Rugpao, Sungwal; Beyrer, Chris*; Tovanabutra, Sodsai†; Natpratan, Chawalit‡; Nelson, Kenrad E.*; Celentano, David D.*; Khamboonruang, Chirasak†
There is compelling evidence that consistent and correct use of male condoms can be effective in preventing transmission of sexually transmitted pathogens, including HIV-1(1-6). Condom breakage and slippage, however, can reduce the effectiveness of this barrier method, leading to unwanted pregnancies in heterosexual sex and sexually transmitted diseases(STD) via vaginal, anal, or oral sex(7,8). Studies have examined condom breakage and slippage rates in developed and developing country settings among diverse populations including married couples, male homosexual couples, female and male commercial sex workers (CSWs), and STD clinic users(7-11). Rates of condom breakage and slippage vary widely across these studies, from <1% to >12% breakage per 100 sex acts in developed country settings, and considerably higher in developing ones(7-11). The mechanisms of condom failure are poorly understood, and few preventive interventions targeting this important route of exposure have been attempted.
We have previously reported rates of condom breakage and slippage during heterosexual commercial sex in northern Thai brothels; laboratory examination of 5,559 condoms found a breakage rate of 5.9% (95% confidence interval [CI] 5.3-6.5)(12). Self-reported condom breakage and slippage rates were 3.9% and 0.4%, respectively, underscoring the need for laboratory examination of used condoms. This study was conducted in 1992, one year after the initiation of the "100% Condom Campaign" of the Thai Ministry of Public Health. Reported condom use in Thai brothels was already high (>80%), but quantification of actual use and measures of quality of use, were lacking(13). In the present study, we evaluated mechanisms of condom breakage and slippage among brothel-based sex workers to verify condom use self-reports among CSWs, and to identify targets for interventions to reduce condom failure in this high-risk setting.
The study was conducted in Lamphun Province, upper northern Thailand, from August-October 1995 and enrolled 68 female CSWs in seven brothels in Lamphun City. These brothels are characterized by low to moderate price, direct sale of sex on brothel premises, largely local Thai clientele, active participation in the"100% Condom Campaign" and in provincial STD control programs, and high (60%) rates of HIV-1 infection among CSWs. Subjects were female adult Thai volunteers recruited by outreach workers, and informed consent was obtained. HIV status of the women was not measured. The study was approved by the ethical review boards of Chiang Mai University and Johns Hopkins University.
Condom Collection and Examination
CSWs were instructed to place condoms and wrappers in provided containers after each sex act. One container per client was used and CSWs noted the order of use. A unique identifier was assigned to each condom with a code including CSW, establishment, and order of the client served. Used condoms were examined for breakage and for types and size of breaks as previously described(12). Briefly, condoms were examined by direct visual inspection and by two tests; a tap water continence test and an air inflation test, to look for leaks, tears, and pinholes. The width and length of any identified breaks were measured. Exposure breaks were any break with one condom, both broken with two condoms, and all broken with more than two condoms used per sex act. Condoms that had been used multiply were separated manually by examiners after brief soaking to prevent tears during separation. The CSWs kept coital logs and recorded all clients and sex acts, condom breaks and condom slippage events.
When laboratory examination identified a condom break, the woman with whom the break occurred was notified and interviewed the next day regarding the sex act(s) in which the break occurred. She was also interviewed about the first and last sex acts of the same day for intrawoman comparison. CSWs who had only intact condoms by examination on the same night of work were randomly selected to serve as controls for interwoman comparison. Structured interviews addressed client characteristics and sexual activities, numbers of clients, condom use with each client, reasons for condom use and nonuse, and substance use by clients. Where a condom break was detected. CSWs were asked about their knowledge of the event and of the circumstances involved.
