Objective: To estimate the incidence of HIV and study the impact of risk-reduction counseling (RRC) in a cohort of people with high-risk behavior for HIV transmission in Chennai, India.
Design: Prospective cohort follow-up of 500 HIV-negative people (250 men and 250 women) at increased risk for HIV acquisition in Chennai, India for a maximum of 1 year was conducted. They received RRC at 0, 6, and 12 months. Generalized estimating equation methodology was used to determine the statistical significance of differences reported in behavior between baseline, 6 months, and 12 months.
Results: The overall HIV incidence in this cohort was 0.44 per 100 person-years (95% confidence interval: 0.05-1.60). In the course of the study, both male and female participants reported statistically significant decreases in the number of different sexual partners, the number of new partners, and the proportion of sexual encounters with nonprimary partners. Participants who had more than 3 different partners at baseline and/or exchanged money for sex in the 6 months before enrollment demonstrated the greatest reductions in the number of different sexual partners.
Conclusions: Individualized sexual RRC seems to be a useful intervention to reduce risk-taking behavior among at-risk heterosexuals in India.
From the *YR Gaitonde Center for AIDS Research and Education (YRG CARE), Chennai, India; †Statistical Center for HIV/AIDS Research and Prevention (SCHARP), Seattle, WA; ‡Division of Infectious Diseases, Department of Medicine, Miriam Hospital/Brown University, Providence, RI; and §Fenway Community Health, Boston, MA.
Received for publication May 18, 2005; accepted March 23, 2006.
This study was supported by the HIV Prevention Network (grant no. 1U01 A148040), the Brown Tufts University AIDS International Research and Training Program - Fogarty Center (grant no. TW00237). and the Lifespan-Brown Tufts Center for AIDS Research (grant no. P30 A142853). This article was supported, in part, by a fellowship/ grant for the Fogarty International Center/USNIH: Grant No. 2 D 43 trv000010-17-AITRP.
Reprints: Kenneth H. Mayer, MD, The Miriam Hospital, 164 Summit Avenue, Providence, RI 02906 (e-mail: Kenneth_Mayer@brown.edu).
HIV infection was first detected in India in 1986 among sex workers in Tamil Nadu.1 Since then, the number of people living with HIV and AIDS (PLWHA) in India has been steadily increasing,2 and it is estimated that more than 5.1 million Indians were infected with HIV at the end of 2004.3 It has also been established that heterosexual contact is the predominant mode of transmission of HIV in India, accounting for 85.7% of all infections.2 The National AIDS Control Organization, India also estimated that 87.7% of all HIV infections in India occur in the most sexually active age group, 15-44 years.2 A study by Mehendale et al in Pune, India reported an HIV-1 prevalence of 21.2% among people attending sexually transmitted disease (STD) clinics.4
Despite the numerous therapeutic innovations that have followed the discovery of HIV as the cause of AIDS,5 behavioral interventions to promote condom use are still among the most effective means of limiting the sexual spread of this infection.6 It is therefore essential to provide some form of risk-reduction counseling (RRC) or education to such individuals, especially those manifesting high-risk behavior. RRC has been reported to be effective in reducing risk of acquisition of HIV/sexually transmitted infections (STIs) both in India7,8 and in industrialized countries.9 RRC has been shown to reduce the risk of acquiring HIV and other STIs, which have been shown to act as cofactors for HIV transmission.10,11
As part of a study of risk factors of HIV seroconversion among high-risk individuals, our group evaluated the efficacy of RRC in an urban Indian population. We measured changes in risk-taking behavior in relation to HIV incidence among 500 HIV-uninfected participants attending an STD clinic and a Voluntary and Counseling and Testing (VCT) center in Chennai, India over the course of a year.
The study population comprised men and women who visited the VCT center of the YR Gaitonde Center for AIDS Research and Education (YRG CARE) or the STD department of the Stanley Government General Hospital in 2002. YRG CARE is a nonprofit, nongovernmental organization that was established in 1993 in South Chennai. It has since provided VCT services to more than 9000 clients and is one of the largest referral centers for HIV-infected individuals in India.12 The Stanley Government General Hospital is a multispecialty hospital in North Chennai that offers services to more than 1000 patients per day and provides STI diagnosis and treatment services to approximately 30 patients per day. YRG CARE was identified by the HIV Prevention Trial Network (HPTN) as a site for the conduct of the HPTN 033 protocol, which was designed to estimate HIV incidence among high-risk individuals in preparation for new interventional studies.
