“May you live in interesting times” is actually not a Chinese curse.1 However, HIV/AIDS is clearly creating interesting times in China. The initial outbreak of HIV occurred among injecting drug users (IDUs) in Yunan province in 1989.2 HIV spread along the drug trafficking routes from the Golden Triangle through southern China and northern Vietnam toward export cities on the Pacific. HIV also spread throughout the country, and by 2002, there were HIV seropositive IDUs in all 31 provinces and autonomous regions.3 There was also substantial transmission among plasma donors in Henan province. Initial attempts at controlling HIV in China were hampered by many factors, including a weak public health infrastructure, differences in development between the eastern and western regions, the difficulties in working with the large “floating population” of internal migrants, as well as relatively severe stigmatization of the disease. The Severe Acute Respiratory Syndrome (SARS) epidemic, however, taught a severe and important lesson. Even if only a small percentage of the population is infected, infectious diseases can be a threat to economic growth and national security.4 A new national policy of “Four Frees and One Care” (free treatment, free voluntary counseling and testing, free prevention of mother to child transmission and free schooling for AIDS orphans, and provision of social relief for HIV patients) was announced on World AIDS Day 2003. In the most recent estimates, there are 650,000 people with HIV in China,5 though the actual numbers may be considerably greater than the estimates. There are also high rates of syphilis and other sexually transmitted diseases that facilitate HIV transmission.6
China has committed to large-scale implementation of HIV prevention programs for IDUs and commercial sex workers (CSWs) to prevent its current “concentrated” HIV epidemic (among IDUs and CSWs) from becoming a “generalized” epidemic (where HIV infection is transmitted primarily through heterosexual activity and 1% or more of the adult population is infected). The planned interventions include community outreach, methadone maintenance treatment, and syringe exchange for IDUs.7–9 There is considerable evidence that these interventions can be very effective in reducing HIV transmission among IDUs.10 There are also interventions to reduce sexual risk behavior among IDUs, including providing information, skills training in using condoms, safer sex negotiation skills, and HIV counseling and testing (to reduce transmission behavior by HIV seropositives). Reducing sexual risk behaviors of IDUs, however, is considerably more difficult than reducing injection risk behavior.11
The study by Lau and colleagues (article in this issue) of HIV risk behavior among male IDUs in Sichuan province, China, provides an opportunity to consider HIV prevention to date for IDUs in that area. The main findings of the study included moderate to high levels of injecting and sexual risk behaviors among the subjects. For example, 16% reported injecting with a syringe used by others at last injection, 36% reported injecting with syringes used by others during the 6 months before the interview, 19% had visited a commercial sex worker in the month before the interview, and 74% of these did not use a condom in their last episode with a commercial sex worker.
The study also found moderately strong associations between injecting and sexual risk behaviors, e.g., injecting with syringes used by others was associated with having sex with a commercial sex worker, odds ratio, OR = 1.81; injecting with syringes used by others was associated with having sex with a nonregular partner, OR = 1.59; and with having unprotected sex with a nonregular partner; OR = 1.57; and passing on used syringes to others was associated with having unprotected sex with a nonregular partner, OR = 1.28. Approximately 15% of the sample was “double risk,” engaging in both injecting and sexual risk behaviors.
These levels of injection and sexual risk behaviors present a clear danger of rapid HIV transmission among IDUs in Sichuan, followed by transmission to large numbers of noninjecting sexual partners, followed by self-sustaining heterosexual transmission. Analyses of rapid HIV transmission among IDUs suggest that the most important factor is the pattern of sharing among IDUs. (Does sharing occur in “random mixing” types of settings such as shooting galleries, public injection sites, or with injection equipment supplied by dealers vs. sharing confined to within small, relatively stable groups of close associates?).12,13 Modeling of HIV transmission from IDUs to the “general” population suggests that the critical factors are how frequently IDUs engage in unprotected sex with CSWs, the frequency of unprotected sex between sex workers and males in the general population, and the prevalence among sex workers of other sexually transmitted diseases that would facilitate HIV acquisition and transmission.14
Lau et al. do not present data on sharing patterns or sexually transmitted diseases for their subjects, so that it is difficult to assess the likelihood of either very rapid transmission among IDUs or large-scale HIV transmission from IDUs to sex workers. The report does raise a number of other issues that need to be addressed in considering the potential transition from a concentrated to a generalized HIV epidemic.
