Portal de Periódicos da CAPES

The epidemics of injecting drug use and HIV in Asia

2001; Lippincott Williams & Wilkins; Volume: 15; Linguagem: Inglês

10.1097/00002030-200100005-00012

1473-5571

Peter D. Ghys, Wayne Bazant, Maristela Monteiro, Sandro Calvani, Stefano Lazzari,

Syphilis Diagnosis and Treatment

Introduction In recent years, two concomitant and related epidemics have swept across many Asian countries: the rapid increase in injecting drug use (IDU), and the spread of HIV. In some countries, this combination has already created significant public health problems, while it is of major concern in others. IDU is a global phenomenon occurring in an increasing number of countries. The number of countries and territories worldwide that report IDU increased from 80 in 1992 to 121 in 1995 and to 136 in 1999 [1-3]. The increasing use of injectable drugs is strongly linked to the increased consumption of opioids, especially heroin. IDU has rapidly spread in Asia, as heroin is increasingly produced in the region while opium has become less available, new overland heroin trafficking routes have developed, and other injectable drugs have been introduced. Sharing needles and other injecting equipment is a very effective mode of transmission of the HIV virus. In combination with the spread of IDU, this has resulted in injecting being the primary mode of HIV transmission in many Asian countries. Today's worldwide HIV/AIDS epidemic is far worse than was predicted 10 years ago. UNAIDS and the World Health Organisation (WHO) estimate that at the end of 2000 more than 36 million people are living with HIV or AIDS [4]. The worst affected region is Africa, with an estimated 3.8 million adults and children newly infected during 2000, bringing the total number of people with HIV/AIDS to 25.3 million. The Asian region comes second, with an estimated 900 000 adults and children newly infected with HIV during 2000. At the end of 2000, 6.4 million adults and children in the region were living with HIV/AIDS, 5.8 million in South and Southeast Asia, and 640 000 in East Asia and the Pacific. Figure 1 shows the spread of HIV in Asia between 1984 and 1999. In the Asian region, Cambodia, Myanmar and Thailand had adult HIV prevalence rates of 2% or more by the end of 1999 [5]. Although India's adult prevalence rate seemed relatively low at 0.7%, an estimated 3.7 million Indians were living with HIV/AIDS at the end of 1999, more than in any other country in the world except South Africa. Moreover, whereas some states of India show very low levels of HIV infection, others have reached adult prevalence rates of 2% and above. In China, HIV infection rates remained relatively low during 1999 with an adult HIV prevalence rate estimated at 0.07%.Fig. 1: Spread of HIV over time in Asia, 1984-1999. HIV prevalence in the adult population (%).This paper reviews recent data on the twin epidemics of IDU and of HIV infection in the Asian region. Included in this review are the countries in South, Southeast, and East Asia, ranging from Iran in the West to Indonesia and Japan in the East. The newly independent Central Asian republics are not included in this review as they have been included in a recent review of HIV/AIDS epidemiology in Eastern Europe [6]. The need for strengthened surveillance and control of injecting drug use and HIV infection in the Asian region is discussed. Data sources The information presented in this paper was compiled from a variety of sources. For the level and trends of IDU, these sources include reports by National AIDS Programs, workshop and field-trip reports available to WHO, UNAIDS and UNDCP, and published papers and articles. The methodologies used to estimate the level and trends of IDU vary considerably, from expert opinion to more scientifically rigorous methods. The validity of the estimates of the number of injecting drug users therefore varies, precluding direct comparisons between countries. Information on HIV infection in injecting drug users and other groups was drawn from the June 2000 release of the HIV/AIDS surveillance database maintained by the International Programs Division of the US Bureau of the Census' Population Division [7]. The database contains data from published papers and articles, conference abstracts, and country surveillance reports. A search of the MEDLINE and AIDSLINE databases was performed using AIDS, HIV, IDU and Asia as keywords, and recent issues of general medical journals and journals specialized in HIV/AIDS and drug use were reviewed. Abstracts of the recent international AIDS conferences were reviewed: the 12th World AIDS Conference, Geneva, June 1998; the 5th International Congress on AIDS in Asia and the Pacific, Kuala Lumpur, October 1999; and the XIII International Conference on AIDS, Durban, July 2000. Spread of injecting drug use in Asia Determinants and characteristics of IDU epidemics in Asia The rapid rise in IDU in Asia is recent, with a notable increase starting in the 1980s [8], and is linked to several factors. Injecting drug users use a number of substances for injection, including opiates such as heroin and buprenorphine, tranquilizers and barbiturates, and to a lesser extent amphetamines and cocaine. In Asia, the most widely injected group of substances is opiates. Until the 1960s, local consumption was confined