The risk behaviors of human immunodeficiency virus (HIV)-positive people who are unaware of their seropositive status have serious implications for the secondary transmission of HIV. As Beckwith et al wrote in 2005, “increasingly, the challenge for the health care community is not how to prevent progression of HIV disease in a person with known infection, but, rather, is how to identify persons who are unknowingly infected with HIV.”1 Numerous studies find that HIV-positive individuals show a tendency to reduce to their sexual risk behaviors after a positive test notification,2–6 fueling a movement in the scientific community to scale-up counseling and testing services. Some have called for universal nonvoluntary testing for all sexually active persons7 whereas others point to the need for an intervention that takes the social aspects of individual choice into consideration.8 This article is one of few to date that examines the effects of different modes of testing on postnotification risk behaviors, and analyzes both sexual and drug-using risk behaviors. This is an important consideration in Southern China where high prevalence in certain areas is driven by both injection drug use and unprotected sex,9 which demands consideration of the overlapping effects of these risk behaviors in assessing postnotification behavioral changes.
In 2005, UNAIDS reported that even though China had an estimated 650,000 people living with HIV, only 141,241 or 21.7% seropositive individuals had been identified through testing, meaning that the large majority of HIV-positive individuals do not know that they are infected.10 Government sponsored voluntary counseling and testing (VCT) was introduced to China in 2004, but the national testing campaign has come under intense scrutiny due to what is now seen as its failure to inform many of those who are infected, and alternative strategies for widespread testing include routinized testing and mass screenings are currently under discussion in China.3,11 In the face of diversified counseling and testing methods, developing a testing strategy that can effectively notify HIV-positive individuals of their serostatus, protect the rights and identities of such individuals, and increase overall awareness on HIV/AIDS is a considerable challenge. Understanding the extent to which HIV testing formats and programs affect the risk behavior of at-risk groups must also take into account local environments and other factors that also influence such behaviors.
This study builds on existing research to further investigate the psychological and social motivations behind reductions in posttest risk behaviors among groups involved in drug use and/or commercial sex work to better inform the development and implementation of HIV testing strategies that can also function as preventative interventions. The study site of Kaiyuan in Yunnan province of Southwestern China provides an opportunity to study HIV-related risk behaviors in a high prevalence area. Because Yunnan province is located near the Golden Triangle of Southeast Asia—so called because of its extensive opium production—the illicit drug trade has ushered along with prevalent heroin use and a subsequent HIV epidemic among injecting drug users (IDU). The commercial sex industry has also flourished in many parts of China and research has shown that it may be fueling the sexual HIV epidemic.12 With increasing evidence of HIV spreading to the general population through unprotected sex between high-risk individuals and their low-risk sexual partners,13 there is a growing need for cost-effective prevention methods, particularly in regions such as Southern China where low levels of education and an underdeveloped health delivery system present severe resource constraints.
MATERIALS AND METHODS
Adapted convenience sampling methods were used to systematically identify and select 177 eligible subjects. Participants were recruited by social outreach workers from local community-based organizations such as the Red Ribbon office, methadone maintenance clinics, the municipal hospital, municipal drug rehabilitation center, and local maternal health clinics. After obtaining verbal informed consent, private interviews were conducted by authors (X.J. and S.Y.) in community-based locations between September to October, 2007. Eligibility criteria required that subjects be over 16 years of age (all institutional review for human subject research in China operates on the foundation of domestic law that dictates that individuals attain legal adult status at 16 years of age), had taken at least one HIV test and knew their test results, and also had ever used drugs through injection. Participants were compensated 20 RMB (about US $2.80) for their time and travel costs.
