Krystal CK LEE, Ka-Hing WONG
Epidemiology is the study of the distribution and determinants of health-related states or events in specified populations, and the application of this study to the control of health problems.1 Epidemiology contributes to the practice of public health, especially as regards communicable diseases, including the description of the pattern of a condition in populations, examination of factors at individual/population levels that affect occurrence and disease progression, evaluation of preventive strategies as well as control programmes. Modern day epidemiology involves the collaboration of multiple disciplines from laboratory science to community medicine. In this chapter, the epidemiology of HIV infection in both global and local context is described and discussed.
Epidemiological studies of HIV began in 1981 when there were reports of outbreaks of Kaposi's sarcoma and Pneumocystis carinii (now Pneumocystis jiroveci) pneumonia (PCP) in homosexual men in several cities in the United States.2 Other reports followed, all sharing a similar clinical syndrome in injection drug users, persons with haemophilia and transfusion recipients. In 1982, the U.S. Centers for Disease Control defined and named the syndrome as 'Acquired Immune Deficiency Syndrome' (AIDS). The virus causing AIDS, HIV, was only identified in 1983 and serological tests became widely available in 1985. Three transmission routes have been established early in the epidemic: sexual, parenteral and perinatal. Without treatment, on average half of the people infected with HIV develop symptoms and signs of immunodeficiency in ten years?time. The prolonged latency of HIV infection has caused the epidemic to go unchecked and grow in exponential proportion in many parts of the world.
Over the years, HIV/AIDS has evolved to become one of the most devastating pandemics in history. The estimated number of people living with HIV worldwide went up from 10 million in 1993 to 40 million (37-45 million) through 2005. Despite the fact that effective prevention is possible since the early days, less than one fifth of the people received the necessary preventive services globally. Since the mid 90s, effective treatment has become available, but to this day less than one in ten in Africa and one in seven in Asia received the necessary treatment. Combination treatment has turned HIV infection into a chronic manageable illness in the developed world, but in places where treatment is inaccessible morbidity and mortality pattern remain the same as they were decades ago. In 2005 alone, 5 million new infections have occurred and 3 million lives were lost to the infection. Cumulatively, some 25 million deaths have resulted since the start of the epidemic.
The HIV epidemic spreading around the world is not a solitary one. The global pandemic consists of many diverse epidemics occurring at different pace and locations, and affecting heterogeneous and often marginalised groups of populations. Dissemination of HIV has not occurred as a random process, as there's complex interplay of the virus, the host, and ecologic (behaviours, social environment) factors. In year 2000, the World Health Organization suggested a classification schema that describes the epidemic by its current state - low level, concentrated or generalised (Box 2.1).
Generalised epidemics are mainly found in African countries where HIV has been fuelled by its unique sociocultural settings. Sub-Sahara Africa is home to 60% of the HIV infection population worldwide and in six African countries, including South Africa, HIV prevalence reaches 20-30% in the general population. In many cases HIV has contributed to radical decrease in life expectancy and substantial change to sociodemographic composition.
Concentrated epidemics have occurred in many parts of the world and are still occurring nowadays in different at risk populations, which most commonly include the injection drug users (IDU), men having sex with men (MSM), clients of commercial sex workers and commercial sex workers. Pockets of concentrated epidemics among MSM were first observed in North American and Western European countries and are continuing to occur today. In some cities including San Francisco, Toronto, London, Barcelona, the reported prevalence among MSM in recent years has ranged between 10% and 25%.3,4 Lately, there're concerns about the escalating problem among MSM in some Asian countries, with prevalence reported to be as high as 15% to 30% in Phnom Penh and Bangkok, 5% to 10% in Taipei, and 3% in one Beijing study.
HIV epidemic in IDU is characterised by its potential for explosive spread. Prevalence could increase to as high as 40% (as in Bangkok) or even 80-90% (as in cities in Myanmar and Yunnan in late 80s) in just over one or two years. Countries in South America, Asia (China, Vietnam, Nepal and Taiwan in 2003)5 and lately in Eastern Europe are most severely affected by the IDU epidemic.