to opium since heroin was not available locally. With increased heroin refinement in the late 1960s, heroin use spread. The increase of injecting is intimately linked to the widespread availability of injectable opiates in the region. Asia is home to the world's two major opium-producing areas. The 'Golden Triangle' is situated in Southeast Asia in the bordering areas of Myanmar, Thailand, and the Lao People's Democratic Republic. The 'Golden Crescent' is where Pakistan's northwest frontier meets the Badakhshan area of Afghanistan and the Baluchistan area of Iran. In fact, two countries in these areas are the world's two largest producers of opiates. Afghanistan became the world's largest producer of opium poppy in 1999, while Myanmar is the second largest producer of illicit opium and heroin, accounting for approximately 80% of the total production of Southeast Asian opium [9]. Not only is the region a major producer, opium is increasingly processed and transformed into injectable heroin locally. This development has increased the availability of pure heroin among the local population. For example, in Myanmar, where IDU is increasing, injectable heroin is refined in factories of remote mountainous regions in the country's north and east [10]. IDU is also increasing along the overland heroin trafficking routes [11], especially from the Golden Triangle region. Spread of injecting is also associated with mobility and migration. Mobile populations such as truck drivers travelling north from the Golden Triangle have brought heroin with them to other regions. In Myanmar, the Shan and Kachin states have particularly high rates of heroin users, many of them young migrants who work in the jade and ruby mines. Young people from Myanmar are also known to leave their villages to escape poverty to the neighboring countries of India and China, where they contribute to the spread of heroin use and injection practices [12]. In Chennai, India, injection of heroin has been linked to migrants from Manipur [2]. Level and trends of IDU in selected countries Table 1 summarizes recent data available on the estimated number of injectors for most Asian countries. Some countries show a continuing but stable trend in both heroin use and drug injection, but other countries have shown a phenomenal increase in these practices in recent years [13].Table 1: Estimated number of injecting drug users in Asian countries.While Afghanistan is a major opium producer, little is known about the local use of heroin by injection. In Bangladesh, an estimated 25 000 people inject drugs [14], with around 7600 male injecting drug users in the capital Dhaka [15]. Injecting is thought to have been introduced in Dhaka around 1995 [16]. In a 1999 survey, the most commonly injected drug was buprenorphine, while only 2% reported injecting heroin [16]. In Iran, an estimated 710 000-1 000 000 people may consume opiates [17], with a minority of them injecting. In Pakistan, there were an estimated 1 million heroin users by 1990, but few reports of injection. However, more recent reports from Karachi suggest that about 25% of heroin users inject [2], and injecting has also been identified in areas other than Karachi [18]. In India, the state of Manipur in the northeast borders with Myanmar and is on a major drug distribution route to other parts of India and Nepal. At the beginning of the 1980s there was little use of heroin, but heroin smoking and then injecting spread and, by the end of the decade, there were an estimated 15 000 heroin injectors in Manipur, particularly along the main highway leading from Myanmar. The prevalence of IDU in Manipur ranged from 0.2% in remote villages to 1.3% in villages close to the national highway [19]. Injecting has been spreading in several other areas and cities in India, and has involved other substances besides heroin. Injecting was reported in many areas of Chennai by 1990, involving buprenorphine as well as drug 'cocktails'. A recent rapid assessment found that opioids and other injectable drugs, including buprenorphine, are widely used in Delhi, Calcutta, Chennai and Mumbai [20]. The drug-using population is almost entirely male, except in Manipur and Misoram where female admissions to drug treatment facilities were 5 and 10%, respectively [21]. In Thailand, in a period of 20 years between the late 1950s and 1970s, there was a transition from opium smoking to heroin smoking and then injecting. There are now reports of drug injecting from all major cities of Thailand as well as rural areas, and the category of "ever injected" is reported by more than 60% of those in drug treatment. It is estimated that there are approximately 40 000 opiate users in Bangkok. Worryingly, the prevalence of IDU among military conscripts in northern Thailand increased from 1.1% in 1991 to 4.2% in 1997 [22]. Like Thailand, Myanmar saw an expansion of opium production and heroin refining during the Vietnam war. Myanmar became a major consumer of heroin from the mid-1970s and heroin injecting replaced smoking. The number of dependents may be as high as 500 000 with a large proportion injecting. There are reports of injecting from cities, mining areas, among fishermen, tribal groups and insurgent armies involved in the heroin trade. In China, the registered number of drug users has steadily risen from