Needle exchange programmes (NEP) have been shown to be an effective measure to reduce unsafe drug injection and prevent HIV transmission among injection drug users (IDUs) . Programmes have been established in many countries including China, which began its first NEP as a trial effort in 2000–2002 . National NEP have been gradually scaled up, especially since 2004, and by the end of 2005 included 92 sites in 12 provinces (Fig. 1). Over two-thirds (70.7%) were located in south-western and north-western provinces, where the HIV epidemic among drugs users is most severe, notably Guangdong, Guangxi, Sichuan, Yunnan, and Xinjiang. Of the 92 NEP operating, 44 (47.8%) were primarily funded by domestic government finances and 40 (43.5%) by international funding, such as the UK Department for International Development. The remainder were funded by both domestic and international funds. As the programmes progress, however, they will be entirely funded by government monies.
Little evaluation has been done of the NEP since the initial trial for efficacy. The purpose of the present study was to evaluate the NEP operating in China, with a specific focus on two provinces heavily affected by the HIV epidemic among drug users: Sichuan and Guangxi.
Fifteen NEP in two provinces, six in Sichuan and nine in Guangxi, were selected to study the operational and individual factors influencing attendance at NEP. The 15 NEP were all governed by the local Center for Disease Control and Prevention (CDC) and operated during normal working hours. Each NEP has a fixed physical site for distributing new needles to and collecting used needles from primary exchangers (i.e. drug injectors who directly get needles from the fixed needle exchange site located in the local CDC). The NEP also employed a certain number of peer educators (or peer outreach staff) who periodically obtained new needles from the CDC to distribute to IDUs in the community. Peer educators, in turn, recruited current or former drug injectors who got new needles from and gave in dirty needles to peer educators, termed ‘secondary exchangers’. The peer educators engaged the fixed needle exchange site located in local CDC, whereas the secondary exchangers did not have a direct relationship with the fixed needle exchange site located at the local CDC. All needles were distributed free of charge. Data were collected from NEP managers, attendees, IDUs in detoxification centres and public security officials.
The managers of the 15 NEP were approached directly and asked to complete a questionnaire about the NEP mode of operation, needle turnover, the number of peer educators, and the number of attendees. The managers were then interviewed in more detail as to how the NEP operated, including how peer educators were selected, whether the NEP provided additional injecting paraphernalia, and whether other services were offered by the NEP.
Public security officials
Fifteen public security officials (police), one each from 15 NEP sites, were also interviewed about their opinions and attitudes towards drug users and NEP, including whether they thought it was legitimate to arrest drug users who attended NEP.
Needle exchange programme attendees
IDUs attending NEP were surveyed to identify factors associated with the accessibility of services provided by the NEP. Peer educators working at each site were the first to be invited to answer the questionnaire. They were each then asked to recommend several secondary exchangers also to respond to the questionnaire. IDU were asked for demographic data, drug use behaviours, how they knew about the NEP, whether they had been arrested or feared arrest when using the NEP, their needle preferences (the type of needle used in NEP versus other types), the number of needles they needed and how many they were able to get from the NEP, and where they usually obtained their needles.
IDUs interned in local detoxification centres were selected through random sampling to identify factors associated with attendance. Attendees were asked the same questions described above for attendees recruited from the NEP. Non-attendees were asked for demographic data, drug use behaviours, whether (and how) they knew about the NEP and why they did not use NEP services.
Data collected from NEP managers and public security officials were used to summarize the operational features of the NEP. Quantitative data were coded and summarized quantitatively. Quantitative data were summarized as mean ± standard deviation for continuous variables and frequency and percentages for categorical variables. Needle turnover was compared during the study period and since opening the NEP to identify any trends in distribution. The data were then used to identify factors associated with needle turnover, a proxy for coverage. Univariate analyses were first performed to identify factors associated with the outcome variables and those with a P < 0.1 were selected for multivariate linear modelling. Variables in the final model were also selected at α = 0.1.
Characteristics of peer educators and secondary exchangers were compared. For continuous data, comparisons were made using t-tests or non-parametric methods if data were not normally distributed. For categorical variables, data were compared by the chi-squared test or Wilcoxon's test.
Data collected from all IDUs were analysed to identify differences between attendees and non-attendees. Demographic and drug using characteristics were compared by t-test or chi-squared tests or non-parametric methods when data were not normally distributed. There data were input into a logistic regression model to identify factors associated with using NEP services.
Ethical approval to conduct the study was granted by the institutional review board (IRB) of the National Centre for AIDS/STD Control and Prevention.
