Monitoring the AIDS pandemic: The Status and Trends of the HIV/AIDS/STD Epidemics in Asia and the Pacific

The World Health Organization (WHO) Regional Office for the Western Pacific in Manila, Philippines, graciously hosted the first Monitoring the AIDS Pandemic (MAP) Network symposium, The Status and Trends of the HIV/AIDS/STD Epidemics in Asia and the Pacific, on 21-23 October 1997, which was an official satellite symposium of the 4th International Congress on AIDS in Asia and the Pacific.

The three-day MAP Network symposium held in Manila was the first symposium formally organized by this new global network formed in December 1996 by the AIDS Control and Prevention (AIDSCAP) Project of Family Health International, the François-Xavier Bagnoud Center for Health and Human Rights of the Harvard School of Public Health and the Joint United Nations Programme on HIV/AIDS (UNAIDS). It was the third in a continuing series of regional and global symposia that have been organized to understand the trajectory of the HIV/AIDS pandemic. Starting with the Status and Trends of the HIV/AIDS Epidemics in Africa Symposium that was held in Kampala, Uganda, in December 1995, a team of internationally recognized technical specialists in epidemiology, modeling, economics, demography, public health and international development was formed to monitor the dynamics of the HIV/AIDS pandemic and various regional epidemics. By collecting, analyzing and disseminating information on HIV/AIDS, this team of experts, which has grown rapidly over the course of two years into a global network, seeks to assist governments, organizations and the world at large to respond more actively and effectively to the challenges posed by the HIV/AIDS pandemic.

The MAP symposium in Manila brought together 40 global and regional experts, including MAP members and some specially invited participants, to achieve the following objectives:

· To present and share information on the status and trends of the HIV/AIDS/STD epidemics in Asia and the Pacific and analyze this information in a global context;

· To review the epidemiological and behavioral patterns among the HIV/AIDS/STD epidemics affecting the different populations in the region;

· To identify specific data needs for monitoring and forecasting the HIV/AIDS/STD epidemics in the region; and

· To produce and disseminate a consensus report on the current status of the HIV/AIDS/STD epidemics in Asia and on the current and projected trends for these epidemics in the region.

Because a large percentage of the world's population resides in the Asia-Pacific region, the symposium held in Manila was important in enabling MAP to focus strategically on the evolving HIV/AIDS and sexually transmitted disease (STD) epidemics in the Asian and Pacific countries, fuse current knowledge, identify gaps therein and determine topical and geographical areas for action.

The symposium began by first reviewing the global HIV/AIDS/STD situation to position the various Asia-Pacific epidemics within the worldwide context. From then on, the team concentrated on the situation in the overall region and the Western Pacific and Southeast Asian subregions, presented country-specific epidemic profiles on Australia, China, India, Japan, Malaysia, Myanmar, Philippines, and Vietnam and discussed past, present and potential future issues of relevance.

This provisional report, coauthored by the MAP Manila Symposium participants and produced in some 24 hours, reflects a consensus of the analysis, determinations, projections and recommendations brought forward during the symposium. Its aim is to provide information that can be used by international as well as local bodies to briefly review the most important aspects of the history of the HIV/AIDS epidemics in the Asia-Pacific region to date, recognize the current status of and trends within these epidemics, and take immediate action to affect the course of these epidemics in the future.

2. The Status and Trends of the HIV/AIDS/STD Epidemics in Asia and the Pacific

With a population in excess of 2.5 billion -- representing more than sixty percent of the world's population -- the Asia-Pacific region has the potential to influence greatly the course and overall impact of the global HIV/AIDS pandemic. This report examines the status and trends of HIV infection and AIDS in countries of Asia and the Pacific, a region which, for the purpose of this analysis, stretches from and includes India on the west, to Japan and island nations in the Pacific, and from China in the north, to countries forming Oceania in the south.

The spread of HIV in this region began in the early to mid-1980s. Early infections could be traced to sexual contacts with infected persons residing outside the region, as well as some apparent further spread within the region itself. By the late 1980s, however, it had become evident that the transmission of HIV was increasing among several populations, in some cases with great velocity, and that two sets of factors strongly influenced the course of the emerging epidemics: participation in sex work and patterns of injecting drug use (IDU).

