Factors associated with acceptability of circumcision among male drug users in western China: a cross-sectional study

Abstract

This study is aimed to investigate whether male circumcision (MC) is feasible among IDUs in China. 1304 drug users who attended methadone maintenance therapy clinics in Guangxi, Chongqing and Xinjiang were selected for participation by using convenience sampling, and completed a self-administered questionnaire. The factors associated with the acceptability of MC were examined via multiple logistic regression models. 45.2% (589/1304) of the participants reported an acceptance of MC. Many of the participants who were initially not willing to accept MC (715/1304) had changed their mind when they were informed that MC would reduce the risk of HIV and STDs (43.4%; 310/715), that MC is associated with few surgery-related complications (23.1%; 165/715), that the surgical procedure could be arranged free of charge (20.1%, 144/715). In the multivariate analysis, higher acceptability of MC was associated with knowledge of the hazards of phimosis (OR=2.22), the presence of phimosis (OR=14.87), and knowledge that MC can prevent AIDS and STDs (OR=1.49); while lower acceptability was associated with residing in Chongqing province (OR=0.41) and an educational level of junior (OR=0.64) and senior high (OR=0.63) school. The MC policy targeting IDUs in China should take into account these factors associated with MC acceptability.

Keywords

HIV AIDS men transmission prevention male circumcision acceptability factors sexual behaviour China

BACKGROUND

As of the end of October 2009, there were a total of 740,000 (range 540,000 to 1,000,000) reported cases of human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome (AIDS) in China.¹ Yunnan, Guangxi, Henan, Sichuan, Xinjiang and Guangdong provinces, four of which are located in western China, accounted for 77.1% of the total cases.² More than half of the reported HIV cases in China were transmitted through sexual contact,^{3
–5} while injection drug users (IDUs) accounted for 24.3% of the total reported cases in 2009.⁶ China has the largest numbers of IDUs worldwide, with 12% HIV prevalence among this population;⁷ the HIV prevalence among IDUs is highest in Sichuan (29.0%), Guangxi (21.0%) and Xinjiang (10.8%).⁸

IDUs are a major population at risk of HIV infection due to their risky behaviours. While needle sharing is the most serious risk factor for acquiring HIV within this population, high-risk sexual behaviours are also an important HIV risk factor among IDUs.^{9

–14} In addition, via these behaviours IDUs could acquire HIV from other populations^{15
–17} and continue to transmit it.¹⁸ Especially in MMT (methadone maintenance therapy) facilities and needle exchange centres, interventions for clean needles reduce risk of transmission among IDUs^19,20 but leaves possibility for sexual transmission from/to their sexual partners. So a comprehensive approach that includes prevention of HIV not only through intravenous transmission but also through sexual transmission will be required to combat the HIV epidemic among IDUs. In this study, we evaluated the acceptability and challenges surrounding the use of male circumcision (MC) as an adjunctive strategy to further prevent the spread of HIV within this high-risk population. The control of HIV infection among the IDU population is likely to have a significant impact on the control of HIV in this population.

Studies have shown that MC can reduce HIV transmission rates in men by 51.0–61.0%.^{21
–23} The World Health Organization (WHO) and the Joint United Nations Program on AIDS (UNAIDS) recommend that MC programmes should be rolled-out wherever there is a generalized HIV epidemic and few men are circumcised.^24,25 WHO/UNAIDS had also issued guidelines in 2007 aiming to reduce new HIV infection in men.²⁵

Despite its observed benefits on the reduction of HIV transmission, circumcision is not common among Chinese men. The overall rate of MC in China is less than 20%,²⁶ which is lower than that in Africa, the USA, Thailand and Republic of Korea.^{27

–30} The acceptability of MC has been studied in men and women in several countries including in Africa, the USA, Thailand and India.^{28,31

–34} However, the acceptability of MC among the Chinese population has yet to be systematically studied. One 2009 study reported an acceptability rate of 40.6% among the Yi ethnic population in China.³⁵ A 2012 study found that the acceptability of early infant MC among parents in Nanjing, China was 34.4%.³⁶ Other studies have reported acceptance levels among men who have sex with men (43.3%),³⁷ male miners (25.1%)³⁸ and the general public (45.0%).³⁹ However, we found no literature that focused on the acceptability of MC among IDUs specifically. In this study, we explored the factors associated with the acceptability of MC among IDUs in western China, a region with low prevalence of MC and higher prevalence of HIV.