In-depth interviews collected subject information including demographic characteristics, HIV testing history including testing mode—voluntary and nonvoluntary (in this study, voluntary testing includes the following: VCT, as part of a routine health check up for work or ante-marital examination where refusal is permissible, through participation in a past research study, upon voluntary admittance to a methadone maintenance treatment clinic, or upon the recommendation of a physician for patients showing signs of an AIDS-related opportunistic infection. Non-voluntary testing is defined as compulsory testing for individuals committed to compulsory detoxification centers, and routine testing for prenatal and pre-surgical examinations)—and test results, and HIV-related risk behaviors both before and after testing. Outcomes were self-reported risk behaviors including condom use with regular, casual, and commercial partners and needle-sharing behavior—before and after awareness of HIV serostatus. Regular sexual partners were defined as a spouse or long-term sexual partner with whom the subject has regular sex without monetary compensation; a casual partner as someone a subject has occasional sex without compensation; and commercial sexual activities were defined as monetarily compensated sex, often in a venue reserved for such purposes. Condom use was reported as the proportion of sexual acts in the past week in which condoms were used divided by the total number of sex acts that week. Similarly, needle-sharing behaviors were reported as number of nonsharing injections in the past week divided by the total number of injections that week. To compare pre- and posttest risk behaviors for subjects who had been tested more than once, we defined the test in question for HIV-negative subjects to be the most recent test for which they had received test results, and for HIV-positive subjects as the test in which they learned of their seropositive status.
All analyses were performed using SPSS l5.0 statistical software (SPSS Inc., Chicago, IL). Descriptive demographic variables and risk behavior are presented as categorical frequencies. Risk behaviors were compared with Fisher exact test. Relationships between the risk factors and risk behaviors were analyzed with nonconditional logistic regression by calculating odds ratios with 95% confidence intervals (CI). All variables were included in multivariate logistic regression analysis in a stepwise manner. Results fit for the full models are presented as odds ratio (OR) with 95% CI (or P <0.05).
Of the 172 subjects, the mean age of subjects was 35 years. Thirteen percent had no elementary education, 62% had a junior high school education (9 years), and about 24% had a higher education (>9 years). Over 30% of the subjects were married or cohabiting a regular sexual partner. Approximately, half (54%) of the subjects earned less than 500 RMB a month (about US $70), whereas 16% had more than 500 but less than 1500 earning in month. About 75% of subjects spent an average of 30RMB (about US $4.30) a day on heroin.
Fifteen of the study subjects (9%) were being held in compulsory detoxification centers or reeducation-through-labor camps at the time of interview. The mean reported age of first intercourse was 18.6 years. Nearly 41% was under the age of 18 at the time of first intercourse. And approximately 79% subjects’ self-reported sexual objects were boy/girl friend. The mean reported drug use time was 10.88 years (Table 1).
HIV Testing History and Test Results
Regarding HIV-testing mode, 71 (41%) learned of their most recent serostatus through voluntarily testing and 101 through nonvoluntary procedures (59%). Eighteen subjects who had undergone nonvoluntary testing at detention centers were unaware of their results from their first test due to current policies that dictate that detainees can only obtain test results upon their release. It was also not uncommon for those who tested voluntarily to not return for their results out of fear of an HIV-positive notification. All 21 subjects who had been unaware of the results of their first test became aware of their serostatus through later testing.
Regarding the most recent test results, 98 (57%) of subjects were of confirmed HIV seropositive at the time of the study and 74 (43%) were negative as of the last test. Over half of the subjects (66%) had only tested once for HIV infection, but 49% had been tested between 2004 and 2006, and 34% (59) in 2007, which is partial evidence of the effects of the implementation of the 5-year China Integrated Programs for Research on AIDS and a national VCT campaign in Kaiyuan (Table 2). A large majority of subjects reported their test results to others, including spouses (78%), family members (63%), sexual partners (38%), and friends (34%).
Condom Use and Needle-Sharing Behavior Changes
In our study sample, most subjects (89.5%) reported having a regular sexual partner; however, up to 43.0% also reported that they engaged in commercial sex. Fisher exact tests were used to compare behavior changes before and after test result notification for IDU, across testing mode. Analysis revealed that after test result notification IDUs of both testing modes reported increased condom use with regular sexual partners (P = 0.00 for voluntary testers, P = 0.02 for nonvoluntary). In sex with either casual or commercial partners, subjects did not report any change in condom use behavior regardless of testing mode or serostatus. Finally, all IDUs—regardless of testing strategy or serostatus—reported significantly reduced incidence of needle-sharing behavior (P = 0.000 for all 4 categories) (Table 3).