Among the 15 officials interviewed, only three mentioned that drug addiction was also a kind of disease, whereas the rest emphasized that drug use was illegal. Only one leader firmly opposed NEP, six were ambivalent, and eight were supportive, three of whom regarded NEP as still consistent with drug prohibition policies. Four officials thought that using NEP services was legitimate grounds for arresting an IDU, nine opposed the idea, and two worried that it would arouse criticism from the community.
Recruiting peer educators
Each peer educator managed an average of 32 IDU (range 20–40). When employing peer educators, NEP managers generally stated that they sought sociable and responsible individuals who were seen as opinion leaders in the IDU community. Three NEP required educators to be non-users (either never used or not currently using) and one chose clinic doctors as educators to avoid any objection from the public security bureau. In contrast, three NEP deliberately employed active IDU to be educators, with one manager saying ‘if they are not using drugs any more, how can they still be good friends with other IDU?’. The other nine NEP had no preference for whether or not their educators were currently or had formerly used drugs.
A median of 6125 needles were distributed per month during the study period [interquartile range (IQR) 2800–11 020]. Since opening, the median number of needles distributed was 3787 per month, indicating that needle distribution had increased over the lifetime of the sites (P = 0.02 Wilcoxon pair marching). The overall needle return rate during the study period was 94.0%; the lowest rate was 74.5%, nine sites reported more than 90% return and one reported a return rate in excess of 100%. The cumulative needle return rate was 92.6% (IQR 88.6–95.5%).
Factors affecting turnover
Univariate analyses indicated that needle turnover was related to eight variables: police support (r = 0.60, P = 0.02); offering a range of syringe volumes (r = −0.52, P = 0.049); the availability of the most popular syringe sizes (r = −0.68, P = 0.005); the availability of other injecting paraphernalia (t = −2.44, P = 0.04); funding from an international agency (t = −2.61, P = 0.04); higher peer educator wages (r = 0.75, P = 0.001); the presence of methadone maintenance treatment services in the area (t = −2.28, P = 0.04); and the availability of additional services (r = 0.63, P = 0.01).
Variables that were significant at the α = 0.10 level were then included in the multivariate model to identify those that were the most significant factors influencing needle turnover. The analysis indicated that needle turnover was most influenced by paying peer educators a higher wage (β-coeff = 0.66, P = 0.001) and having police support (β-coeff = 0.42, P = 0.02).
Obtaining needles from needle exchange programmes
A total of 501 NEP attendees were recruited from the 15 NEP, including 108 peer educators and 393 secondary exchangers. In addition, there were 86 NEP attendees interned in detoxification centres, bringing the total number of attendees examined in this analysis to 587. Among the 108 peer educators, 79 were active IDU. Peers educators were older and had a higher education level than secondary exchangers, but were similar in terms of sex and ethnicity. Peer educators were also more likely to have been using drugs for longer than secondary exchangers, but there were no differences in terms of the length of time they had been injecting or the frequency of injecting (Table 1).
The 79 active IDU peer educators said they could obtain 95.6% of the needles they needed from their NEP, whereas only 66.2% of secondary exchangers said they could do so (P < 0.0001). At one NEP, secondary exchangers could get only 24.7% of the needles they needed from the NEP. Moreover, 20.4% of secondary exchangers actually got more needles than they needed. Based on local estimates of the size of the IDU population, approximately 7.3% of local IDU attended the NEP in their county/district.
More than half of peer educators (54.4%) and 66.9% of secondary exchangers reported that they had obtained needles from sources other than the NEP in the past month. Other sources included pharmacies, clinics, veterinary clinics, medical wholesale markets and other drug-using friends. The reasons given by peer educators for obtaining needles from other sources included ‘needed a hit but was not carrying a needle at the time’, ‘fear of arrest’, and ‘did not like the needles provided by NEP’. The reasons given by secondary exchangers for not being able to get enough needles included that ‘the peer educator had been arrested or was not contactable’, ‘the NEP only offered one needle per person per day’, and ‘unfair distribution of needles by peer educators’. Another important reason was that the number of secondary exchangers each peer educator managed was linked to their wages, so educators tried to manage as many secondary exchangers as possible, which meant that they had fewer needles for each IDU.
Some 14.7% of attendees continued to share needles, despite the availability of free needles from their NEP. The reasons cited for doing this included the belief that the person they were sharing with was a good friend and he/she was not infected with HIV.
Factors affecting needle exchange programme attendance
There were 324 IDUs in detoxification centres who participated in the survey. Among them, 238 IDUs (73.5%) had never used NEP services. This gave a total of 587 attendees and 238 non-attendees who were compared to identify factors associated with NEP attendance. There were no differences between attendees and non-attendees with respect to sex and ethnicity (Table 2). Attendees in Guangxi had a higher education level than non-attendees. Attendees in both provinces were older, had been injecting for longer and injected more frequently than non-attendees. In Guangxi, 14.7% of attendees and 43.7% of non-attendees had shared needles in the past month; in Sichuan these figures were 14.6 and 32.7%, respectively. The frequency of sharing needles in the past month was also lower among attendees (Table 2).