By early 1997, South and South East Asia accounted for an estimated 5.2 million (23 percent) of the 22.6 million adults and children living with HIV in the world. About one-third of adults living with HIV in the region are female. As the HIV epidemic is still relatively recent, HIV disease, including AIDS, is only beginning to emerge and the associated needs for care are rising steeply. The estimated HIV prevalence in 15 to 49 year-old populations varies from zero (DPR Korea) to one per several thousand in most countries in the region, up to 2 to 3 percent in Cambodia, Myanmar and Thailand.

HIV epidemics in Asia and the Pacific are diverse, localized and have different trends over time (see Figure 1). Against this backdrop, however, it is becoming increasingly clear that the intensity of HIV epidemics associated with sex work, affecting both female sex workers and their clients, is primarily determined by the daily or weekly number of sex partners (clients) per sex worker, the frequency of use of commercial sex by men, and such other factors as the rate of regular condom use in commercial sex and the magnitude and quality of the response to the epidemics. Epidemics associated with injecting drug use have, in many situations, led to explosive outbreaks in the IDU population and then to their sexual partners (e.g., in the late 1980s in Thailand; Myanmar; the Yunnan province of China; and the Manipur state of India, Vietnam and Malaysia). The prevalence of HIV infection in IDUs in these areas reached staggering levels with prevalence reaching 50 to 90 percent within a few months.

Although HIV can spread rapidly among IDUs who share contaminated injection equipment, and then from them to their sexual partners, these epidemics have so far resulted only in limited spread of HIV to the heterosexual population at large. It may be assumed that for a variety of reasons including social isolation of some IDU populations and their sexual partners from other communities, and/or stigmatization to which they are subjected, there are only tenuous bridges between them and other sexually active adults. Strikingly, in Asia and the Pacific, HIV epidemics associated with commercial sex and those involving IDUs do not appear to fuel each other significantly. These epidemics appear to emerge and evolve almost independently from each other, as exemplified by the two concurrent HIV epidemics in Thailand, which were caused by two different subtypes of HIV, with minimum cross-over.

From a regional perspective, the magnitude and short-term trends of HIV epidemics are largely dependent on the extent of ongoing epidemics in a few countries: Cambodia, India, Thailand, Myanmar and, because of their population size, Indonesia and China. With a population close to 1 billion and multiple epidemic foci, India projects the image of a complex epidemic, involving focal outbreaks among injecting drug users and extensive HIV spread among female sex workers and their clients in several regions.

Of increasing concern in the region is the issue of blood safety and HIV transmission. In 1995, WHO/GPA estimated that less than 50 percent of blood transfusions in the region were being routinely screened for HIV. Currently in Bangladesh, virtually no screening for HIV antibodies is performed in the nearly 200,000 units of blood transfused annually. In India and Myanmar, screening of donor blood for HIV remains far from complete and measures are being taken through improved donor selection to address this issue.

If the HIV epidemics were analyzed on a country-by-country basis, as if HIV epidemics respected national geopolitical boundaries, most countries in the Asian-Pacific region project the reassuring image of low prevalence (proportion of adults living with HIV/AIDS) and low incidence (proportion of adults newly infected each year). Such is the case of Australia and New Zealand which, early in the epidemic, experienced sudden epidemics in men having sex with men (MSM) with a peak of incidence in the mid-1980s, followed by a rapid decline. The spread among IDUs in these two countries has been and remains limited (less than 2 percent), and heterosexual transmission remains at low levels (the prevalence of HIV among pregnant women is lower than 1 per 10,000).

Such is the case as well for Japan, where an initial dramatic outbreak of HIV infection among people with haemophilia was brought under control in the mid-1980s (see Box 1), and where other modes of transmission are only contributing minimally to a limited HIV burden in the country (less than 1 HIV-infected person per 10,000 adults in 1997). Also reassuring would be the situation in China where, in spite of increasing trends in HIV incidence in IDUs in selected southern provinces (Yunnan, Guanxi), the overall national rate of HIV prevalence in adults in this large country remains below 4/10,000 -- one tenth to one-fifth the prevalence found in Western Europe and in North America, and the spread of HIV from IDUs to their sexual partners accounts for most of the heterosexually acquired HIV infections.