The main driving force of concentrated epidemic through heterosexual contact is the phenomenon of commercial sex between male clients and female sex workers (FSW). This is observed in a few Asian countries such as India, Cambodia, Thailand and Vietnam with prevalence among FSW of about 5%. When a large number of men have visited sex workers, the epidemic can be felt in the general population, as in the case of Cambodia and Thailand where the adult prevalence has ranged between 1% and 3% in 2005. There are also interactions between epidemics among different sub-populations. For example in Vietnam there has been a clear association of women who are both IDU and FSW. Epidemiologic patterns of HIV infection are shaped by the social network in the community, the characteristics of which may vary from one country to another.
Low level epidemics denote a state when, relatively speaking, HIV has not affected a significant proportion of populations or sub-populations. The state is volatile and changes could occur over short period of time. Hong Kong is an example of the HIV low prevalence areas, though constantly challenged by the concentrated epidemics in the nearby regions (see below).
In Hong Kong, the first case of HIV infection was reported in 1984. Under the voluntary HIV/AIDS reporting system, Department of Health has received a total of 3198 reports of HIV infection at the end of 2006. All except perinatally transmitted (19) or transfusion related (21) cases were adults (Box 2.2). It is estimated that in 2005 there were 3240 people living with HIV infection in Hong Kong, giving an overall prevalence in adult population of < 0.1%. The number of AIDS reports has been maintained at a stable level of about 60 cases annually since the availability of HAART in the mid-90s. At present, PCP and tuberculosis are the commonest primary AIDS-defining illness in Hong Kong. The HIV epidemic in Hong Kong can be roughly divided into three phases, according to the rate of growth in number of reports and predominant risk for transmission.
The first phase, between mid-80s and early-90s, began with the identification of about seventy cases of HIV infection acquired through transfusion of blood or blood products in Hong Kong before HIV screening became available in 1985. A majority of these were patients with haemophilia (66 local cases were reported so far). At the same time, cases of infection acquired in MSM as a result of sexual contacts outside Hong Kong were reported, with more than half of which being non-Chinese. Until early 90s, the absolute number of reported infections has remained relatively low and stable with less than fifty cases added per year.
Heterosexual transmissions became prominent starting in the early 90s and marked the start of the second phase of the epidemic. This was coupled with the thriving heterosexual epidemic in a number of Southeast Asian countries frequented by local people. The heterosexual spread started with a male predominant pattern with the 5-year averaged male to female ratio gradually decreasing from 7 in 1990, 3.6 in 1995, 2.4 in 2000 to 2.0 in 2005. A majority (about 80%) of the infected men were ethnic Chinese. For female, the ratio of Chinese to non-Chinese rose slowly to one in two near the turn of this century, while the remainder was largely from Asian countries. Despite the increasing number of infections reported during the period, prevalence among attendees at the Social Hygiene Clinics has remained low at <0.1%. Similarly, HIV has not kicked off in injection drug users in the territory. Only scattered number of infection was reported from drug rehabilitation services including methadone clinics.
At the turn of the century, the HIV situation has quietly entered a new phase. The important observation leading to this was the increase in infections among local men having sex with men (MSM), while at the same time, heterosexual transmission assumed a stable level. A recent review of epidemiological data revealed that HIV transmission in MSM has gained momentum in Hong Kong. The alarm was backed by the following observations: (a) a four fold increase in HIV reports in Chinese MSM (21 reports in 2000 to 76 in 2005); (b) an increase in proportion of MSM tested positive (by one NGO) from <1% in 2002 to 2.5% in 2005; (c) identification of 2 clusters of HIV-1 subtype B infection involving at least 40 men for the first time in Hong Kong. These infections were believed to have occurred locally over a short period of time and investigation showed that unprotected sex between men, sex networking through internet, and use of soft drugs were key risk factors among these cases.
