people thought to be at high risk of HIV. This may include men who have sex with men, intravenous drug injectors, sex workers and people attending sexually transmitted infections (STI) clinics. Surveillance is usually repeated at the same sites at regular intervals (serial surveillance) to indicate how levels of infection are changing over time in specific areas and certain population groups. Levels of HIV in the wider population and among those at low risk are also important indicators of the parameters of the epidemic. Data may sometimes be obtained by screening blood donors, although their representativeness of the wider population is limited. Population-based studies, though complex and costly, can provide a better measure of the prevalence of HIV in the general population. However, the results of several population-based studies have shown that, in generalized epidemics, sentinel surveillance in pregnant women can be used as an indication (a proxy) for the prevalence in the adult, sexually-active population.

Strengths and weaknesses of the data

AIDS case surveillance

A number of factors need to be kept in mind when interpreting these data. In the first place, they come from surveillance systems of varying quality, and as such are subject to all the limitations of international comparisons. For example, the proportion of AIDS cases which are reported ranges widely, from less than 10% in some countries to almost 90% in others. In addition, countries use different AIDS case definitions. Stigma and discrimination associated with the disease may contribute to the reluctance in diagnosing and reporting AIDS cases. Next, the development of AIDS occurs fairly late in the natural history of the disease. For the most part, those who have developed AIDS in 1999 are those who were infected 5-10 years ago or even earlier. Thus, the AIDS data presented here reflect HIV transmission patterns that took place years ago. Also, there is considerable variation in the speed of progression from HIV to AIDS between children and adults. Very few of the children infected at birth survive beyond the age of five. Progression rates have also changed dramatically in industrialized countries, where the introduction of Highly Active Anti-Retroviral Therapy has contributed to decreases of up to 70% in the number of reported AIDS cases and related AIDS deaths (Fig. 9.1).

The term AIDS refers to the most severe clinical manifestations of infection with HIV. It includes a number of specific opportunist infections and/or associated diseases or cancers. People with AIDS usually die of associated illnesses like tuberculosis, chronic diarrhoea and wasting, pneumonia, meningitis, tumours or other infections that their immune system can no longer fight. The underlying cause of death, immunodeficiency caused by HIV, may not be recognized.

The AIDS figures given are reported numbers, not percentages of respective populations. This may lead to a biased perception of the seriousness of the epidemic in different populations unless relative population sizes are taken into account. Furthermore, AIDS figures only reflect the final, terminal stage of HIV infection, not the number of people who have the virus . The term "AIDS case" refers only to those people with full AIDS syndrome, that is, the final stage of HIV infection, who meet the national AIDS case definition. It does not include anyone with only mild symptoms of disease or those with HIV infection but no symptoms. Some people may even die of HIV-related problems without meeting the strict criteria for AIDS. Some claim that the term "AIDS case" itself is becoming less useful over time. Many medical practitioners are tending to use terms such as HIV or AIDS related illness instead.

HIV sentinel surveillance

HIV sentinel surveillance can provide more accurate indications of trends of HIV infection in the selected population groups and sites, particularly when conducted regularly at yearly intervals. HIV sentinel surveillance is a relatively simple and cheap epidemiological tool. Its flexibility and low cost make it feasible and sustainable even in resource poor settings. More than 10 years of experience have shown that, when conducted regularly, HIV sentinel surveillance can provide valuable information on the general trends in HIV prevalence in different population groups.

HIV sentinel surveillance may be conducted in existing health facilities (e.g. antenatal clinics (ANC), STI clinics) or in communities or sub-population groups, often through outreach work (e.g. sex workers, injecting drug users). Most surveys of HIV seroprevalence, particularly sentinel surveillance, are not based on a representative sample of the national populations but on convenient samples in selected sites that may at best represent only the specific population at the selected site. Therefore, while the use of unlinked anonymous testing methodologies can reduce the potential participation bias, selection biases remain a potential source of error. Other sources of bias or confounding surrounding HIV estimates based on limited convenient samples include:

Non-representative samples: Samples of convenience, may be used, e.g. hospital or ANC patients. These people may not be truly representative and have higher or lower levels of HIV than the general population or population sub-groups.
Geographic bias:
If more easily accessible populations are sampled, they may represent people at higher risk than those in less accessible areas where there is lower prevalence of HIV; where facilities are used for sentinel surveillance, only areas with functioning facilities and sufficient patient load can be included in the system.


(3)Source: UNAIDS/WHO working group on HIV/AIDS and STI surveillance.