Box 1
The Lingering Effect of a Controlled HIV Epidemic:
HIV infection from blood coagulation products in Japan

As of the end of August 1997, the cumulative reported number of AIDS cases in Japan was 1,657, which was still very low in Asia comparatively, even among developed countries. A little less than half of the cases (41.6 percent) are, however, those infected through blood coagulation factor products. This high percentage of hemophilia cases is still the distinctive characteristic of HIV infection in Japan and not seen in other countries in the world.

These infections resulted from transfusion of infected blood products in the early 1980s. In July 1985, the Japanese Ministry of Health and Welfare approved the usage of heat-treated blood coagulation factor products. Since the introduction of these heat-treated products, virtually no HIV infection has occurred through infected blood products. The epidemic of HIV in this population is now reflected in an increasing number of people developing AIDS and dying from the disease each year.

Almost all those infected from blood products are hemophiliacs type A or B, and the rest are related coagulopathy, such as von Willebrand disease and few cases of administrated coagulation factor for the prevention of excessive bleeding. An estimated 35 to 40 percent of hemophiliacs in Japan are infected with HIV.

Cases from blood products show a tendency to have a longer incubation period than the cases via other routes of infection. Most of these cases, however, will have the onset of AIDS in the near future. Consequently, every prefectural government in Japan is now preparing base hospitals for the treatment of AIDS. As of the end of May 1997, almost 1,300 people infected through blood products were living in Japan, accounting for 32.4 percent of the total population living with HIV in the country.

Less reassuring would be the estimated and projected trends in India. There, a sharp increase in the estimated number of HIV infections from a few thousand in the early 1990s to a cumulative minimum of 2.5 million in 1997, in a context of a severe gap of knowledge about prevailing risk-taking sexual behaviors, creates great uncertainty about the future course and impact of the epidemics. Yet, the 1997 national prevalence of HIV infection in adults may be between 0.2 to 0.5 percent which, on first analysis, would not place India high on the world list of "hardest hit" countries where prevalence rates may range from 10 to 25 percent.

HIV prevalence among sex workers in India varies widely from state to state, with high HIV prevalence in western and southern India to low levels of HIV in eastern and northern India. HIV prevalence among sex workers tested in Mumbai (formerly Bombay) rose from 1 to 51 percent between 1987 and 1993. Prevalence among sex workers in Calcutta was consistently low at about one percent until 1994, but there are indications that it might be rising. In Mumbai, HIV prevalence increased from two to three percent in STD clinic attendees before 1990 to 36 percent in 1994. Among antenatal clinic attendees tested in Mumbai, two percent tested positive for HIV in 1995 and around 5 percent in 1996. In Manipur, rates of HIV infection among the antenatal clinic attendees are rising.

In some countries, HIV prevalence has remained very low (less than 0.1 percent in the 15- to 49-year-old population. In the Philippines, AIDS case reporting has slowly increased to a total of 958 reported as of September 1997. HIV surveillance has found only a few cases of HIV infection among female sex workers, male STD clinic patients and men who have sex with men. Similarly, a small number of AIDS cases and low levels of HIV infection have been reported from Indonesia.

When examined through the lens of current national HIV prevalence and incidence rates, most other countries in Asia and the Pacific would conform to a pattern of low prevalence and slow HIV spread (see Figure 2).