METHODS

Study design and participants

A cross-sectional survey was conducted in the form of face-to-face structured interviews in MMT clinics in Guangxi, Chongqing and Xinjiang provinces in western China. The study utilized convenience sampling, screening all patients who attended the MMT clinics (2 clinics in Guangxi, 3 in Chongqing and 3 in Xinjiang) from September 2009 to December 2010. Subjects were recruited for participation if they were men, age 18–45 years and reported using injection drugs for more than six months. Subjects were excluded if they were unable to provide voluntary informed consent or were suffering from severe mental illness, mental retardation or a language disorder, which were diagnosed by doctors in MMT clinics. The study was approved by the Ethics and Human Subjects Committee of the Guangxi Medical University.

Data collection

An 83-item structured questionnaire was used for data collection. The questionnaire sought information in the following areas: (i) demographic characteristics, (ii) general knowledge about AIDS, (iii) general knowledge about MC, (iv) willingness and reasons to undergo circumcision and (v) drug use and sexual behaviour. Details of the questionnaire are described elsewhere.³⁹

Briefly, 10 questions were asked to assess knowledge about AIDS and 11 questions to assess knowledge and attitudes towards MC.³⁷ For AIDS knowledge the average score among all interviewed subjects was computed, with each correct response receiving a score. Willingness to accept circumcision was assessed via the following question: ‘Do you want to be circumcised to prevent HIV?’. The response categories were ‘definitely willing’, ‘probably willing’, ‘definitely not willing’ and ‘probably not willing’. For analysis, the variable was dichotomized; the categories ‘definitely willing’ and ‘probably willing’ were combined into a single category labelled ‘willingness to accept MC’, and the categories ‘definitely not willing’ and ‘probably not willing’ were combined into a single category labelled ‘refusal to accept MC’. To assess reasons to accept or refuse circumcision, 12 open-/close-ended questions were asked that enquired about the advantages and disadvantages of MC. To assess the impact of delivering a brief educational intervention to improve the acceptability of MC, participants who were not willing to be circumcised were asked whether they would change their mind if they were informed of the benefits of MC (e.g. a reduction in the risk of HIV and sexually transmitted infections [STIs], very few surgical complications and the arrangement of circumcision free of charge). Data were collected by trained research assistants (RAs). After the subjects provided their written informed consents to participate in the study, RAs conducted the detailed interviews following structured written guidelines.

Statistical analysis

Data were entered in EpiData 3.1 for Windows (The EpiData Association, Odense, Denmark) and analysed using SPSS for Windows Version 16.0 (SPSS, Chicago, IL, USA). Descriptive statistics and univariate analysis were generated for each of the variables corresponding to specific questions in the survey. To compare the basic characteristics of the two groups, the chi-squared test was used. Multivariate logistic regression analysis was performed to identify factors associated with the acceptability of MC. The variables included in the logistic regression model were those that showed a statistically significant association (P < 0.05) with the willingness to be circumcised in the univariate analyses. All statistical tests were two-sided with a significance level of P < 0.05.

RESULTS

Demographics of the participants

Of the 1386 IDUs who agreed to be interviewed, 1304 (94.1%) completed the entire questionnaire. Of those 1304 participants, 556 (42.6%) were from Guangxi, 408 (31.3%) from Chongqing and 340 (26.1%) from Xinjiang. The majority of the participants was aged 36 years or above (72.6%), was of Han ethnicity (92.5%) and had a high school (junior or senior) level of education or beyond (82.8%). In addition, 46.3% were married and 43.9% were employed; 41.6% of participants had taken drugs at the age of 20–30 years and 78.6% had received MMT in the past year. Almost all of the participants have had sexual intercourse (97.8%, n = 1275) and a majority had received AIDS prevention advice or detection of HIV from public health providers (73.1%) and received publicity materials of AIDS or STIs in the past year (60.7%) (Table 1).

Table 1

Demographic characteristics of the participants (N = 1304)

