The Pattern of AIDS Reporting and the Implications on HIV Surveillance (extract from the Public Health and Epidemiology Bulletin Volume 11, Number 4, Aug 2002)

Dr Kelvin Low ¹ Dr S S Lee ²

¹ Senior Medical & Health Officer
² Consultant (Special Preventive Programme)

Introduction

Acquired Immune Deficiency Syndrome (AIDS) did not announce itself but appeared rather inconspicuously. The June 5, 1981 issue of the Morbidity and Mortality Weekly Report (MMWR) first alerted the medical community about five cases of Pneumocystis carinii pneumonia (PCP) diagnosed in previously healthy young homosexual men within a six month period between late 1980 and early 1981 in Los Angeles, California[1]. The occurrence of this condition in young men with no apparent causes of immunodeficiency was unusual, and had prompted further investigation on the links between the infection and the lifestyles unique to the infected individuals. Human immunodeficiency virus (HIV) was later discovered to be the cause of AIDS. The infection has since led to one of the major epidemics in the ensuing two decades. An estimated 40 million people were living with the infection as of the end of 2001[2].

Most countries had used AIDS surveillance as their mainstay of monitoring mechanism for HIV epidemiology, especially during the first few years before the causative virus was identified. Serological tests have become widely available after 1985, enabling HIV surveillance to be undertaken as a regular public health programme in many countries. In this article, we take readers to a journey of how AIDS was defined in the past and how this has affected Hong Kong, a population with a relatively low HIV prevalence. The epidemiology of reported AIDS cases will be discussed together with the strengths and weaknesses of our current AIDS surveillance system.

Defining "AIDS" Epidemiologically

AIDS represents the late stage of the course of HIV disease characterized by severe immunodeficiency. The first case of AIDS in Hong Kong was reported in 1985 when a patient fell ill and was admitted to a hospital where he died shortly. The syndrome fitted the definition proposed by the Centers for Disease Control and Prevention (CDC) in the United States which described AIDS as a disease "at least moderately predictive of a defect in cell-mediated immunity, occurring in a person with no known cause for diminished resistance"[3].

The original CDC definition underwent refinement in the following years, which emphasized on specific "indicator" diseases, while the laboratory evidence of infection was not mandated[4]. The CDC 1987 definition was subsequently adopted by the World Health Organization, other developed countries and Hong Kong[5].

The wide access to CD4 (T-helper cells) monitoring and HIV testing provided the foundation for an even more specific definition for AIDS in the western world. In 1993, CDC revised the criteria by mandating the laboratory diagnosis of HIV infection, the inclusion of indicator diseases (which reflect the underlying immune deficiency) and/or a low CD4 level of less than 200/ml for AIDS surveillance[6]. In Hong Kong, AIDS surveillance case definition for adults and adolescents was drawn up in 1995 by the Scientific Committee on AIDS^*. The CDC 1993 definition was adopted with the following modifications: (a) disseminated penicilliosis (a fungal infection common in South East Asia) was included in the list of AIDS defining conditions, (b) pulmonary or cervical lymph node tuberculosis was counted only if CD4< 200/ml, (c) a low CD4 (< 200/ml) alone was not considered as an AIDS defining condition. (Table 1)

Table 1 AIDS Surveillance Definition for Adults and Adolescents in Hong Kong

The Trends of AIDS Reporting in Hong Kong

In Hong Kong a voluntary HIV/AIDS reporting system has been in place since 1984, comprising the reporting of laboratory diagnosed cases of HIV infection meeting the AIDS surveillance definition as proposed by the Scientific Committee on AIDS. It is a dual system involving the participation of laboratories and physicians, the latter submitting the reports using a standard form⁺. The system is operated by the Special Preventive Programme of the Department of Health (DH).

As of the end of 2001, a cumulative total of 1 755 HIV infected cases have been reported, of whom 560 were known to have progressed to AIDS. The number of reported AIDS cases rose steadily in the first decade, from less than ten per year in the first few years to 70 per year in 1996, then levelled off since 1997. (Figure 1) In the past five years the ratio of reported HIV infection and AIDS had remained stable. For every three to four cases of HIV reported, one AIDS case was diagnosed.