approximately 140 000 in 1990 to 861 000 in 2000 [23-25], as shown in Fig. 2. It is unclear how much of this increase may be due to increased coverage of the registration. However, unofficial estimates have placed the number of drug users several times higher [26]. Of these drug users, an estimated 600 000-1 000 000 inject [14,27]. Drug use is primarily a male phenomenon in China. However, it is noteworthy that, unlike most other countries in the region, there is a growing proportion of female drug users entering treatment facilities. Opium smoking has been largely replaced by heroin injection, albeit in different population groups. Heroin is the drug most commonly injected, although methamphetamine is becoming more popular [28].Fig. 2: Number of registered drug users in China, 1991-2000. (Data from [23-25].)In Indonesia within the past 5 years, drug abuse has emerged as a serious concern. Reports have estimated the number of users (both injecting and non-injecting) at 1.5 million [29,30]. During 2000, a rapid assessment of HIV infection in eight cities was conducted through a consortium of international and local agencies [31]. The data revealed a relatively young population of users, with 70% younger than 24 years of age. Between 80 and 90% of users were male and many had a relatively short drug taking history of 1-2 years before resorting to injecting [32]. The study further identified heroin and crystal methamphetamine as the drugs of choice. Injecting has remained relatively rare in both Cambodia and Laos despite their sharing borders with countries with high numbers of injectors, as well as in the Philippines. Little information is available on the situation in Sri Lanka and the Maldives. Spread of HIV among injecting drug users in Asia IDU is the first mode of transmission of HIV in many Asian countries with concentrated HIV epidemics, including China, Malaysia, Nepal, and Vietnam. Countries with extensive heterosexual spread of HIV, including Thailand, Myanmar and India, also have high rates of HIV prevalence in groups of injecting drug users. Determinants and characteristics of HIV epidemics among injecting drug users Sharing needles and other injection equipment is a very efficient way of transmitting HIV, as HIV is inoculated directly in the bloodstream and HIV may remain viable for up to 4 weeks in syringes [33]. HIV epidemics can occur very rapidly among injecting drug users when sharing is widespread. The level and pattern of sharing equipment among injecting drug users determines the transmission risk in IDU communities in countries. Figure 3 shows some examples of the extremely rapid spread of HIV infection among injecting drug users in China, India, Thailand and Vietnam. The long-standing epidemics in Thailand and Manipur, India also illustrate that the HIV prevalence levels remain high after they have peaked.Fig. 3: HIV prevalence among injecting drug users in Manipur state (India), Bangkok (Thailand), Ha Noi (Vietnam), and Kaiyuan (China), 1985-1999. (Source: National AIDS Programmes.)Besides the rapid spread, another important characteristic of HIV epidemics among injecting drug users is their discrete nature. Epidemics take place in geographically defined communities of IDU, linked by their sharing practices. Table 2 shows for China and Vietnam that, while IDU communities in some sites have high rates of HIV infection, others were still at very low levels by 1998/1999.Table 2: Seroprevalence in injecting drug users in selected sites in China and Vietnam, 1995-1999.Level and trends of HIV epidemics among injecting drug users in selected countries In China, HIV first appeared in injecting drug users in Yunnan province, bordering Myanmar. According to the Chinese Ministry of Public Health, 80% of all reported HIV infections and 60% of all reported AIDS cases in the country are found in Yunnan. No HIV was detected in drug users in seven provinces outside Yunnan in a 1993 study, although injecting was present in several provinces [34]. In 1998, sentinel surveillance among injecting drug users found HIV infection in 17 out of 19 national surveillance sites [35]. This rapid spread of HIV-1 infection has been linked to diffusion of injecting along the drug trafficking routes from the northern areas of Myanmar into China [11]. The occurrence of an explosive epidemic in Guangxi province, bordering Yunnan province and Vietnam, has been well documented. HIV infection among injecting drug users was found only in the Baise site in 1996, but had spread to the six other sites in the province by 1999 [36]. A 1996-1997 study of injecting drug users in Guangxi province showed an HIV prevalence of 40%. Most injecting drug users had used drugs for 3 years or less, and sharing of injection equipment and unprotected sex were significantly associated with HIV-1 infection [37]. In Vietnam, HIV has spread in all regions [38]. A large epidemic emerged in the northern region in recent years (see Table 2), with 66% of injecting drug users reporting sharing injection equipment in Quang Ninh province in 1999 [39]. In 1994, the highest HIV prevalence rates worldwide were found among Myanmar injecting drug users. The results showed that 91% of addicts were infected with HIV in the city of Myitkyina, capital of Kachin State on the Chinese border, as were 84% in Mandalay and 74% in