Variables	n	%
Source of samples (different regions)
Guangxi	556	42.6
Chongqing	408	31.3
Xinjiang	340	26.1
Ages
≤24	44	3.4
25–35	313	24
≥36	947	72.6
Ethnic group
Han	1206	92.5
Other ethnicity	98	7.5
Marital status
Never married	523	40.1
Married or live together	604	46.3
Divorced/separated/widowed	177	13.6
Education level
Elementary school or below	222	17
Junior high school	682	52.3
Senior high school or above	398	30.5
Employment status
Not employed	731	56.1
Employed	573	43.9
Grades of AIDS knowledge*
<9.3	531	40.7
≥9.3	773	59.3
The age of first taking drugs
≤19	491	37.7
20–30	543	41.6
≥31	179	13.7
Refused to answer	91	7
Received methadone maintenance therapy in the past year?
Yes	1025	78.6
No	279	21.4
Received AIDS advice or detection from public health service providers in the past year?
Yes	953	73.1
No	351	26.9
Received publicity materials of AIDS or sexually transmitted infections (STIs) in the past year?
Yes	792	60.7
No	512	39.3
Received peer education of AIDS in the past year?
Yes	327	25.1
No	977	74.9
Received any free condoms from public health service providers in the past year?
Yes	255	19.6
No	1049	80.4
Exchanged needles or received clean needles from Needle Exchange Centre in the past year?
Yes	211	16.2
No	1093	83.8
Ever had sexual intercourse?
Yes	1275	97.8
No	29	2.2
Did you have urogenital infection symptoms, such as painful urination in the past year?
Yes	149	11.4
No	1149	88.1
Refused to answer	6	0.5

*The average score was 9.3 for all interviewed male drug users. The maximum score is 10.0

Acceptability of MC and reasons for accepting or not accepting MC

Of the 1304 participants, 589 (45.2%) reported acceptance of MC. The major reasons were: 58.4% had self-reported phimosis, 53.0% believed that MC could help prevent penile cancer, 46.2% believed that HIV and STIs could be partially prevented by MC, 43.5% believed that MC could improve the hygiene of their sexual partners and 23.8% believed that MC could enhance sexual pleasure. Of those who refused MC (n = 715), the majority (93.3%) believed that MC would not be effective for them, 2.9% were worried about the reduction of sexual ability and 2.9% were concerned about the expensive cost of MC surgery. Many of the participants who were initially unwilling to accept MC (715/1304) changed their mind when they were informed that MC would reduce the risk of HIV and STIs (43.4%; 310/715), or that MC was associated with few surgery-related complications (23.1%; 165/715), or the surgical procedure could be arranged free of charge (20.1%, 144/715) (Table 2).

Table 2

Reasons for accepting or refusing MC among male drug users in China

	Frequency	%
Willing to accept MC	589	100.0
Having phimosis	344	58.4
Prevention of penile cancer	312	53.0
Protection against HIV and STIs	272	46.2
Improve partners’ hygiene	256	43.5
Enhance sexual pleasure	140	23.8
Better penile appearance	61	10.4
Traditional or religious reason	19	3.2
Other reasons	5	0.8
Refuse MC*	715	100.0
Not necessary or not effective	667	93.3
No idea of benefits of MC	46	6.4
Concern about potential danger associated with surgery	32	4.5
Concern about reducing sexual ability	21	2.9
Concern about expensive surgery cost	21	2.9
Other reasons	19	2.7

MC = male circumcision; STI = sexually transmitted infection

*After they were told that MC would reduce HIV and STIs, had only very few surgery-related complications, and could be arranged free of charge, 43.4% (310/715), 23.1% (165/715), and 20.1% (144/715) would accept MC, respectively

Knowledge about AIDS and MC

For all participants, the average score of AIDS knowledge was 9.3 (out of 10), with 773 (59.3%) scoring higher than the average level (Table 1).

Those participants who were willing to be circumcised (i.e. WTC group) possessed greater knowledge of AIDS and MC than the non-WTC participants. A significantly larger proportion of subjects in the WTC group knew that MC could prevent penile inflammation and cancer compared with those in the non-WTC group (49.1% versus 42.8%, P < 0.05). Overall 326 (25.0%) subjects knew that MC could prevent AIDS and STIs, but this knowledge was more prevalent in the WTC group than in the non-WTC group (29.9% versus 21.0%, P < 0.05). A larger proportion of subjects in the WTC group compared with the non-WTC group knew about the hazards of phimosis (93.0% versus 85.3%, P < 0.05) and had friends who had undergone MC (53.5% versus 44.1%, P < 0.05) (Table 3).