Are HIV epidemics likely to expand abruptly in India, or in other countries in Asia and the Pacific? To answer this question, possible clues can be drawn from past history in several countries, including Thailand, Myanmar and Cambodia. In Thailand, information was available on IDU needle sharing practices and sex work in the country when, in the mid-1980s, HIV had not begun to spread in epidemic form. The epidemics that were predicted but insufficiently prevented did occur later in this decade, infecting an estimated 800,000 women, men and newborns by 1997. Prevalence rates in pregnant women reached 2 to 3 percent in 1995 nationwide, while rates in army recruits rose to around 4 percent in 1993, then levelled off and began to decline. In Myanmar, rates of HIV infection in IDUs, sex workers and pregnant women are similar to those found in Thailand. Consistent with an East-West gradient of HIV prevalence in Myanmar, the epicenter of the epidemics appears to lie east of the country. The analysis of geographical differentials in HIV infection rates and trends over time underscores both the deepening severity of the epidemics and the potential for national, aggregate data to mask important sub-national epidemic trends. In Cambodia, HIV spread rapidly in the early 1990s, reaching high levels in sex workers (about 40 percent HIV prevalence) and from 2 to 3 percent of the 15- to 49-year-old population (see Box 2). The evolution of the epidemic in Cambodia country illustrates the potential for HIV to spread rapidly and widely when patterns of commercial sex involve high mobility, a high sexual partner exchange rate and low condom use.

Box 2 Cambodia:
An Explosion of HIV

The Kingdom of Cambodia has seen a rapid increase in HIV infection over the past few years, giving it the likely dubious distinction of the most widespread and serious HIV epidemic in Asia. Although only about 600 AIDS cases had been reported by mid-1997, the epidemic has progressed so quickly that it is now estimated that approximately 100,000 individuals are infected with HIV. Although substantial variations exist regionally, national rates derived from 1997 serologic surveillance results indicate that approximately 40 percent of sex workers, 6 percent of police/military, and 3 percent of antenatal women are HIV-infected.

Other sexually transmitted diseases are likewise high in Cambodia, contributing to HIV spread in the country. A 1996 survey in selected cities indicated prevalence rates of gonorrhoea of 12 to 31 percent among sex workers and 2 to 6 percent among antenatal women. Similarly, syphilis rates were 4 to 24 percent in sex workers, 0 to 12 percent among police and the military, and 2 to 8 percent among antenatal women.

Despite the widespread HIV epidemic in Cambodia, there are some recent HIV prevention sucesses which may promote a slowdown. A condom social marketing program by Population Services International reports condom sales increasing from about 5 million in 1995 to 9.5 million in 1996. Behavioral surveys confirm that many of these condoms are being used for commercial sex. Men frequent sex workers in high proportions in Cambodia (about 75 percent of military/police and 37 percent of students report visiting a sex worker in the past year), but condom use has substantially increased over the past few years. However, given the state of the Cambodia's political and economic situation, these successes are fragile and continued concentration and strengthening to maintain and increase behavioral changes will be required.

Empirical evidence thus exists showing that sudden and sharp increases in HIV incidence can and have occurred in Asia. However, the lack of quantitative and qualitative epidemiological, behavioral and social information on the nature of and linkages between sexual networks in any of these countries rules out any reliable prediction of the future course of HIV epidemics in countries that would intuitively appear vulnerable to rapid spread. These countries include, in particular, Malaysia, Nepal and Vietnam, where rapid increases of HIV incidence in various vulnerable populations are being noted. There is an urgent need to collect and analyze systematically the information needed for the dual purpose of projecting epidemic trends and targeting prevention toward factors that seem to influence the vulnerability of the population to the further -- and possibly rapid -- spread of HIV.

The analysis of HIV epidemic trends in the region becomes more meaningful when a focus is placed on populations whose cultural and social affinity and networks transcend geopolitical borders. A new geography of HIV/AIDS in the region then emerges that helps recognize the foci of intense HIV spread. (See the map, Figure 3.) These include large metropolitan areas in western and southern India (Mumbai, Tamilnadu) (see Box 3); the India/Nepal border area; the larger "Golden Triangle," which reaches out to Northern Thailand, eastern Myanmar, but also encompasses the areas of Manipur in India and Yunnan in China; and the Mekong delta area, which includes Cambodia and southern Vietnam. To gain better understanding of the dynamics of HIV epidemics, factors of affinity between populations as well as mobility patterns must be explored and mapped.

Box 3
India: An uncertain future

With more HIV infections than any other country in the world, India gives the impression that HIV infection is common and that there is a severe epidemic in the country. However, the estimated 2.5 million HIV infections should be considered in the context of the close to 970 million population of India. The prevalence of HIV is about 0.3 percent, a rate much lower than many other countries in the Asia-Pacific region.