A number of HIV prevention interventions had been implemented for IDUs in Sichuan at low to moderate levels of coverage:
1. 31% of subjects reported that they had ever had an HIV antibody test,
2. 10% reported that they had had an STD checkup in the previous 12 months,
3. 37% reported that they had received HIV-STD “general services” in the previous 12 months,
4. 20% reported that they had received services related to syringe rinsing in the previous 12 months,
5. 22% reported that they had received syringe exchange services in the previous 12 months, and
6. 9% reported that they had received methadone maintenance services in the previous 12 months.
There are not-yet-clear standards for the desired “coverage” (percentage of IDUs to be reached) for different interventions. (A special issue of the International Journal of Drug Policy will be devoted to the issue of coverage.) Despite the lack of definitive standards, there is little doubt that the interventions in Sichuan need to be scaled up. As the authors note, expansion of the current interventions is “urgently” needed.
Trends Over Time
One of the more interesting findings in the report was the reduction in membership in the “double risk” group (IDUs reporting both unsafe injection and unsafe sexual risk behaviors) from 2003 to 2004. This reduction was substantial, from 18.6% of subjects in 2003 versus 11.7% in 2004, and highly significant in the multivariate analysis, adjusted odds ratio, AOR for 2004 survey = 0.36, 95% confidence interval, CI 0.28–0.46, P < 0.001. It would be important to determine the possible reasons for this reduction and, in particular, whether the reduction was associated with any increase in intervention services.
Knowledge of and Supplies for Risk Reduction
Despite the modest coverage of the interventions, HIV/AIDS knowledge appeared to be fairly good among the subjects. Eighty-three percent of the subjects correctly answered 4 or 5 of the questions on the 5-question knowledge test. Given the limited coverage of the interventions in Sichuan, it is very likely that HIV/AIDS information diffused within the IDU population, so that IDUs who were not in contact with the interventions may still have learned important information about HIV/AIDS.
If there is sufficient knowledge about HIV/AIDS among IDUs in Sichuan, then the relatively high rates of risk behavior are probably due to problems with the availability of necessary materials (clean needles and syringes, condoms) for risk reduction. Does the syringe exchange program distribute large numbers of clean needles and syringes? Does it encourage “secondary exchange,” where an IDU who attends the program can also exchange for others who do not personally attend the exchange? Do the various programs distribute condoms to IDUs, and if so, do they distribute large numbers of condoms? And do the activities of law enforcement officials create barriers to IDUs obtaining and using clean syringes and condoms? (Similarly, do the activities of law enforcement officials create barriers to CSWs obtaining and using condoms.)
Although Lau and colleagues do not discuss possible barriers to IDUs obtaining and using clean syringes and condoms, removing any such barriers may be a critical issue for HIV prevention in Sichuan. Similarly, any barriers to CSWs obtaining and using condoms should be addressed.
One of the most interesting aspects of the Lau et al. study is the variation among their 6 sites. This is dramatically illustrated in their analysis of factors associated with belonging to the “double risk” (both injecting and sexual risk) group:
1. site 1, AOR = 1.0 referent
2. site 2, AOR = 1.70 (95% CI 1.06–2.72)
3. site 3, AOR = 3.58 (95% CI 2.36–4.95)
4. site 4, AOR = 3.19 (95% CI 2.05–4.95)
5. site 5, AOR = 3.61 (95% CI 2.34–5.56)
6. site 6, AOR = 15.12 (95% CI 9.37–24.41)
This is a remarkably wide range of adjusted ORs (1.0–15.12) for a modest number (6) of sites. Note also that with sites entered into the model, participation in interventions was not associated with membership in the double-risk group. In terms of the potential for rapid spread of HIV through the sharing of drug injection equipment and then through heterosexual activity with noninjecting sexual partners, the important differences in Sichuan would appear to be among the sites. Understanding the variation among the sites may be critical to implementing effective HIV prevention for IDUs and their sexual partners in Sichuan.