Table 3

Factors associated with the willingness to be circumcised

	WTC group, n (%)	Non-WTC group, n (%)	χ²	P value
Total	589 (100.0)	715 (100.0)
Source of samples (different regions)			37.9	<0.001
Guangxi	266 (45.2)	290 (40.6)
Chongqing	136 (23.1)	272 (38.0)
Xinjiang	187 (31.7)	153 (21.4)
Ages			1.74	0.420
≤24	24 (4.1)	20 (2.8)
25–35	143 (24.3)	170 (23.8)
≥36	422 (71.6)	525 (73.4)
Ethnic group			1.00	0.318
Han	540 (91.7)	666 (6.9)
Other ethnicity	49 (8.3)	49 (6.9)
Marital status			1.21	0.547
Never married	244 (41.4)	279 (39.0)
Married or live together	263 (44.7)	341 (47.7)
Divorced/separated/widowed	82 (13.9)	95 (13.3)
Education level			6.18	0.046
Elementary school or below	117 (19.9)	105 (14.7)
Junior high school	296 (50.3)	386 (54.1)
Senior high school or above	175 (29.8)	223 (31.2)
Employment status			1.06	0.303
Not employed	321 (54.5)	410 (57.3)
Employed	268 (45.5)	305 (42.7)
Grades of AIDS knowledge*			2.12	0.146
Less than average grades	227 (38.5)	304 (42.5)
Average grades or more	362 (61.5)	411 (57.5)
Ever heard MC?			7.54	0.006
Yes	521 (88.5)	594 (83.1)
No	68 (11.5)	121 (16.9)
Do you know that MC can prevent penile inflammation and cancer?			5.12	0.024
Yes	289 (49.1)	306 (42.8)
No	300 (50.9)	409 (57.2)
Do you know that MC can prevent AIDS and STIs?			13.6	<0.001
Yes	176 (29.9)	150 (21.0)
No	413(70.1)	565 (79.0)
Do you know that MC can improve sexual partners’ hygiene?			10.5	0.001
Yes	295 (50.1)	294 (41.1)
No	294 (49.9)	421 (58.9)
Do you know that MC can enhance sexual pleasure in the future?			0.00304	0.956
Yes	159 (27.0)	194 (27.1)
No	430 (73.0)	521 (72.9)
Do you know MC can improve penile appearance?			0.453	0.501
Yes	73 (12.4)	80 (11.2)
No	516 (87.6)	635 (88.8)
Whether or not your friends undergo MC?			11.5	0.001
Yes	315 (53.5)	315 (44.1)
No	274 (46.5)	400 (55.9)
Do you know the hazard of phimosis?			19.4	<0.001
Yes	548 (93.0)	610 (85.3)
No	41 (7.0)	105 (14.7)
Do you feel that you have phimosis?			256	<0.001
Yes	247 (41.9)	37 (5.2)
No	342 (58.1)	678 (94.8)
The age of first taking drugs			6.44	0.092
≤19	232 (39.4)	259 (36.2)
20–30	230 (39.0)	313 (43.8)
≥31	77 (13.1)	102 (14.3)
Refused to answer	50 (8.5)	41 (5.7)
Ever had sexual intercourse?			1.20	0.274
Yes	573 (97.3)	702 (98.2)
No	16 (2.7)	13 (1.8)
Received methadone substitution treatment in the past year?			8.89	0.003
Yes	441 (74.9)	584 (81.7)
No	148 (25.1)	131 (18.3)
Received AIDS advice or detection from public health service providers in the past year?			0.197	0.657
Yes	434 (73.7)	519 (72.6)
No	155 (26.3)	196 (27.4)
Received publicity materials of AIDS or sexually transmitted infections (STIs) in the past year?			0.0214	0.885
Yes	359 (61.0)	433 (60.6)
No	230 (39.0)	282 (39.4)
Received peer education of AIDS in the past year?			2.91	0.088
Yes	161 (27.3)	166 (23.2)
No	428 (72.2)	549 (76.8)
Received any free condoms from public health service providers in the past year?			0.0132	0.908
Yes	116 (19.7)	139 (19.4)
No	473 (80.3)	576 (80.6)
Exchanged needles or received clean needles from Needle Exchange Centre in the past year?			0.503	0.478
Yes	100 (17.0)	111 (15.5)
No	489 (83.0)	604 (84.5)

WTC = willingness to be circumcised; Non-WTC = no willingness to be circumcised; MC = male circumcision

*The average score was 9.3. The maximum score is 10.0

Factors associated with acceptability of MC

Univariate analysis identified 10 potential factors that were significantly associated with circumcision preference: source of samples, education level, whether ever heard of MC, knowing that MC can prevent penile inflammation and cancer, knowing that MC can prevent AIDS and STIs, knowing that MC can improve sexual partners’ hygiene, having friends who have undergone MC, knowing the hazard of phimosis, feeling of having phimosis, receiving methadone substitution treatment in the past year (Table 3). The multivariate logistic regression model including the above ten significant variables identified three factors that were associated with WTC: knowledge about the hazards of phimosis (OR = 2.22), feeling of having phimosis (OR = 14.87) and knowing that MC can prevent AIDS and STIs (OR = 1.49). In contrast, participants who were residents in the Chongqing province (OR = 0.41) and who were educated at the junior (OR = 0.64) and senior high (OR = 0.63) school level were more likely to reject MC (Table 4).