In India, HIV infection is not evenly distributed throughout the country. While it is true that HIV has now been reported from all except one of the 32 states and Union Territories, the infection is actually highly localized.

AIDS Cases Pie

As the pie chart shows, almost half the total 4,828 cases of AIDS reported so far are from the state of Maharashtra, and of these 80 percent are from Mumbai city. The state of Tamilnadu contributes another 22 percent. It is worth noting that 21 of the 32 states add only 4 percent of the total AIDS cases reported. The state of Manipur, which contributes 6 percent to the total, reports that all the cases of AIDS recorded so far are in IDUs. There are an estimated 25,000 IDUs in the state, which borders on the "Golden Triangle," and at last estimate the HIV prevalence in this group varied from about 50 to a staggering 80 percent.

Another group showing high HIV prevalence is that of long-distance truck drivers and their helpers. It has been shown that this group has a steadily increasing prevalence. In Chennai (Madras) a seroprevalence of close to 10 percent has been reported.

These data bring out the fact that it is often misleading to consider a country as a homogenous entity as far as HIV/AIDS is concerned. Many countries have more than one epidemic, often at different stages of development and in different communities and regions. An appreciation of this fact is important in planning suitable interventions.

The potential for continued spread of HIV/AIDS in Asia and the Pacific is real and requires determined and sustained prevention efforts. Several countries have already experienced intense HIV epidemics in certain population groups or, in some cases, in the population at large. In these countries, including India, Thailand, Myanmar and Cambodia, the individual impact of HIV has begun to be felt as AIDS has imposed new demands on the health care systems. It is essential that countries reinforce their prevention and care efforts in order to enhance their response to the existing HIV/AIDS challenge. In addition, countries should make every effort to collect and analyze the information needed to assess and monitor the evolving potential for large-scale HIV epidemics.

Recognizing the threat of emerging or fast-growing epidemics in certain populations is essential to an early and effective response. Acknowledging the possibility that, in other populations, rapid and extensive spread of HIV may not occur is equally crucial as policy and decision makers may, on the basis of this information, orient efforts and resources towards people who are most at risk.

Countries with epidemic spread	Current HIV Epidemic Trends		Main Populations Affected	Projected HIV Epidemic Trends
.	HIV incidence	HIV prevalence	.	(3-5 Years)
Australia	Low and decreasing	Low and stable	MSM	Decline
Cambodia	High and increasing rapidly	High and increasing	Individuals with high and moderate risk heterosexual behavior	Sustained upward trend
China	Low except in Yunnan	Low and increasing	IDU	Increasing
India	Moderate and increasing(significant regional variation)	Still low but increasing(significant regional variation)	Individuals with high-risk heterosexual behavior and IDUs	Increasing
Malaysia	Moderate and increasing	Low and increasing	Principally IDUs but increasing among individuals with high risk sexual behavior	Increasing
Myanmar	High and increasing	High and increasing	Individuals with high-risk heterosexual behavior, IDUs and their spouses	Increasing
New Zealand	Low and decreasing	Low and stable	MSM and IDU	Decline
Papua New Guinea	Moderate and increasing	Low but increasing	Individuals with high-risk heterosexual behavior	Slowly increasing
Thailand	Moderate and stabilising in specific groups	High but stabilizing	IDUs and individuals with high and moderate risk heterosexual behavior	Tending to stabilize
Vietnam	Moderate and increasing	Still low but increazing	Principally IDUs but increasing among individuals with high risk sexual behavior	Increasing

Countries with low transmission	Current HIV Epidemic Trends		Main Populations Affected	Projected HIV Epidemic Trends
.	HIV incidence	HIV prevalence	.	(3-5 Years)
Bangladesh	Low	Low	Individuals with high-risk heterosexual behavior	Slowly increasing
Indonesia	Low	Low	MSM, Bisexual and high-risk heterosexual behaviour	Slowly increasing
Japan	Low	Low	Previously blood product related, currently sexual	Slowly increasing
Hong Kong	Low	Low	MSM	Slowly increasing
Nepal	Low except in IDUs	Low except in IDUs	Individuals with high-risk heterosexual behavior and IDUs	Slowly increasing
Philippines	Low	Low	Individuals with high-risk heterosexual behavior	Slowly increasing
Singapore	Low	Low	MSM, IDUs	Slowly increasing
Sri Lanka	Low	Low	Individuals with high-risk heterosexual behavior and MSM	Slowly increasing