Individuals were invited to participate if they were at least 18 years of age, available for the entire course of the study, and HIV-1 antibody seronegative at their baseline visit. All participants included in the study provided written informed consent and contact information for locator purposes in local languages, Tamil and Telugu. To be considered as high-risk, all participants had to satisfy one or more of the following criteria during the 6 months before enrollment: (1) having had 5 or more different sexual partners of the opposite sex, (2) having been diagnosed/treated for an STI; or (3) having had vaginal/anal sex once a week with an HIV-infected partner of the opposite sex. Individuals were excluded from the study if they were diagnosed as having a psychological or psychiatric condition that might interfere with the study procedure or if they expressed an interest to migrate within the next year.
A sample of size of 500 HIV-negative individuals was calculated to provide a half width of 2.5% for the 95% confidence interval (CI) for HIV incidence based on a 12-month retention rate of 90% and a true incidence rate of 8%. A total of 630 consecutive patients were screened from June to December 2002 to participate in this prospective cohort study. Of this total, 544 satisfied the inclusion criteria and provided written informed consent to participate in the study. Seventeen participants did not provide informed consent, 2 were underage, and 66 failed to meet the high-risk criteria. The others then underwent pretest counseling for determination of HIV serostatus at baseline, using a rapid test (The Determine HIV1/2 Rapid test, Abbott Laboratories, USA). Forty-two of those screened (6.6%) were found to be HIV-infected and 2 had indeterminate results. One person dropped out of the study after baseline testing. These 45 individuals were excluded from the study, and HIV-infected participants were referred to YRG CARE for confirmation of status using the Western blot and further clinical management. The 500 individuals were recruited between the 2 sites as follows: YRG CARE, 339; and Stanley GH, 161.
The 500 study participants were followed every 6 months for 1 year. These individuals received RRC sessions at every visit using the World Health Organization counseling module. The RRC covered basic information of HIV transmission, assessed the level of risk and cognitive understanding, and checked for myths and misconceptions. HIV prevention techniques were discussed, including safer sex techniques and reducing the number of partners. The counselors explored participant's cultural obstacles to changing behavior and used case histories as counseling scenarios. Optimal condom use was demonstrated, and condoms were provided free of charge. Each session lasted approximately an hour. At baseline, all the participants underwent standardized sessions, but during the follow-up visits, they were provided individualized peer counseling. A sociodemographic and risk-assessment questionnaire was also administered at each visit. A follow-up visit was scheduled for each individual for 6 and 12 months after their date of enrollment. At each follow-up visit, participants were again tested for HIV after RRC and HIV pretest counseling. Locator visits were conducted during the third and ninth month of follow-up to confirm the contact information given by the study participants and to reconfirm the date of the follow-up visit. All the RRC sessions and the administration of the risk behavior questionnaires were performed by the same set of counselors at baseline, 6 months, and 12 months. All participants received posttest counseling, and seroconverters were referred to YRG CARE for further management.
The statistical analysis was performed using SAS Version 8.2. For the calculation of the incidence, we assumed that a person lost to follow-up between 2 visits contributed half the time between the 2 visits.
Wherever possible, generalized estimating equation methodology was used to determine the statistical significance of differences reported in behavior between baseline, 6 months, and 12 months. In those instances where generalized estimating equation methodology could not be used, due to the nature of the data, the level of significance of reported changes in behavior was analyzed using an exact McNemar test.
A husband/wife or a steady boyfriend/girlfriend was considered a primary partner for the analysis, and anyone besides these was considered a nonprimary partner. Giving money/gifts or receiving money/gifts in exchange for sex was considered a commercial sexual encounter in the analysis.
As a stratified analysis, the already high-risk participants were further divided into subgroups. Participants who had either had 4 or more partners in the 6 months before enrollment, or participants who had either paid or received money/gifts in exchange for sex in the 6 months before baseline, were considered to be at highest risk for acquiring HIV based on an earlier Indian study.13 The 3 parameters considered in the stratified analysis were the number of different sexual partners, percentage of engaging in sexual encounters with nonprimary partners, and the number of new sexual partners. A P value of less than 0.05 was considered statistically significant.