Yangon [7]. Particularly hard-hit are ethnic minorities who live in the border areas where opium is produced and transformed into heroin, and that are also subject to civil unrest. In 1996-1997, 61% of heroin injecting drug users in the northern city of Myitkyina reported sharing needles [40]. HIV prevalence rates among injecting drug users in India range from 2% in Calcutta State to 44.5% in Delhi. Manipur has the highest HIV infection rate among injecting drug users in India; it rose from almost zero in 1988 to around 80% just 5 years later and has remained at this high level since [7]. Among military recruits in northern Thailand, a growing proportion of HIV infections is due to injecting [22]. A cohort study in northern Thailand estimated that the HIV incidence rate among injecting drug users was stable at 8 per 100 person-years between 1989 and 1997 [41]. This is in sharp contrast with a reduction in HIV prevalence in military recruits in Chiang Rai [42]. In Nepal, HIV prevalence surveys had indicated low HIV prevalence of 1.6% in 1991 and 0% in 1994 [43,44]. However, a nationwide seroprevalence survey in 1999 found high levels of HIV infection, with 40% in Kathmandu and a national average of 50% [45]. This is consistent with findings from rapid assessments in Nepal, where 50-65% of injecting drug users admitted to freely sharing injection equipment [46,47]. In a 1999 study among 200 injecting drug users in Lahore, Pakistan, no cases of HIV infection were identified, although 64% said they shared equipment [18]. An earlier study among drug users at a psychiatry center in Lahore had revealed an HIV prevalence of 11.5% [48]. In Iran, major outbreaks of HIV infection among injecting drug users have occurred in two prisons in 1997-1998, involving several hundreds of cases. In Malaysia, 'professional injectors' are associated with IDU. Known as 'street doctors', they inject multiple customers with the same uncleaned equipment. A related practice has also been reported at the village level, where heroin is prepared in a syringe and then used on multiple occasions among many people. Sentinel surveillance among injecting drug users at drug rehabilitation centers showed HIV prevalence ranging from 10.5 to 35.5% in different regions in 1998 [7]. In Bangladesh, sentinel surveillance among in-treatment injecting drug users in 1998 indicated 2.5% HIV prevalence, while in 2000 it was 0.2% among in-treatment injecting drug users and 0-1.4% in injecting drug users attending a needle exchange program [16]. A very high proportion of injecting drug users (93-96%) reported sharing injecting equipment in the past week [16]. In Indonesia, a rapid assessment has revealed sharing and exchanging of injecting equipment to be very common among injecting drug users, while no HIV prevalence data are as yet available from the surveillance system [31]. In the Philippines, HIV infection among injecting drug users remains rare, with an HIV prevalence >1% recorded at only one sentinel surveillance site in 1997, and only 6% of injecting drug users reporting the sharing of equipment [49]. Although HIV prevalence rates remain very low in Japan, with an estimated overall prevalence rate among adults of 0.02% [5], a high level of sharing at 88% was reported in a study of non-hospitalized injecting drug users in 1996 [50]. Spread of HIV from injecting drug users to their sex partners and the general population The epidemics in Cambodia, Myanmar, Thailand, and the southern states of India are generalized epidemics fueled by heterosexual transmission [5,51]. Early studies in Thailand suggested that the epidemics among injecting drug users and among heterosexuals were unrelated. A subtype B epidemic among injecting drug users starting in 1988 preceded a separate subtype E epidemic driven by heterosexual transmission starting in 1989 [52,53]. However, in recent years, subtype E has also spread among injecting drug users in Thailand, especially among younger injecting drug users [54]. There is also evidence from Manipur, India that generalized heterosexual epidemics can originate from an initial group of infected injecting drug users. In one sentinel surveillance site in Manipur, HIV prevalence in pregnant women was 2.3% in 1999, while in injecting drug users it has been around 80% since 1993 [55]. Molecular studies indicate that the recent heterosexual spread is related to the earlier IDU epidemic [56,57]. In many countries, injecting drug users have reported trading sex for drugs, having several sexual partners and not using condoms regularly [57-59]. In 1997, in a Myanmar study, 14% of injecting drug users reported multiple sex partners [60]. In a study of HIV-infected male injecting drug users and their wives in Manipur, India, only 15% of these couples reported regular condom use, and 45% of the wives were also infected with HIV [61]. In a rapid assessment in Nepal, 72% acknowledged premarital sex practices with multiple partners and approximately 65% did not use condoms. One-half of the group admitted to unsafe sex and slightly more than one-half had no knowledge of sexually transmitted diseases and their risk of infection. The Lahore, Pakistan study has also suggested transmission links to the general population through unprotected sexual contact [18]. Nearly one-half of the study sample reported