Most NEP attendees (82.3%) knew about their NEP through drug-using friends, 25.6% had seen pamphlets about NEP that had been distributed by the CDC, and 7.7% were introduced to the NEP by a policeman. More than half of non-attendees (55.0%) had never heard about the NEP. Participants who knew about the NEP were asked why they did not use them. Reasons cited included: they knew they existed but did not know where or with whom they could exchange needles; the NEP services were too far away; they did not like the needles provided by the NEP; and they were scared of being arrested.
The logistic regression model indicated that higher education [odds ratio (OR) 1.46, 95% confidence interval (CI) 1.14–1.86], more frequent drug injection (OR 1.16, 95% CI 1.02–1.32) and being divorced (compared with unmarried; OR 2.29, 95% CI 1.38–3.80) were associated with NEP attendance.
The majority of NEP in China were set up after 2003, and since 2004 they have developed rapidly in terms of the activity (measured by needle distribution and collection). Their operation models are, however, for the most part limited, with the majority run by CDC during normal working hours. Mobile clinics, vending machines and night services, all of which have demonstrated their efficacy in other countries [3,4], are rarely used in China and may contribute to the very low coverage observed in this study (an average of just 7.34%); IDUs cited distance and difficulty getting to the NEP as a barrier to accessing services. Flexibility in terms of hours of operation and types of services offered are critical components of successful NEP programmes [5,6].
Our results also suggest the need to reform the wage structure for peer educators. Needle turnover was related to peer educator wages, and secondary exchangers complained that they were not always able to get all the needles they needed from peer educators. This was especially true in places where peer educators were paid according to how many secondary exchangers they served, making these services less effective by stretching them over a greater number of individuals. The employment of more peer educators to cover a larger number of IDUs as well as the use of alternative operating models could see an increase in coverage and satisfaction with NEP services.
Coverage is also affected by awareness among IDUs. For example, more than half of non-attendees had never heard of the NEP services in their areas. As the vast majority of IDUs were introduced to the NEP through friends, IDUs need to be encouraged to spread the word. A missed opportunity for educating IDUs about NEP (and about the general risks of drug use) is within drug detoxification centres, but is rarely used . On the occasions when NEP has been promoted in detoxification centres, it has been shown to be effective . The use of detoxification centres for IDU education is controversial, however, even in the face of exceedingly high recidivism rates, and is thus highly dependent on support from local police. The data collected in this study indicated that police attitudes and support for NEP were key to the smooth implementation of the programmes and were associated with increased needle turnover. Health workers and NEP staff need to work closely with police to help increase their understanding of the benefits of harm reduction . In the current political climate this can be difficult, and even at the highest levels of government support for harm reduction has been limited, with police generally viewing NEP as encouraging drug use .
Even in areas where police may show passive support (inasmuch as they do not arrest NEP users), when local or central governments launch crackdowns on drug use, they need to fulfil their arrest quotas and NEP attendees are easy targets . This has previously been a problem with NEP in Guangxi , and may partly explain the high arrest rates seen in this study. In addition, it is probable that in some of the sites included in the study, police followed peer educators to distribution points and later arrested secondary exchangers; peer educators were less likely to be arrested by police and secondary exchangers were arrested typically after the peer educator had already left.
Despite limited coverage, the NEP included in this study were at least attracting the IDUs most in need, i.e. those who inject more frequently. This trend has also been seen elsewhere [12–16], and can limit the apparent benefits of NEP when measured by injecting frequency because a reduction in attendees' injecting frequencies may not be lower than the injecting frequency of non-attendees. This is another point that can stoke the scepticism of law enforcement officials as to the benefits of NEP .
NEP have been massively scaled up in China since the time this study was conducted, with 729 sites established by the end of 2006 . Central government has mandated that enough NEP be established to serve at least 50% of the IDU population by 2010 . The evaluation performed here indicates that coverage by the NEP already operating is extremely limited, and it is likely that although the service is reaching a larger number of counties, it may still not be able to reach a large enough proportion (modelled to be ≈60%) of IDUs to be truly effective at curbing the HIV epidemic . Strategies to overcome barriers to high coverage, including greater cooperation with public security, equitable payment structures for peer educators, and increased awareness of NEP among IDUs, are needed if the programme is to be maximally effective.
The authors gratefully acknowledge the help from staff at the Sichuan and Guangxi CDC who facilitated communication with local CDC, including Rongjian Li, Wenbo Huang (Guangxi) and Tian Li (Sichuan).