The proportion of Chinese AIDS cases had increased from 33.3% in 1987 to 81.7% in 2001. Cumulatively 78% of all reported AIDS cases (436/560) were Chinese. The overall male-to-female ratio was 7.1:1, which reflected a gradual narrowing of the ratio from 59:1 between 1985 and 1991, to 8.7:1 in 1992 to 1996, and 5.4:1 in 1997 and beyond. The gap between male and female AIDS cases was getting closer as the proportion of female AIDS cases increased in the past few years, from 7.1% in 1996 to 20% in 2001. (Table 2) The ethnicity pattern of reported AIDS cases was not the same between males and females. Whereas majority of female AIDS cases were non-Chinese (66.7%), non-Chinese only accounted for 15.9% of male cases. The majority (98.2%) of AIDS cases were adults, a phenomenon similar to the reported HIV infection (98% adults). This is not surprising, as majority of the reported AIDS cases had resulted from infections through sexual contact. Overall, 65.9% (369/560) and 22.3% (125/560) contracted the virus through heterosexual route and in men having history of sex with men (MSM), respectively. While the number of heterosexual infection in men with reported AIDS had increased over the years, MSM and bisexual men continued to account for a significant proportion in recent years. This latter pattern was seen also in reported HIV infection. The increase in heterosexual men with AIDS was lagging behind those reported with infection. Paediatric AIDS was uncommon, and was either related to blood or blood product transfusion (4 cases) or perinatal infection (6 cases).

Table 2 Characteristics of AIDS Cases Reported in Hong Kong, 1985 - 2001

^* Ratio not shown because no female cases were reported during that period.
⁺ Diagnosis of AIDS within 3 months of HIV detection.
^--- Not applicable because no AIDS cases were reported in that year.

Primary AIDS Defining Illnesses in Hong Kong

Primary AIDS defining illness (ADI) is the qualifying condition that leads to the classification of a case of HIV infection as AIDS. The pattern naturally depends on the criteria adopted and the diagnostic accuracy of these conditions in the health care system. PCP has remained the single most important ADI over the years. In 2001, PCP accounted for 43.3% of the ADIs (cumulatively, 38.6% of all ADIs). This was followed by tuberculosis (28.3%) and then fungal infections (10.0%). Cumulatively, they accounted for 22.5% and 18.4% of all ADIs respectively. (Table 3)

Pneumocystis carinii is an ubiquitous organism which causes disease in individuals who are immunocompromised. The risk of PCP increases with the fall in CD4 counts. The regular use of prophylactic agents when an HIV infected patient's CD4 count falls below 200/ml has been proven to be effective in preventing PCP. PCP is more common in persons not previously known to be HIV infected, and in those without prior HIV care[7]. In Hong Kong, a higher proportion of reported AIDS with PCP as ADI was diagnosed concurrently with HIV (defined as the reporting of AIDS within three months of the HIV diagnosis), compared to the non-PCP AIDS cases (88.4% vs 71.8%, p-value < 0.001). The number of PCP cases reported per year had risen steadily in the first decade but had been fluctuating in the past five years. (Figure 2a)

Unlike PCP and fungal infections, tuberculosis (TB) occurs when one's immunity is still relatively preserved. The causative agent, Mycobacterium tuberculosis, is a prevalent pathogen in Hong Kong, a phenomenon which has been here long before the AIDS era. In 2001, 7 262 cases of TB were reported to the DH under a statutory system. The yearly notification rate remained high at more than 100 per 100 000 population in the last decade[8]. Almost all cases were unrelated to HIV, and the very low prevalence of HIV made it an insignificant contributory factor to the high rate of TB in the background. The upsurge of TB as an ADI since 1996 may in part be a result of the change in AIDS case definition, when pulmonary and cervical lymph node TB with CD4 lower than 200/ml became included as ADI upon recommendations of the Scientific Committee on AIDS in 1995. (Figure 2b)

Finally, the occurrence of fungal infection heralds the stage of severe immunodeficiency. Penicillium marneffei is a dimorphic fungus that is found largely in South East Asia, including Hong Kong. Its inclusion as an ADI in Hong Kong reminds us of the importance of understanding the geography of microbes in the development of control strategy. A fluctuating number of cases had been reported since 1990. (Figure 2c) The diagnosis of penicilliosis is dependent on a high index of suspicion and the skilful detection by microbiologists.