HIV Prevention for IDUs as a Community Process
There are a number of reasons to consider effective HIV prevention for IDUs as a community process. Information about HIV/AIDS can diffuse within the population of IDUs. New social norms that encourage safer injection and safer sex can develop and influence the behavior of individuals. Supplies needed for safer behavior (clean needles and syringes, condoms) can also diffuse within the IDU population without requiring all individuals to attend specific programs. Cooperation from law enforcement officials can reduce barriers to IDUs obtaining clean needles and syringes and condoms (and reduce barriers to sex workers obtaining and using condoms).
Considering HIV and STD prevention for drug users as a community process raises community-level research questions. What levels of “coverage” are needed for different interventions? What are the social change processes (as opposed to individual change processes) that lead from intervention implementation to reductions in HIV transmission? How can altruistic motivation among IDUs—to protect their peers and their sexual partners—be maximized? How can interventions be implemented so that they support each other and possibly have synergistic effects? How can prevention of injection-related transmission be best integrated with prevention of sexual transmission? In particular, we need better methods for multisite analyses—why is HIV transmission under control in one area and out of control in another? How can interventions that were effective in specific cultural settings be successfully adapted for different cultural settings? These community-level research questions are probably best studied through ethnographic research, detailed case histories of HIV epidemics in different sites, and mathematical modeling.
While research is urgently needed on these community change processes, there is no justification for delaying large-scale implementation of HIV prevention programs for IDUs that already have evidence of effectiveness.10,15 Resources, including money and trained staff, for preventing HIV infection among IDUs and their sexual partners, are limited in many different countries. The most important limited resource, however, is the remaining time available for preventing additional HIV epidemics among IDUs and their sexual partners.
2. Zeng Y, Fan J, Zhang Q, et al. Detection of antibody to LAV/HTLV-III in sera from hemophiliacs in China. AIDS Res 1986; 2:S147–S149.
3. Feng C, Des Jarlais DC. HIV among drug users in China. Science 2002; 298:1171.
4. Shao Y. AIDS epidemic at age 25 and control efforts in China. Retrovirology 2006; 3:87.
5. UNAIDS/WHO. AIDS Epidemic Update: December 2006. Geneva: Joint United Nations Programme on HIV/AIDS (UNAIDS) and World Health Organization (WHO), 2006.
6. Chen Z-Q, Zhang G-C, Gong X-D, et al. Syphilis in China: Results of a national surveillance programme. Lancet 2007; 369:132–138.
7. Isbell M. The money trail: financing the global AIDS response. Frontline: The age of AIDS, 30 May 2006.
8. Kanabus A. HIV & AIDS in China: AVERT 2006.
9. Russell S. China finally taking steps to fight its HIV problem: Methadone clinics, condoms in hotels, free testing offered. San Francisco Chronicle. 2006.
10. Committee on the Prevention of HIV Infection among Injecting Drug Users in High Risk Countries. Preventing HIV Infection Among Injecting Drug Users in High Risk Countries: An Assessment of the Evidence. Institute of Medicine of the National Academies, 2006.
11. Des Jarlais DC, Semaan S. Interventions to reduce the sexual risk behavior of injecting drug users. Int J Drug Policy 2005; 16S:S58–S66.
12. Des Jarlais DC, Friedman SR, Choopanya K, et al. International epidemiology of HIV and AIDS among injecting drug users. AIDS 1992; 6:1053–1068.
13. Vickerman P, Hickman M, Rhodes T, et al. Model projections on the required coverage of syringe distribution to prevent HIV epidemics among injecting drug users. J Acquir Immun Defic Syndr 2006; 42:355–361.
14. Saidel TJ, Des Jarlais DC, Peerapatanapokin W, et al. Potential impact of HIV among IDUs on heterosexual transmission in Asian settings: The Asian Epidemic Model. Int J Drug Policy 2003; 14:63–74.
15. Des Jarlais DC. Preventing HIV transmission among IDUs and from IDUs to non-injecting sexual partners in Sichuan, China. Sex Transm Dis. In press.