Sponsorship: This study was partly supported by the China Multidisciplinary AIDS Prevention Training Program with NIH Research Grant # U2R TW06918 funded by the Fogarty International Center, National Institute on Drug Abuse, and the National Institute of Mental Health.
Conflicts of interest: None.
1. Wodak A, Cooney A. Effectiveness of sterile needle and syringe programmes. Int J Drug Policy 2005; 16(suppl 1):S31–S44.
2. Wu Z, Lin P, Liu W, Ming ZQ, Pang L. Randomized community trial to reduce HIV risk behaviors among injecting drug users using needle social marketing strategies in China. In: 15th International AIDS Conference. Bangkok, Thailand, July 11–16 2004. LbOrC16.
3. Wood E, Kerr T, Spittal PM, Small W, Tyndall MW, O'Shaughnessy MV, et al. An external evaluation of a peer-run “unsanctioned” syringe exchange program. J Urban Health 2003; 80:455–464.
4. Miller CL, Tyndall M, Spittal P, Li K, Palepu A, Schechter MT. Risk-taking behaviors among injecting drug users who obtain syringes from pharmacies, fixed sites, and mobile van needle exchanges. J Urban Health 2002; 79:257–265.
5. Riley ED, Safaeian M, Strathdee SA, Marx MA, Huettner S, Beilenson P, et al. Comparing new participants of a mobile versus a pharmacy-based needle exchange program. J Acquir Immune Defic Syndr 2000; 24:57–61.
6. Brahmbhatt H, Bigg D, Strathdee SA. Characteristics and utilization patterns of needle-exchange attendees in Chicago: 1994–1998. J Urban Health 2000; 77:346–358.
7. Sullivan SG, Wu Z. Rapid scale up of harm reduction in China. Int J Drug Policy 2007; 18:118–128.
8. China–UK HIV/AIDS Prevention and Care Project. Best practice case studies. Beijing, China; London, UK: Department for International Development; 2004.
9. Centre for Harm Reduction. Manual for reducing drug related harm in Asia, 2nd ed. Melbourne: The Burnett Institute; 2003.
10. Hammett TM, Wu Z, Duc TT, Stephens D, Sullivan SG, Liu W, et al. “Social evils” and harm reduction: the evolving policy environment for HIV prevention among injection drug users in China and Vietnam. Addiction 2007 Nov. 20: [Epub ahead of print].
11. Hammett TM, Chen Y, Ngu D, Cuong DD, Van LK, Liu W, et al. A delicate balance: law enforcement agencies and harm reduction interventions for injection drug users in China and Vietnam. In: Kaufman J, Kleinman A, Saich T, editors. AIDS and social policy in China. Cambridge, Massachusetts: Harvard University Asia Center; 2006. pp. 214–231.
12. Hagan H, Des Jarlais DC, Purchase D, Friedman SR, Reid T, Bell TA. An interview study of participants in the Tacoma, Washington, syringe exchange. Addiction 1993; 88:1691–1697.
13. Fisher DG, Reynolds GL, Harbke CR. Selection effect of needle exchange in Anchorage, Alaska. J Urban Health 2002; 79:128–135.
14. Schechter MT, Strathdee SA, Cornelisse PG, Currie S, Patrick DM, Rekart ML, et al. Do needle exchange programmes increase the spread of HIV among injection drug users?: an investigation of the Vancouver outbreak. AIDS 1999; 13:F45–F51.
15. Bruneau J, Lamothe F, Franco E, Lachance N, Desy M, Soto J, et al. High rates of HIV infection among injection drug users participating in needle exchange programs in Montreal: results of a cohort study. Am J Epidemiol 1997; 146:994–1002.
16. Miller M, Eskild A, Mella I, Moi H, Magnus P. Gender differences in syringe exchange program use in Oslo, Norway. Addiction 2001; 96:1639–1651.
17. Hahn JA, Vranizan KM, Moss AR. Who uses needle exchange? A study of injection drug users in treatment in San Francisco, 1989–1990. J Acquir Immune Defic Syndr Hum Retrovirol 1997; 15:157–164.
18. Wu Z, Sullivan SG, Wang Y, Rotheram-Borus MJ, Detels R. Evolution of China's response to HIV/AIDS. Lancet 2007; 369:679–690.
19. State Council of P.R. China. China's action plan for reducing and preventing the spread of HIV/AIDS (2006–2010) [in Chinese]. Beijing: State Council of P.R. China; 2006.
20. Burrows D. Rethinking coverage of needle exchange programs. Subst Use Misuse 2006; 41:1045–1048.
© 2007 Lippincott Williams & Wilkins, Inc.