The study cohort comprised 250 men and 250 women, as specified by the protocol. The mean age among men in the cohort was 31 years. Of them, 44.8% were single. Most men were ethnic Tamilians (90.4%). Most men had undergone primary schooling (26.4%) or middle/high school (58.8%); 70.8% of them were employed full time.
The mean age of the female participants was 31.8 years. Of them, 60.4% were married, 85.6% were Tamilians, 28.4% had not attended any school, 30.4% were unemployed, and 28% reported part-time employment. Detailed results of participant demographics are presented in Table 1.
During the course of this study, only 2 participants, 1 male and 1 female, seroconverted over the 457.5 person-years of follow-up resulting in an incidence rate of 0.44 per 100 person-years (95% CI: 0.05-1.60 per 100 person-years).
Over the course of the year, men were likely to report decreases in their number of different female sexual partners, their likelihood to have nonprimary female partners, and their number of new sexual partners over the 6 months before visit (P < 0.0001 in all categories). We also found a significant decrease in the proportion of men who reported engaging in anal sex with a nonprimary female partner (P = 0.0036). In addition, there was a decrease in the proportion of men who reported engaging in anal sex with another male (P = 0.0003). However, the sample sizes for these latter 2 analyses were relatively small (8% and 11% of the men at baseline). We also found that although men reported having fewer nonprimary female sexual partners in the prior month (P < 0.0001), their reported decrease in their overall number of vaginal sexual episodes was not significant (P = 0.08; Table 2).
In comparing changes in the men's reported behavior between the 6- and 12-month visits, the only parameters that showed statistically significant reductions were the proportion of men engaging in sex with a nonprimary partner (P = 0.0003) and the proportion of men who had new sexual partners (P < 0.0001). There was additionally a significant decrease in the proportion of men who reported engaging in vaginal sex with a nonprimary female partner in the month before visit (P = 0.001).
Similarly, over the course of a year in study, women tended to report decreases in their number of different male sexual partners and in their number of new male sexual partners. There was also a decrease in the proportion of women who reported engaging in sex with a nonprimary male partner. Examining data for the month immediately before each visit, the proportion of women who reported engaging in vaginal sex with a nonprimary male partner also decreased. P values were estimated to be less than 0.0001 for each of the above-reported behaviors. At baseline, 75% of the women in the cohort reported using condoms when engaging in vaginal sex with a nonprimary partner. There was an increasing trend in the percentage of the time a condom was used among those who used a condom (P = 0.0166; Table 3).
Comparing reported behavior between the 6th and 12th month visits among the women, significant reductions were found in the number of new sexual partners and in the proportion who engaged in anal or vaginal sex with a male partner (P < 0.05 in all categories).
The analyses based on the number of partners before baseline indicated that the men who had 4 or more partners at baseline showed greater reductions in risk through the entire course of the study (P < 0.01 in all categories: BL vs. 6; 6 vs. 12; BL vs. 12; Table 4). However, the men with 3 or fewer partners failed to demonstrate any reductions between the 6th and 12th months (P > 0.05 in all categories: BL vs. 6; 6 vs. 12; BL vs. 12). Similarly the women with 4 or more partners demonstrated greater changes in risk-taking behavior through the course of this study when compared with the women with 3 or fewer partners (Table 4).
The male participants in the study who had not had a commercial sexual encounter in the 6 months before baseline failed to show a decreasing trend for the number of different sexual partners in the past 6 months and also for all 3 parameters under consideration on comparing the 6th and 12th month visits (P > 0.05 in 6 vs. 12; Table 5). The women with no commercial encounters in the 6 months before enrollment had no changes in sexual behavior through the entire course of the study. The participants, both male and female, who reported having had sexual encounters in the 6 months before baseline demonstrated significant decreasing trends for all the parameters considered in the sensitivity analysis (P < 0.0001 in BL vs. 12). The men also demonstrated significant decreases in risk-taking behavior between the 6th and 12th month visits (P < 0.001 in 6 vs. 12). However, the women with history of commercial sexual encounters only demonstrated a significant decrease in the number of different sexual partners between the 6th and 12th month visit (P = 0.032; Table 5).
The HIV epidemic is increasing in India.2 More than 80% of HIV-infected Indians belong to the most sexually active age group where the epidemic is spreading most rapidly.2 Therefore, it is important to target this population in our interventions to control the spread of this disease more effectively.