The three phases of HIV epidemic in Hong Kong echo the pattern of distribution of molecular subtypes and CRFs (circulating recombinant forms) locally and the region. In Hong Kong, the commonest subtypes identified from 2002 to 2005 were CRF_01AE (49%) and B (35%). Subtype C, CRF07_BC and CRF08_BC each accounted for 5%, 3% and 2% of total cases respectively. The remaining 5% was made up of ten other subtypes or CRFs. Approximately 70% of CRF01_AE occurred in male, Chinese and heterosexually transmitted cases, whereas subtype B is more common in male (96%), Chinese (86%) and MSM (62%). Subtype C is commoner in female (56%), non-Chinese (90%) and heterosexually transmitted cases (63%). A rising proportion of subtype B has been observed from 34% in 2002 to 40% in 2005, coupling with the rising trend of infections among MSM. Distribution of CRF07_BC/CRF08_BC is similar to CRF01_AE, except that a higher proportion is found in injecting drug users (33%). The CRF07_BC and CRF 08_BC were initially identified in mid-90s in Yunnan, and have been circulating widely in China along distinct drug trafficking routes with CRF07_BC has travelled far north to Xinjiang and CRF08_BC to Guangxi province.
The complexity of the much diversified HIV epidemics globally has intrigued researchers who have used and developed different models to describe the mechanism and to understand the factors affecting the dynamics of the epidemic. A fundamental approach is to conceptualise the HIV epidemic as one form of infectious disease epidemic. The growth of an infectious disease epidemic is expressed by the basic reproductive number, R0, which describes the number of secondary infections that arise from a primary infection. The basic reproductive number is a function of the probability of infection per contact (s), the number of contacts (C), and the duration of infectivity (D) (R0 = βCD). Efforts to control infectious disease are set out to reduce the empirical value of these terms by modifying the social conditions under which individual risk factors lead to disease.
Indeed, observations from the past two to three decades have informed us that HIV transmission dynamics and the differential distribution are affected by a broad range of factors. Among the different paradigms described by various researchers, the social epidemiological approach provides one simple and practical framework to study the determinants of HIV/AIDS. Poundstone and others proposed a three-level framework on such and includes individual (and viral), social and structural factors.6 All these directly or indirectly affect the variables in the determination of the basic reproductive number.
Individual factors include biological, demographic, and behavioural risk factors that may influence risk of HIV acquisition (β) and disease progression (which then affect β, C and D). Examples of these include:
(a) Type of exposure. Needle sharing and anal sex carries higher risks than male-to-female to female-to-male transmission.7
(b) Viral burden as measured in plasma. Studies have shown that transmission is rare among persons with levels of less than 1500 copies of HIV-1 RNA per millilitre.8 Effective clinical treatment therefore plays an additional role of controlling HIV spread.
(c) Stages of HIV infection. Both acute and advanced stages of HIV infection are associated with increased risk of HIV transmission. A recent study demonstrated that acute HIV infection resulted in an 8-10-fold increase in the risk of male-to-female transmission.9
(d) Genital ulcer disease. Studies have shown that HSV-2 infection, for example, increases risk of HIV acquisition by 2 folds and transmission by 5 folds.10
(e) Circumcision. Though controversial as a preventive intervention, a recent randomised controlled trial in South Africa demonstrated that circumcision offered 60% protection against HIV infection after controlling for a range of behavioural factors.11
(f) Immunogenetic factors. Studies of individuals uninfected with HIV despite high level of behavioural risks demonstrated that various immunogenetic factors may contribute to the phenomenon. e.g., mutant (homozygous in delta32 deletion) in the CCR5 co-receptor causes non-expression of the co-receptors, thus rendering the individuals resistant to HIV infection.