Table 4

Multivariate analysis of willingness to be circumcised among 1304 male drug users interviewed

	Adjusted OR(95% CI)	P value
Source of samples (different regions)
Guangxi	1.000
Chongqing	0.41 (0.30–0.56)	<0.001
Xinjiang	1.12 (0.82–1.52)	0.480
Education level
Elementary school or below	1.000
Junior high school	0.64 (0.45–0.90)	0.011
Senior high school or above	0.63 (0.43–0.93)	0.020
Do you know the hazard of phimosis?
No	1.00
Yes	2.22 (1.43–3.47)	<0.001
Do you feel that you have phimosis?
No	1.00
Yes	14.87 (10.11–21.87)	<0.001
Do you know that MC can prevent AIDS and STIs?
No	1.00
Yes	1.49 (1.12–2.00)	0.006

OR = odds ratio; STI = sexually transmitted infection; MC = male circumcision

DISCUSSION

This study is the first to report on the acceptability of MC among IDUs. We found that in China, 45.2% of the drug users interviewed in these three western provinces expressed their willingness to undergo MC. This rate of acceptability is higher than that of the young Yi men in Sichuan province of China (40.6%), but still lower than the reported rate of 65.0% (median) in sub-Saharan Africa.³¹ In the multivariate analysis, factors associated with the acceptability of MC included knowledge of the hazards of phimosis, the presence of phimosis and knowledge that MC can prevent AIDS and STIs. Several previous studies also reported partially similar factors associated with the acceptance MC, including maintaining penile hygiene, improvement of MC knowledge, increasing sexual pleasure and preventing HIV infection.^{40
–42}

Data from multivariate analysis indicate that Xinjiang and Guangxi showed similar acceptability of MC, while Chongqing had lower acceptability. It may be due to the differences in the STI/AIDS health education campaigns that were promoted in these provinces. Province- and city-based Centres for Disease Control and Prevention (CDC) offices often plan and design different forms of educational campaigns (i.e. pamphlets, media) according to their available budgets. Unfortunately, we do not have the detailed information about the STI/AIDS-related health education campaigns in these provinces. We also found that a higher level of educational attainment (being at junior high school or at senior high school and above) was not associated with accepting MC. In a Ugandan study the awareness of MC was associated with increasing educational level.⁴³ In Chinese society, cultural beliefs play a greater role in accepting health-promoting behaviour.⁴⁴ Those with more education might have known about different routes for HIV transmission and culturally was not in favour of MC, a society where circumcision is not commonly practiced, thus rejecting MC acceptability. In contrast, those with low education were not knowledgeable about the potential routes of HIV transmission and perceived greater benefits from MC regardless of cultural debut and accepted MC. A future qualitative study should explore this in greater details.

Although multivariate analysis showed that higher educational level was associated with a rejection of MC, some subjects changed their minds after receiving information about the benefits of circumcision. We believe the increased knowledge about the protective effects of MC had had an impact on participants’ attitudes towards MC acceptability. IDUs were more likely to accept MC if they understood the association between MC and AIDS prevention, which is not currently integrated into public health education campaigns. This finding supports the idea that knowledge of MC and HIV should be disseminated via the school or other public education campaigns. The dissemination would increase public knowledge on understanding MC as one measure against HIV spread in addition to other protective measures. These findings of association between increased HIV knowledge and MC acceptability are consistent with earlier findings in the Dominican Republic,^29,33 Africa^31,45,46 the USA^32,47 and among the general public in China.³⁹ In addition, we also found that there was no significant difference in the acceptability of MC between different age groups, which differs from the studies conducted in the Dominican Republic and Kenya.^33,48 This indicates that among IDUs in China, the older population does not have a greater knowledge of MC and HIV than the younger population, and that education among all ages should be enhanced in the future.