Discussion

About twenty years ago, AIDS reporting was introduced as a surveillance mechanism to track the epidemiology of an unknown infectious agent. This was deemed necessary before HIV was discovered, and before a reliable objective test was invented. As an epidemiological tool, AIDS reporting suffers many drawbacks including insensitivity, delay in reporting, operator dependence and discrepancies in the condition's definition. Despite these inadequacies, AIDS reporting has informed us the morbidity pattern of HIV infection. Moreover, AIDS epidemiology has provided information to supplement the analysis arising from HIV surveillance. The understanding of the natural history of the infection has also enabled the use of back-calculation in reconstructing HIV epidemiology from reported AIDS figures. In the developing world, AIDS reporting may be the only method to determine the pattern of the epidemic.

Over the years there were two turning points in the application of AIDS epidemiology in Hong Kong. The first one could be traced back to the year 1993 when CDC revised the AIDS case definition, which has led to an increase in the specificity of AIDS reporting and is applied mainly in the United States and European countries. There was however an ever-increasing discrepancy between countries in their determination of what constituted AIDS. Comparability of AIDS data is becoming difficult, especially between developed and developing nations. The CDC revision has set the scene for the revision of our own AIDS case definition in 1995. Comparison of the patterns before and after 1995 should therefore be treated with caution. The second turning point was the year 1996 when Highly Active Anti-retroviral Therapy (HAART) became available. The efficacy of HIV treatment implies that clinical complications (thus ADIs) are no longer the inevitable outcome of HIV infection, if early diagnosis can be made and appropriate management can be given. In future, the number of AIDS cases is expected to be stable or even on the decline. The levelling off of AIDS cases reported after 1996 also means that back-calculation could no longer be applied to determine the HIV loads in the population.

Is there still a role for AIDS surveillance, and therefore, AIDS epidemiology? With the diminishing impacts of AIDS reporting in HIV surveillance, AIDS epidemiology is repositioning itself. An effective AIDS reporting mechanism can help generate useful morbidity data, which in turn could facilitate health service planning. On the other hand, the pattern of AIDS reporting can also reflect on the access of HIV patients to effective care. In this connection, the number of episodes of PCP is a good example of the kind of information which can be turned into important morbidity marker. The pattern of ADIs would however need to be interpreted with care as there are other confounders which may affect the figures, for example, the coverage of HIV test, facilities for diagnosing specific complications, and the consistency and timeliness of reporting AIDS cases to the system by the medical practitioners.

In conclusion, advances in laboratory technologies and treatment strategies have changed the landscape of AIDS in the past decade. AIDS epidemiology can no longer be a reflection of HIV epidemiology, but a description of disease pattern and burden arising from the infection. This is particularly true in western countries and also in Hong Kong.

Footnote

[*] Scientific Committee on AIDS is one of the three committees underpinning the Government-appointed Advisory Council on AIDS in Hong Kong.

[+] The standard AIDS report form (DH2293) can be downloaded from www.aids.gov.hk.

References

[1] CDC. Pneumocystis pneumonia - Los Angeles. MMWR 1981;30:250-2

[2] Joint United Nations Programme on HIV/AIDS and World Health Organization. AIDS epidemic update-December 2001. Geneva:UNAIDS, 2001.

[3] CDC. Current trends update on acquired immune deficiency syndrome (AIDS) - United States. MMWR 1982;31:507-8, 513-4.

[4] CDC. Revision of the CDC surveillance case definition for acquired immunodeficiency syndrome. MMWR 1987;36(suppl 1S):3S-15S.

[5] Buehler JW, De Cock KM, Brunet J-B. Surveillance definition for AIDS. AIDS 1993;7(suppl 1):S73-S81.

[6] CDC. 1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. MMWR 1992; 41:RR-17.

[7] Huang L, Hecht M. Why does Pneumocystis carinii pneumonia still occur? [editorial comment] AIDS 2000; 14:2611-2.

[8] Department of Health and Hospital Authority. Notification and death rate of tuberculosis (all forms) 1947-2001. http://www.info.gov.hk/tb_chest/index_2.htm accessed on 1.7.2002.