Other factors that increase HIV risk include cervical ectopy and menstruation in female while appropriate use of condoms can effectively reduce transmission risk by 98%. Public health interventions could contribute to a reduction of the occurrence of infections in individuals through modifying any of the individual factors, nonetheless the demonstrated effectiveness of which varies in different context. Viral factors such as its subtype and its replicative capacity have also been proposed and found to be related to pathogenicity and transmissibility in some studies.12
Social-level factors include critical pathways by which community and network structures link persons to society, and the relationship between individuals within networks that influence the occurrence of some of the risk factors. For examples, social network such as sexual network and injecting drug use network has been found to be important in understanding STI (sexually transmitted infection) epidemiology. Specific network characteristics such as size and distribution of subgroups and concurrent sexual partnerships are associated with HIV spread. The prevalence and distribution of the individual protective or at risk behaviours described above, e.g., prevalence of unprotected anal intercourse among MSM or prevalence of STI in the female sex workers (FSW) population are some of the societal level risk factors.
Finally, structural-level factors include social (such as racial/ethnic differentials, gender inequalities, stigma and discrimination, demographic change through population mobility and migration) and economic factors, as well as laws and policies (such as those relating to access of sterile injecting equipments) and warfare. The socio-cultural context of a society plays important role in affecting HIV spread. A case in point is the explosive HIV epidemic among injecting drug users in South Asia along the drug trafficking route from Myanmar, Thailand and China etc where the communities of injection drug users formed and facilitated the spread of HIV. Observations in developed countries where effective treatment is readily accessible hypothesised that optimism from effective treatment has caused increase in HIV risk behaviour, although there is no conclusive data to confirm yet.13 In other words, risk behaviours do not occur at random, neither is a population homogeneous in risk behaviours. Characteristics of social networks and their contextual relationships need to be incorporated in epidemiologic studies in order to yield meaningful analysis. Conventional biostatistical approach to HIV epidemiology is simply inadequate.
Taking Hong Kong as an example, there are many possible inter-related reasons to account for the maintenance of a low HIV prevalence despite rise in neighbouring cities. The basic premise is the low level of risk behaviour against the background of small number of virus that has been introduced into the population. Todate, it is likely that a significant proportion of the HIV cases so far have been acquired outside Hong Kong. About half of the heterosexual male attending the public HIV clinic reported the suspected place of infection outside Hong Kong, and molecular study of HIV-1 of more than 1000 samples did not show significant clustering of cases before the detection of the MSM cases in the recent years.
What then have contributed to this favourable combination of low behavioural risk and low prevalence over the years? A supportive environment for sustaining broad coverage public health programmes,14 even before the first case of HIV report, was crucial. As evidenced in most Asian countries like Thailand and Vietnam, the HIV outbreaks in IDU have acted as a catalyst for subsequent growth of the epidemic in the community, while commercial sex has fuelled the heterosexual spread. The dynamics of the linkage of the two HIV networks is a hot topic for public health research. In Hong Kong, a low threshold approach has been adopted when methadone clinics were introduced one decade before HIV arrived in the territory. The high coverage of these clinics turns out to be an effective harm reduction strategy which, coupled with an easy access to syringes without need for prescription, had protected IDUs in Hong Kong, despite the lack of an explicit policy to the effect. The operation of free STI services at the territory's Social Hygiene Clinic serves the same purpose of providing accessible treatment services and prevention advice to a significant proportion of people practicing risk behaviours.
Our privilege of enjoying a low level HIV epidemic is being challenged by at least two distinct phenomena. Firstly, the expanding local HIV epidemic in MSM is a cause for concern (Box 2.3). Applying the Asian Epidemic Model with Hong Kong data, it was speculated that an explosive MSM epidemic is imminent, with incidence that may double in two and a half years time, unless preventive interventions are strengthened in the near future (Reference 1 under "Further reading"). Secondly, Hong Kong is geographically situated between the rising epidemics in Mainland China and nearby regions. This is particularly relevant as the HIV spread in IDU and in commercial sex context are both accelerating, against the background of a highly mobile population. So far, as much as half of all HIV infected drug users reported are from other Asian countries (non-Chinese). One-fifth of the drug users identified in methadone clinics in recent years are not local resident and two thirds were not locally born. At the same time, the HIV prevalence in Guangdong is in the range of 3-6%, ten to twenty times of that in Hong Kong. Similarly, there's a highly mobile population of FSW in Hong Kong, and many may not have been in contact with local public health programmes. The high coverage of methadone clinics and Social Hygiene Service clinics may mean very little to this unique mobile risk-taking populations. The change of social context could play a determining role in the dynamics of HIV epidemic in the territory, even though host and virus factors have remained the same.