Multivariate Analyses of Risk Behaviors
We used multivariate regression analysis to analyze factors associated with increased condom use (with both regular and commercial partners) and lowered needle-sharing behavior. Only those variables significantly associated with the relevant outcome variables in univariate analysis were included in the final regression model. Regarding sex with regular partners, subjects who had received a positive test result notification reported a significant increase in condom use(P = 0.00, OR = 5.37, CI: 1.80 ∼ 16.04). In commercial sexual activities, IDUs who had tested for HIV voluntarily were more likely to use condom in sexual activities (P = 0.01, OR = 10.94, CI: 1.74 ∼ 68.66), while age over 35 was found to be a risk factor for lowered condom use (P = 0.04, OR = 0.09, CI: 0.02 ∼ 0.57). Regarding patterns in needle sharing, an HIV-positive test results was associated with a higher frequency of needle-sharing (P = 0.03, OR = 0.30, CI: 0.10 ∼ 0.90) whereas voluntary testing and notification of test results led to less sharing (P = 0.00, OR = 4.51, CI: 1.61 ∼ 12.62 and P = 0.04, OR = 0.34, CI: 0.12 ∼ 0.93) (Table 4).
The primary results of this study found that getting tested for HIV, regardless of the testing mode, is significantly associated with a posttest increase in condom use with regular sexual partners, and that those who were HIV-positive had a more significant increase in condom use than those who were negative. Analysis also revealed that being male was significantly associated with increased condom use in commercial sexual activities. Finally, this report also found that decreased needle-sharing was significantly associated with being HIV-positive, having notified the HIV test results to one’s regular partner, and voluntarily testing for HIV.
These basic findings largely concur with past research on behavior changes in sexual practices, particularly in sex with regular partners, among seropositive individuals who are aware of their status.6,14,15 Interestingly, increased condom use was not associated with many key factors such as partner notification or type of testing strategy, possibly because individual choice regarding condom use is influenced by multiple factors beyond the simple awareness of one’s own serostatus. For example, women who sell sex for a living may abstain from increasing condom use with clients for fear that they would be suspected of infection, and indeed, of the 26 women in this sample who sold sex 23 of them continued to sell sex after becoming aware of their serostatus. This hypothesis is further supported by the fact that being male was found to be a predictor for increased condom use in commercial sex, implying that men exercise greater decision making power for condom use in commercial sex. Finally, this report also found that partner notification was a meaningful predictor for lowered sharing, possibly because much of needle sharing takes place among individuals with a close relationship including regular sexual partners.
The policy implications for these findings are that testing and counseling services must not only be expanded and improved, but must also consider the broader social context of individual behavior change. Past research has shown that in addition to awareness of one’s own serostatus, additional factors such as drugs addiction and sharing test results with friends and sexual partners all play a significant role in long-term behavior change.16,17 In light of this past research, we extend two primary recommendations based on the new findings of this report. First, we propose that voluntary testing methods eventually replace all nonvoluntary testing in China, based on the fact that nonvoluntary testing does not predict positive behavior change, and because compulsory testing raises concerns of human rights abuses as raised by the international community in the Third Conference on HIV/AIDS International Cooperation Projects in China in 2005.18 Second, based on our finding that partner notification predicts decreased needle-sharing, we propose that notification habits be emphasized in existing VCT practices and other voluntary testing programs.
In addition to these recommendations we also support the expansion of the emerging practices of “active testing” in the field of HIV testing in China. Wu et al define active testing as a strategy in which community health workers actively target high-risk groups for testing through outreach. Our study results show that individuals at risk for HIV due to their behaviors are responsive to the measureable impacts of testing on posttest behaviors, and this approach is suitable for the HIV epidemic in China, which is still largely concentrated among vulnerable populations including IDU or commercial sex workers. However, based on these same results, we would like to extend a word of caution regarding two aspects of active testing. First, active testing condones routine testing conducted in institutions such as prisons; however, our study findings show no support for the idea that compulsory testing can generate positive posttest behavior change. Second, because active testing seeks to replace individual counseling services with social marketing campaigns, we feel that this strategy risks bypassing an important opportunity to provide individuals with personalized counseling that can acknowledge factors such as drug addiction or poverty, which this study has found to be associated with the likelihood of positive behavior change.