having sex with sex workers with minimal or irregular use of condoms. In Indonesia, there was also significant evidence of multiple sex partners and low rates of condom use among injecting drug users [31]. These findings have also been supported by a study of a small group of drug users admitted to a treatment facility in West Java [30]. In China, HIV prevalence was over 6% in one sentinel site of sex workers in 1999 [62]. Among spouses of HIV-infected injecting drug users in Yunnan province, HIV prevalence increased from 3.1% in 1990 to 12.3% in 1997 [63]. Discussion In many countries in Asia, a strong trend towards increasing use of injectable drugs is apparent. This trend is strongly linked to the consumption of opiates, especially heroin, and it is affecting younger populations. HIV transmission via IDU has already generated large epidemics among injecting drug users in many countries of the region, while other countries have hitherto been relatively spared. The twin epidemics of IDU and HIV present interrelated challenges and opportunities for prevention. Comprehensive programs need to be developed to address the problem of drug dependence and related behaviors that are linked to the spread of HIV. Efforts should include primary prevention of drug use, by addressing social and economic changes that are driving a greater number of young people to resort to drug use as a means of coping with their daily problems. For those already using drugs, prevention of progression to injection should be a focus. There is limited experience on how this can best be achieved, but targeted educational programs on the risks of injection can have an effect [64]. For those who started injecting, clean needles, bleach, and syringe exchange services need to be available and accessible through outreach services. With qualified staff and peer educators, these services can provide HIV education and counseling regarding condom use and sexually transmitted infections, can distribute condoms, and can encourage HIV testing and entering into drug treatment. While the effectiveness of harm reduction programs has been demonstrated in a variety of settings [65], governments and partners fail to fund or support their wide-scale implementation [66,67]. For those willing to be treated for their dependence, effective treatment programs should be community based, and available where people need them, providing a range of options, including long-term agonist pharmacotherapies and psychosocial support. Finally, to those who are HIV infected and chronically dependent, family support services, affordable clinical and home-based care, essential legal and social services and psychosocial support are needed, in addition to treatments for their dependence. This continuum of options is likely to reach most risk populations and can change more effectively the course of the epidemic in Asia. Countries should focus on appropriate interventions within this continuum. Within the constraints of existing resources, countries will also need to re-examine current policies that minimize the opportunities for these options in favor of a heavy reliance on punitive measures and medium-term institutional-based treatment without follow-up. The level and the trends over time of IDU are difficult to study due to limitations with data collection and reporting systems. The majority of countries in Asia cannot provide reliable estimates of the magnitude of IDU. The estimated number of injecting drug users needs to be interpreted with caution, taking into account the underlying methodology. Reliable information on the level of risk behavior is also hard to come by, and has not been collected in a standardized fashion that will allow comparison between areas and assessing trends over time. Hence there is a need for improved epidemiological tools. UNDCP has recently launched its 'Global Assessment Programme' (see www.undcp.org) to improve drug abuse epidemiology, including IDU. WHO has developed a drug abuse epidemiology guide [68] as well as rapid assessment guides [69,70]. Family Health International has developed tools for repeated behavioral surveillance surveys [71]. Behavioral surveillance aims to include a representative sample of injecting drug users, and is well suited to inform the design of interventions among injecting drug users and to monitor their impact. For example, an increase in condom use in injecting drug users with sex workers from 8 to 18% over a 1-year period was documented in Bangladesh, although the increase was not statistically significant [16]. In Asia, HIV has badly affected large numbers of injecting drug users as well as the communities around them. With a population of over 2 billion in just India and China, Asia may have the greatest potential for further IDU-driven growth of the HIV epidemic. As strategies are well known, it is imperative that more harm be prevented, by investing in effective, large-scale programs. Acknowledgments The authors would like to acknowledge the contributions of Paul Griffiths (UNDCP, Vienna), Kamran Niaz (UNDCP, Tehran), Jennifer Hillebrand (WHO, Geneva), Anindya Chatterjee (UNAIDS, Bangkok), and Ofelia Monzon (AIDS Society of the Philippines, Manila). They would like to thank Neff Walker (UNAIDS, Geneva) and Tim Brown (East-West Center, Bangkok) for helpful comments.