Source of data in the article were obtained from Special Preventive Programme, Department of Health unless otherwise stated.
1. Brown T. HIV/AIDS in Hong Kong. Living on the edge. Hong Kong: Department of Health, 2006. Available from www.aids.gov.hk
2. Monitoring the AIDS Pandemic (MAP) Network. IDS in Asia: Face the facts. A comprehensive analysis of the AIDS epidemic in Asia. MAP report 2004. Available from http://www.mapnetwork.org/docs/MAP_AIDSinAsia2004.pdf
3. UNAIDS. Report on the global AIDS epidemic. Geneva: UNAIDS, 2006. Available from www.unaids.org
John Last. Dictionary of Epidemiology. USA: Oxford University Press, 2000.
Centers for Disease Control (CDC). Pneumocystis pneumonia--Los Angeles. MMWR Morb Mortal Wkly Rep 1981;30:250-2.
Centers for Disease Control and Prevention (CDC). HIV prevalence, unrecognized infection, and HIV testing among men who have sex with men--five U.S. cities, June 2004-April 2005. MMWR Morb Mortal Wkly Rep 2005;54:597-601.
EuroHIV. HIV/AIDS Surveillance in Europe. Mid-year report 2005. Saint-Maurice: Institut de veille sanitaire, 2006. No. 72.
Chen YM, Lan YC, Lai SF, Yang JY, Tsai SF, Kuo SH. HIV-1 CRF07_BC infections, injecting drug users, Taiwan. Emerg Infect Dis 2006;12:703-5.
Poundstone KE, Strathdee SA, Celentano DD. The social epidemiology of human immunodeficiency virus/acquired immunodeficiency syndrome. Epidemiol Rev 2004;26:22-35.
Smith DK, Grohskopf LA, Black RJ, et al. Antiretroviral postexposure prophylaxis after sexual, injection-drug use, or other nonoccupational exposure to HIV in the United States: recommendations from the U.S. Department of Health and Human Services. MMWR Recomm Rep 2005;54(RR-2):1-20.
Quinn TC, Wawer MJ, Sewankambo N, et al. Viral load and heterosexual transmission of human immunodeficiency virus type 1. Rakai Project Study Group. N Engl J Med 2000;342:921-9.
Wawer MJ, Gray RH, Sewankambo NK, et al. Rates of HIV-1 transmission per coital act, by stage of HIV-1 infection, in Rakai, Uganda. J Infect Dis 2005;191:1403-9.
Corey L, Wald A, Celum CL, Quinn TC. The effects of herpes simplex virus-2 on HIV-1 acquisition and transmission: a review of two overlapping epidemics. J Acquir Immune Defic Syndr 2004;35:435-45.
Auvert B, Taljaard D, Lagarde E, Sobngwi-Tambekou J, Sitta R, Puren A. Randomized, controlled intervention trial of male circumcision for reduction of HIV infection risk: the ANRS 1265 Trial. PLoS Med 2005;2:e298.
Quinones-Mateu ME. Is HIV-1 evolving to a less virulent (pathogenic) virus? AIDS 2005;19:1689-90.
Elford J. Changing patterns of sexual behaviour in the era of highly active antiretroviral therapy. Curr Opin Infect Dis 2006;19:26-32.
Chan MK, Lee SS. Can the low HIV prevalence in Hong Kong be maintained? AIDS Educ Prev 2004;16(3 Suppl A):18-26.