Several limitations should be taken into account in this study. We only focused on the drug users who were registered in MMT; subjects who were unable to provide voluntary informed consent or who were suffering from severe mental illness, mental retardation or language disorder were excluded, and such convenience sampling may lead to sampling bias and raise concern of its generalizability to all IDUs. We were unable to check the characteristics of the drug users who were not registered in the MMT, but compared the characteristics of those MMT participants who refused to participate in the study and found no differences in the characteristics of the participants and non-participants. Therefore our findings should be generalizable to drug users attending MMT clinics. Moreover, the collected information was primarily based on reported behaviours or characteristics, such as self-reported phimosis without any clinical examination. Furthermore, the influence of different religious beliefs, especially among Muslims and Catholics, on the acceptability of MC was not addressed in this study; more studies are needed to assess the association between the two. In addition, there is a possibility that we may be overestimating the willingness of MC due to the social desirability bias present during the interview. Finally, ‘willingness’ does not always translate into ‘action’; the proportion of action is pretty low.⁴⁹ However, we believe that ‘willingness’ is a necessary first step and that the results of this study illustrate the need for additional education in order to overcome much of the barriers to the acceptability of MC. Finally we did not distinguish between the epidemiological facts that MC reduces the risk for contracting HIV^{21
–23} as opposed to the limited evidence about whether a circumcision can reduce the transmission from an infected man to an uninfected partner.^50,51 It is beyond the scope of this paper to conclude whether circumcising a high-risk population would reduce transmission to uninfected individuals. However our view is to follow the public health recommendations of the USA CDC, which recommends MC as an additional HIV prevention measure, which recognizes that circumcision (1) does carry risks and costs that must be considered in addition to potential benefits; (2) has only proven effective in reducing the risk of infection through penetrative vaginal sex; and (3) confers only partial protection and should be considered only in conjunction with other proven prevention measures (abstinence, mutual monogamy, reduced number of sex partners, and correct and consistent condom use).⁵² However, there is controversy about whether MC should be considered only in conjunction with other proven interventions. Given the fact that MC is 60–70% efficacious in preventing HIV infection, MC could be promoted as a preventive measure in its own right.⁵³ The 2012 policy statement of the American Academy of Pediatrics did not suggest that MC could be promoted only if other interventions are provided; rather it recommended access to MC for families who choose it.⁵⁴

CONCLUSIONS

In conclusion, we found that nearly half of IDUs (45.2%) in western China were willing to accept male circumcision. The acceptability rates varied remarkably in different provinces and at different education levels, and were associated with MC/AIDS-related knowledge. These results indicated that MC-targeted education should be taken into account for MC program towards IDUs in China.

Footnotes

ACKNOWLEDGEMENTS

We thank all participants of the study for their time and sharing their information. We especially thank Dr Yiming Shao and Dr Zunyou Wu from National Center for AIDS/STD Control and Prevention (NCAIDS), Chinese Center for Disease Control and Prevention, Beijing, China for their contribution to the project.

This study was supported by grants from National Key Science and Technology Project (Grant Number 2008ZX10001-016), Guangxi Natural Science Foundation (2010GXNSFD013045), as well as supported by Program for Innovative Research Team of Intellectual Highland in High School of Guangxi (Guijiaoren[2010]38).

Competing interests: The authors declare that they have no competing interests.

Authors’ contributions: The work presented here was carried out in collaboration between all authors. HL, BW and JH designed the overall study. HL, BW, XY, BQ and YR designed methods and experiments. ASA was a consultant in the project and contributed to the conceptualization of the current paper. JH, XY, PX, FW and NX performed the statistical analysis and interpreted the results. JH, JJ and ASA wrote the paper. WD, HU, QW, CZ, QW and YZ helped in the collection of data and interpretation of results. YZ, LY and XY conducted the final analyses, and summarized the results. All authors have contributed to, read and approved the final manuscript.

References

UNAIDS. Global Report: UNAIDS Report on the Global AIDS Epidemic 2010. WHO Library Cataloguing-in-Publication Data, 2010

Ministry of Health of China, UNAIDS, WHO

2010 Update on the HIV/AIDS Epidemic in China. 2011. See: http://www.unaids.org.cn/download/2009%20China%20Estimation%20Report-En.pdf

Teng

Shao

. Scientific approaches to AIDS prevention and control in china. Adv Dent Res 2011;23:10–2

UNAIDS. AIDS Epidemic Update: December 2007. WHO Library Cataloguing-in-Publication Data, 2007

A Joint Assessment of HIV/AIDS Prevention, Treatment and Care in China (2007). State Council AIDS Working Committee Office & UN Theme Group on AIDS in China, 2007

MOH, UNAIDS, WHO: 2009 Estimates for the HIV/AIDS Epidemic in China, 2010. See: http://www.unaids.org.cn/download/2009%20China%20Estimation%20Report-En.pdf

Mathers

Degenhardt

Phillips

Global epidemiology of injecting drug use and HIV among people who inject drugs: a systematic review. Lancet 2008;372:1733–45

Kretzschmar

Zhang

Mikolajczyk

Regional differences in HIV prevalence among drug users in China: potential for future spread of HIV?

BMC Infect Dis 2008;8:108

Abdala

Krasnoselskikh

Durante

Timofeeva

Verevochkin

Kozlov

. Sexually transmitted infections, sexual risk behaviors and the risk of heterosexual spread of HIV among and beyond IDUs in St. Petersburg, Russia. Eur Addict Res 2008;14:19–25

10.