This study had several limitations. Findings were based on a single cross-sectional survey, which limit our analysis to associations between relevant factors. Further research into this topic would benefit from the analysis of longitudinal data from a cohort study that could detect changes in behavior over time and more directly measure the impact of HIV testing on posttest risk behaviors. Another study limitation was that all data on behaviors were self-reported, which can result in reporting bias as well social desirability bias.19 In addition because data were collected retrospectively, recall bias may have occurred, especially for those in who had HIV test for more than once and test year before 2000.
1. Beckwith C, Flanigan T, Del Rio C, et al. It is time to implement non-voluntary, not risk-based, HIV testing. Clin Infect Dis 2005; 40:1037–1040.
2. George N, Green J, Murphy S. Sexually transmitted disease rates before and after HIV testing. Int J STD AIDS 1998; 9:291–293.
3. Wu Z, Sun X, Sullivan S, et al. Public health: HIV testing in China. Science Mag 2006; 12:1475–1476.
4. Marks G, Crepaz N, Janssen RS. Estimating sexual transmission of HIV from persons aware and unaware that they are infected with the virus in the USA. AIDS 2006; 20:1447–1450.
5. McGowan J, Shah S, Ganea C, et al. Risk behavior for transmission of human immunodeficiency virus (HIV) among HIV-seropositive individuals in an urban setting. Clin Infect Dis 2004; 38:122–127.
6. Colfax GN, Buchbinder SP, Cornelisse PG, et al. Sexual risk behaviors and implications for secondary HIV transmission during and after HIV seroconversion. AIDS 2002; 16:1529–1535.
7. Beckwith C, Simmons EM, Lally M, et al. Testing for HIV Infection: Should It Be Non-voluntary? Resid Staff Physician 2004; 50:9–15.
8. Marks G, Burris S, Peterman T. Reducing sexual transmission of HIV from those who know they are infected: The need for personal and collective responsibility. AIDS 1999; 13:297–306.
9. Ministry of Health Expert Consultation Committee. Report on HIV Screening Among Key Populations in Yunnan Province. Beijing, People’s Republic of China: Ministry of Health; 2005.
11. Chen HT, Liang SL, Liao QL, et al. HIV voluntary counseling and testing among injection drug users in South China: A study of a Non-Government Organization Based Program. AIDS Behav 2007; 11:778–788.
12. Yang HM, Li X, Bonita S, et al. Heterosexual Transmission of HIV in China: A Systematic Review of Behavioral Studies in the Past Two Decades. Sex Transm Dis 2005; 32:5.
13. State Council AIDS Working Committee, Office and UN Theme Group on AIDS in China. A Joint Assessment of HIV/AIDS Prevention, Care and Treatment in China. 2007; 21(suppl 8):S109–S114.
14. Marks G, Crepaz N, Senterfitt JW. Meta-analysis of high-risk sexual behavior in persons aware and unaware they are infected with HIV in the United States: Implications for HIV prevention programs. AIDS 2005; 39:446–453.
15. Kilmarx PH, Hamers FF, Peterman TA. Living with HIV: Experiences and perspectives of HIV-infected sexually transmitted disease clinic patients after posttest counseling. Sex Transm Dis 1998; 25:28–37.
16. Janssen RS, Holtgrave DR, Valdiserri RO, et al. The Sero-status approach to fighting the HIV epidemic: Prevention strategies for infected individuals. Am J Public Health 2001; 91:1019–1024.
17. Kalichman SC, Rompa D, Cage M, et al. Effectiveness of an intervention to reduce HIV transmission risks in HIV-positive people. Am J Prev Med 2001; 21:84–92.
18. Chinese Ministry of Health, UN Theme Group on HIV/AIDS in China. Paper presented at: The Third Conference on HIV/AIDS International Cooperation Projects. September 3–4, 2005; China, Kunming.
19. Schroder KE, Carey MP, Vanable PA. Methodological challenges in research on sexual risk behavior: Accuracy of self-reports. Ann Behav Med 2003; 26:104–123.