The current study enrolled high-risk populations. The baseline prevalence of HIV among those who screened for the study was 6.6%, and most participants reported high-risk sexual behavior. A study form Pune, India reported a HIV incidence of 10.2/100 person-years (95% CI: 7.9, 13.1) among the HIV-uninfected patients screened at 2 STI clinics in Pune.14 Another report from Calcutta that used the standardized testing algorithm for recent HIV-1 seroconversion estimated the annualized incidence among a cohort of high-risk individuals to be 7%.15 The baseline risk-taking profiles of many of the participants in this Chennai cohort are comparable to the other high-risk Indian populations, but the current study found HIV incidence after 1 year in this high-risk cohort to be 0.44 per 100 person-years after RRC.
Interventions in the forms of RRC and timely treatment of STIs which were made available to the study participants have been shown to decrease incidence of HIV. Our results indicated that there was indeed a decrease in the risk-taking behavior among the riskiest participants. Some of the people who met the eligibility criteria to be included in this study might not be really considered as high-risk, but met entry criteria because of a recent STI. However, among the riskiest participants, significant behavioral changes were manifested in the course of the study. Among the complete male cohort, there were no significant reductions between the 6- and 12-month visits, as most of the risk reduction took place after the first round of RRC, a finding that has also been demonstrated in other settings.19 However, when male subsets were analyzed, based on the number of different partners at baseline, the men with 4 or more partners at baseline showed significant risk reductions even between the 6th and 12th month visits. We observed a similar pattern when we split our men by whether or not they had engaged in commercial sex in the 6 months before baseline. Among the women with more than 3 partners, the mean number of partners was reduced from 45 to 25 in the first 6 months. However, unlike the men, the women did not demonstrate significant reductions for all the parameters when the 6th and 12th month visits were compared. However, the women who had commercial sexual encounters before baseline had a significant reduction in the number of different partners. A significant number of the women in this study were street-based female sex workers (n = 132) who might be refractory to counseling that endangered their livelihood. It is therefore interesting to note that although they could completely abstain from commercial encounters, they were able to cut down on the number of different partners. The heterogeneity of the women's sexual experience in this cohort underscores the need to develop diverse interventions for the different populations.
RRC, although very effective by itself, can be made even more effective by coupling it with culturally appropriate intervention programs.20,21 The importance of counseling people on alcohol abuse along with sexual RRC has been emphasized in several studies, as alcohol abuse is believed to amplify risky behaviors especially among the youth.22,23 In a recently concluded survey on domestic violence in Chennai, 94% of women who had suffered some form of domestic violence reported their spouse being under the influence of alcohol as one of the most common reasons for the episode; in the same survey, 74% of the women also reported being forced to have sex against their will (unpublished data). Some communities reported unavailability of condoms as a reason for their unsafe sexual practice.24 Herein lies the importance of tailoring RRC for specific target communities or coupling it with other relevant behavioral interventions to improve its efficacy.
One of the limitations of this study was that the inclusion criteria were not specific for age; hence, participants with a wide age range (18-75) were enrolled. There was also a wide range for the number of sexual partners before enrollment. Although these wide variations meant that the cohort was diverse, the heterogeneity of the populations limited the ability to make robust conclusions about some of sociodemographic and behavioral correlates of risk-taking behavior. The subgroup analysis also revealed that a significant number of the participants had no scope to reduce the number of participants beyond the 6th month visit which would affect the overall effect of the impact of RRC in reducing risky behavior. Our data suggest that VCT may be a helpful intervention to decrease risky behavior in some subpopulations, but cannot definitively know whether this was solely due to the RRC or whether other factors like timely management of STIs and free distribution of condoms were involved, or if the participants were demonstrating the epidemiological principle of regression to the mean.
India is a diverse country with high rates of poverty and illiteracy and a burgeoning HIV epidemic. This article describes the utility of individualized sexual RRC tailored in assisting high-risk Indians in reducing their sexual risk-taking behavior.
The authors are also thankful to the study participants and all the study staff affiliated with this study. We also wish to acknowledge the efforts of Family Health International, who oversaw the HPTN 033 protocol development and study conduct, Dr. Benoit Masse and his colleagues at the protocol statistical center at the Fred Hutchinson Cancer Center in Seattle, Washington, the staff at Fenway Community Health, Boston, for their assistance in staff training and protocol monitoring, and Ms. Lola Wright of The Miriam Hospital, for assistance in manuscript preparation.