Brodish

Singh

Rinyuri

Evidence of high-risk sexual behaviors among injection drug users in the Kenya PLACE study. Drug Alcohol Depend 2011;119:138–41

11.

Liu

Luo

Jarlais

Koram

. Role of sexual transmission of HIV among young noninjection and injection opiate users: a respondent-driven sampling study. Sex Transm Dis 2011;38:1161–6

12.

Shacham

Cottler

. Sexual behaviors among club drug users: prevalence and reliability. Arch Sex Behav 2010;39:1331–41

13.

Todd

Earhart

Botros

Prevalence and correlates of risky sexual behaviors among injection drug users in Tashkent, Uzbekistan. AIDS Care 2007;19:122–9

14.

Yao

Wang

Chu

Sexual behavior and risks for HIV infection and transmission among male injecting drug users in Yunnan, China. Int J Infect Dis 2009;13:154–61

15.

Strathdee

Sherman

. The role of sexual transmission of HIV infection among injection and non-injection drug users. J Urban Health 2003;80(Suppl. 3):i7–i14

16.

Kral

Bluthenthal

Lorvick

Gee

Bacchetti

Edlin

. Sexual transmission of HIV-1 among injection drug users in San Francisco, USA: risk-factor analysis. Lancet 2001;357:1397–401

17.

Control CFD, US P. Drug-associated HIV transmission continues in the United States: Centers for Disease Control and Prevention, 2002

18.

Jenness

Neaigus

Hagan

Murrill

Wendel

. Heterosexual HIV and sexual partnerships between injection drug users and noninjection drug users. AIDS Patient Care STDS 2010;24:175–81

19.

Zhang

Yap

Xun

Wilson

. Needle and syringe programs in Yunnan, China yield health and financial return. BMC Public Health 2011;11:250

20.

Smith

Bartlett

Wang

. A harm reduction paradox: comparing China's policies on needle and syringe exchange and methadone maintenance. Int J Drug Policy 2012;23:327–332

21.

Auvert

Taljaard

Lagarde

Sobngwi-Tambekou

Sitta

Puren

. Randomized, controlled intervention trial of male circumcision for reduction of HIV infection risk: the ANRS 1265 Trial. PLoS Med 2005;2:e298

22.

Bailey

Moses

Parker

Male circumcision for HIV prevention in young men in Kisumu, Kenya: a randomised controlled trial. Lancet 2007;369:643–56

23.

Gray

Kigozi

Serwadda

Male circumcision for HIV prevention in men in Rakai, Uganda: a randomised trial. Lancet 2007;369:657–66

24.

Siegfried

Muller

Deeks

Volmink

. Male circumcision for prevention of heterosexual acquisition of HIV in men. Cochrane Database Syst Rev 2009;D3362

25.

WHO-UNAIDS. New data on male circumcision and HIV prevention: policy and programme implications. 2007 . See: http://data.unaids.org/pub/Report/2007/mc_recommendations_en.pdf

26.

Drain

Halperin

Hughes

Klausner

Bailey

. Male circumcision, religion, and infectious diseases: an ecologic analysis of 118 developing countries. BMC Infect Dis 2006;6:172

27.

Ben

Male circumcision is an effective "surgical vaccine" for HIV prevention and reproductive health. Zhonghua Nan Ke Xue 2009;15:395–402

28.

Tieu

Phanuphak

Ananworanich

Acceptability of male circumcision for the prevention of HIV among high-risk heterosexual men in Thailand. Sex Transm Dis 2010;37:352–5

29.

Brito

Luna

Bailey

. The feasibility and acceptability of male circumcision among men, women, and health providers of the Altagracia Province, Dominican Republic. AIDS Care 2010;22:1530–5

30.

Buchbinder

Vittinghoff

Heagerty

Sexual risk, nitrite inhalant use, and lack of circumcision associated with HIV seroconversion in men who have sex with men in the United States. J Acquir Immune Defic Syndr 2005;39:82–9

31.

Westercamp

Bailey

. Acceptability of male circumcision for prevention of HIV/AIDS in sub-Saharan Africa: a review. AIDS Behav 2007;11:341–55

32.

Gust

Kretsinger

Gaul

Acceptability of newborn circumcision to prevent HIV infection in the United States. Sex Transm Dis 2011;38:536–542

33.

Brito

Caso

Balbuena

Bailey

. Acceptability of male circumcision for the prevention of HIV/AIDS in the Dominican Republic. PLoS One 2009;4:e7687

34.