Country	IDU*	HET**	MSM***
Australia	+	+	++
Bangladesh	+	+	+
Cambodia	+	+++	+
China - Yunnan Prov. - Hong Kong - Rest of China	+++ + ++	+ + +	+ ++ +
India - West & South - Central & East - Northeast	+ + +++	+++ + +	+ + +
Indonesia	+	+	++
Japan	+	+	+
Laos	+	+	+
Malaysia	+++	++	++
Myanmar	+++	++	++
Nepal	+++	++	++
Papau New Guinea	+	++	+
Philippines	+	+	+
South Korea	0	+	+
Sri Lanka	0	+	+
Thailand	+++	++	+
Vietnam	+++	+	+

Chart legend:

+++	high or rapidly growing
++	relatively low or plateauing
+	not a major component
0	no evidence of spread
*	IDU: Injecting drug users
**	HET: Heterosexual men and women
***	MSM: Men having sex with men

3. Risk and vulnerability

There are three factors that appear to play a crucial role in HIV transmission in the Asia and Pacific region: female sex work, substance use and mobility. Female sex workers and their clients have been a major factor in the heterosexual transmission of HIV in Thailand, Cambodia and parts of India and Myanmar. Separate but explosive epidemics have been seen in some IDU populations in Thailand, Myanmar, Manipur (India), and Malaysia. And mobile populations, particularly at national borders, are at higher risk of HIV acquisition due to the fact of being away from home and community, and the anonymity and loneliness of traveling. The following three sections focus on monitoring the HIV epidemics in these vulnerable populations.

Female sex workers operate in all countries in the region, but it is important to know how large the sex-worker population may be to adequately interpret surveillance results. It is believed that in some countries, rapid increases in the number of sex workers have resulted from significant political, social or economic changes.

The number of female sex workers in countries can range from a few thousand to a few hundred thousand. Thailand, for example, has approximately 80,000 to 100,000 female sex workers -- an estimate that only recently has gained consensus. In most countries, however, validated estimates of the numbers of sex workers are almost non-existent. Without such numbers, interpretation of surveillance results, policy commitment, planning, intervention designing, and resource allocation are extremely difficult (see Box 4).

Box 4
How Many Sex Workers?

Most estimates of the numbers of female sex workers are either restricted to a subset of licensed commercial sex establishments or ad hoc unvalidated "guesstimates." In Dhaka, Bangladesh, estimates of the number of sex workers frequently quoted range from 3,000 to 100,000. It is impossible to anticipate the scale of services and funds and nature of interventions needed without more meaningful numbers.

One promising direct estimation method being used in Thailand, Laos, Nepal and Taiwan developed by the Thai Red Cross and the East-West Center is the method of comprehensive geographic mapping of sex work sites and types. In Thailand, this method has now been adopted by the Ministry of Public Health for their annual survey of sex work. In several areas, the method has detected up to 50 percent more sites than estimates by using the conventional method of referral by STD patients. In other countries, population estimates have been generated through innovative applications of capture-recapture techniques. In Dhaka, CARE/Bangladesh recruited and trained sex workers to assist in "capturing" street sex workers throughout the city by handing out different colored cards for 24 hours on two different days within a fortnight.

All estimates need regular updating and validation but these two approaches are worthy of replication in other settings. Neither method can work without the trust and involvement of sex wor

Monitoring the AIDS Pandemic (MAP) Network

The Status and Trends of the HIV/AIDS/STD Epidemics in Asia and the Pacific

October 1997, Manila, Philippines

1. Introduction

2. The Status and Trends of the HIV/AIDS/STD Epidemics in Asia and the Pacific

3. Risk and vulnerability