1. Simoes EA, Babu PG, John TJ, et al. Evidence for HTLV-III infection in prostitutes in Tamil Nadu (India). Indian J Med Res
4. Mehendale SM, Shepherd ME, Divekar AD, et al. Evidence for high prevalence & rapid transmission of HIV among individuals attending STD clinics in Pune, India. Indian J Med Res
5. Gallo RC, Montagnier L. The discovery of HIV as the cause of AIDS. N Engl J Med
6. Sexually Transmitted Diseases Treatment Guidelines 2002. Centers for Disease Control and Prevention. MMWR Recomm Rep
7. Singh YN, Malaviya AN. Experience of HIV prevention interventions among female sex workers in Delhi, India. Int J STD AIDS
8. Basu I, Jana S, Rotheram-Borus MJ, et al. HIV prevention among sex workers in India. J Acquir Immune Defic Syndr
9. Gottlieb SL, Douglas JM Jr, Foster M, et al. Incidence of herpes simplex virus type 2 infection in 5 sexually transmitted disease (STD) clinics and the effect of HIV/STD risk-reduction counseling. J Infect Dis
10. Pao D, Fisher M, Hue S, Dean G, Murphy G. Transmission of HIV-1 during primary infection: relationship to sexual risk and sexually transmitted infections. Aids
11. Plummer FA. Heterosexual transmission of human immunodeficiency virus type 1 (HIV): interactions of conventional sexually transmitted diseases, hormonal contraception and HIV-1. AIDS Res Hum Retroviruses
. 1998;14(Suppl 1):S5-S10.
12. Kumarasamy N, Solomon S, Flanigan TP, Hemalatha R, Thyagarajan SP, Mayer KH. Natural history of human immunodeficiency virus disease in southern India. Clin Infect Dis
13. Bentley ME, Spratt K, Shepherd ME, et al. HIV testing and counseling among men attending sexually transmitted disease clinics in Pune, India: changes in condom use and sexual behavior over time. Aids
14. Mehendale SM, Rodrigues JJ, Brookmeyer RS, et al. Incidence and predictors of human immunodeficiency virus type 1 seroconversion in patients attending sexually transmitted disease clinics in India. J Infect Dis
15. Gupta P, Kingsley L, Sheppard HW, et al. High incidence and prevalence of HIV-1 infection in high risk population in Calcutta, India. Int J STD AIDS
16. Celentano DD, Bond KC, Lyles CM, et al. Preventive intervention to reduce sexually transmitted infections: a field trial in the Royal Thai Army. Arch Intern Med
17. Brookmeyer R, Mehendale SM, Pelz RK, et al. Estimating the rate of occurrence of new HIV infections using serial prevalence surveys: the epidemic in India. Aids
18. Kaul R, Kimani J, Nagelkerke NJ, et al. Reduced HIV risk-taking and low HIV incidence after enrollment and risk-reduction counseling in a sexually transmitted disease prevention trial in Nairobi, Kenya. J Acquir Immune Defic Syndr
19. Vanichseni S, Des Jarlais DC, Choopanya K, et al. Sexual risk reduction in a cohort of injecting drug users in Bangkok, Thailand. J Acquir Immune Defic Syndr
20. Hogg RS, Strathdee S, Kerr T, Wood E, Remis R. HIV prevalence among Aboriginal British Columbians. Harm Reduct J
21. Aguilera S, Plasencia AV. Culturally appropriate HIV/AIDS and substance abuse prevention programs for urban Native youth. J Psychoactive Drugs
22. Simbayi LC, Kalichman SC, Jooste S, Mathiti V, Cain D, Cherry C. Alcohol use and sexual risks for HIV infection among men and women receiving sexually transmitted infection clinic services in Cape Town, South Africa. J Stud Alcohol
23. Butcher AH, Manning DT, O'Neal EC. HIV-related sexual behaviors of college students. J Am Coll Health
24. Kusseling FS, Shapiro MF, Greenberg JM, Wenger NS. Understanding why heterosexual adults do not practice safer sex: a comparison of two samples. AIDS Educ Prev