Madhivanan

Krupp

Chandrasekaran

Karat

Reingold

Klausner

. Acceptability of male circumcision among mothers with male children in Mysore, India. AIDA 2008;22:983–8

35.

Feng

Zeng

Acceptability and related factors on male circumcision among young men with Yi ethnicity in one county of Sichuan province. Zhonghua Liu Xing Bing Xue Za Zhi 2010;31:281–5

36.

Pan

Zhang

Shen

Wang

. Acceptability of early infant male circumcision among Chinese parents: strategy implications of HIV prevention for china. BMC Public Health 2012;12:738

37.

Ruan

Qian

Willingness to be circumcised for preventing HIV among Chinese men who have sex with men. AIDS Patient Care STDS 2009;23:315–21

38.

Hao

Jian

Xiao-Bo

Jun-Jun

Wei

. Acceptability of male circumcision among male miners in Baise of Guangxi. Zhongguo Yi Xue Ke Xue Yuan Xue Bao 2011;33:313–7

39.

Yang

Abdullah

Wei

Factors influencing Chinese male's willingness to undergo circumcision: a cross-sectional study in western China. PLoS One 2012;7:e30198

40.

Herman-Roloff

Otieno

Agot

Ndinya-Achola

Bailey

. Acceptability of medical male circumcision among uncircumcised men in Kenya one year after the launch of the national male circumcision program. PLoS One 2011;6:e19814

41.

Siegler

Mbwambo

Diclemente

. Acceptability of medical male circumcision and improved instrument sanitation among a traditionally circumcising group in East Africa. AIDS Behav 2012;16:1846–52

42.

Gasasira

Sarker

Tsague

Determinants of circumcision and willingness to be circumcised by Rwandan men, 2010. BMC Public Health 2012;12:134

43.

Wilcken

Miiro-Nakayima

Hizaamu

Keil

Balaba-Byansi

. Male circumcision for HIV prevention – a cross-sectional study on awareness among young people and adults in rural Uganda. BMC Public Health 2010;10:209

44.

Wang

Mandelblatt

Liang

Schwartz

. Knowledge, cultural, and attitudinal barriers to mammography screening among nonadherent immigrant Chinese women ever versus never screened status. CANCER 2009;115:4828–38

45.

Lukobo

Bailey

. Acceptability of male circumcision for prevention of HIV infection in Zambia. AIDS Care 2007;19:471–7

46.

Ngalande

Levy

Kapondo

Bailey

. Acceptability of male circumcision for prevention of HIV infection in Malawi. AIDS Behav 2006;10:377–85

47.

Begley

Jafa

Voetsch

Heffelfinger

Borkowf

Sullivan

. Willingness of men who have sex with men (MSM) in the United States to be circumcised as adults to reduce the risk of HIV infection. PLoS One 2008;3:e2731

48.

Mattson

Bailey

Muga

Poulussen

Onyango

. Acceptability of male circumcision and predictors of circumcision preference among men and women in Nyanza Province, Kenya. AIDS Care 2005;17:182–94

49.

Buchbinder

Metch

Holte

Scheer

Coletti

Vittinghoff

. Determinants of enrollment in a preventive HIV vaccine trial: hypothetical versus actual willingness and barriers to participation. J Acquir Immune Defic Syndr 2004;36:604–12

50.

Wawer

. Trial of Male Circumcision: HIV, Sexually Transmitted Disease (STD) and Behavioral Effects in Men, Women and the Community. See: http://www.clinicaltrials.gov/ct2/show/NCT00124878?term=circumcision+AND+uganda&rank=2 . (last checked 23 October 2012)

51.

Gray

Wawer

Thoma

Male Circumcision and the Risks of Female HIV and Sexually Transmitted Infections Acquisition in Rakai, Uganda [Abstract 128]. Presented at 13th Conference on Retroviruses and Opportunistic Infections. 5–9 February2006, Denver, CO. See: http://www.retroconference.org/2006/Abstracts/25977.htm . (last checked 24 October 2012)

52.

CDC HIV/AIDS Science Facts. Male Circumcision and Risk for HIV Transmission and Other Health Conditions: Implications for the United States. See: http://www.cdc.gov/hiv/resources/factsheets/pdf/circumcision.pdf . (last checked 10 October 2012)

53.

Wamai

Morris

Waskett

Criticisms of African trials fail to withstand scrutiny: male circumcision does prevent HIV infection. J Law Med 2012;20:93–123

54.

Task Force on Circumcision. Circumcision policy statement. Pediatrics 2012;130:585–6