Breakage and slippage rates for single and multiple use were compared using odds ratios (OR) and 95% confidence intervals(95% CI). Two case-control analyses were done. In the first, cases were defined as CSWs experiencing a laboratory-confirmed broken condom and controls were randomly selected CSWs who did not have a break on the same night of work. The second case-control analysis was done to evaluate client and sex act factors for condom outcomes. Cases were sex acts in which a condom break was detected. Controls were sex acts with the same CSW but with customers (the first and last of the same night on which a break occurred, if these were not the clients with the break) where the condom remained intact. To add statistical power, controls were also randomly selected from among sex acts involving other CSWs who also had only intact condoms on the same night as the break. CSW, client, and sex behavior factors related to breakage were assessed using multiple logistic regression analysis.
The 68 participating CSWs had minimal education(47.1% had primary education or less), were young(mean age 25 yr ± 5.4) and most (66.1%) had been married. Mean number of months working as a CSW was 35.1, with a median of 24 months. The mean age at first CSW work was 21.1 yr ± 5.5. Prices for sexual service averaged 240 Baht, ranging from 40-950 Baht(25 Baht = US $1.00). The women had a median of two clients per day, ranging from 1-10. Virtually all reported exclusively vaginal intercourse with clients. Most reported having had an STD episode in the last year, including gonorrhea (45.6%), pelvic inflammatory disease (29.4%), a genital ulcer (29.3%), syphilis(13.2%), nongonococcal cervicitis (13.2%), and genital warts (8.8%).
A total of 7,594 condoms were examined from 4,734 client visits, representing 5,040 sex acts (Table 1). Of the reported sex acts, 30 (0.6%) were without condoms and 5,010 (99.4%) were with at least one condom. Single condoms were used in 2,485 sex acts(49.3%), two condoms in 2,469 (49.0%), and more than two in 56 sex acts (1.2%) (Table 2).
Condom slippage events were rare, 5 of 5,010 sex acts (0.1%) and did not vary by number of condoms used (see Table 2). There were 45 breaks with single condom use, a rate of 1.8% (95% CI 1.4-2.3), four exposure breaks with two condoms, a rate of 0.2%(95% CI 0.02-0.4), and no breaks with more than two condoms. Single condom use was significantly more likely than multiple use to lead to an exposure break, OR 11.4 (95% CI 3.9-37.2). Condom breaks were found in four regions of condoms; 31.7% were at reservoir tips, 40.2% just below the reservoir, 18.3% along the central (shaft) section, and 9.8% at the condom base.
In the first case-control analysis, coital log data were used to compare women reporting a break with women not reporting a break the same night. Seventy cases and 70 controls were identified, who did not differ significantly in terms of overall condom usage, average price for sex, extra payment not to use condoms, substance abuse on the night of the break, or wearing of rings. Cases were more likely to have had more than two clients on the night of the break, 51.4% of cases had more than two clients, whereas 34.4% of controls reported more than two clients, OR 2.03(95% CI 1.03-4.0).
Natural experiments are unusual in scientific research and can yield important and unexpected findings. We found a considerable reduction in condom breakage over a 3-year-period in northern Thailand, from about 6% in 1992 to <2% 3 years later. Multiple condom use increased from 2.8% to 50.4% over the same period of time. Although the breakage rate for single use in this study was considerably lower than in 1992, the exposure breakage rate for multiple use was lower still, 0.2%, suggesting that multiple condom use was responsible for a significant component of the decline in condom failure.
We believe we have serendipitously measured a natural experiment in which CSWs and their clients have reduced condom breakage rates through multiple condom use. To confirm that multiple condom use was indeed occurring, and not multiple sex acts with the same client in which sequential condoms were worn, CSWs, brothel managers, and several clients were interviewed. These qualitative interviews revealed that multiple condom use for single sex acts was indeed common, and that this behavior was initiated by the CSWs themselves in >75% of condom negotiations before sex acts. This finding suggests that single and multiple condom users may not be behaviorally different groups of clients; multiple condom use is based on the CSW's ability to negotiate this use, and her perception of a client's HIV/STD risks. Laboratory staff who examined the condoms confirmed that multiple condoms were usually"stuck" together, and had to be separated for individual examination.
There has been no identifiable intervention to promote multiple condom use in this population, nor is this a part of the "100% Condom Campaign"(13). Rather, this appears to have been a community response to condom promotion messages and to fears of HIV infection.
It is not clear why the single condom use breakage rate was lower in 1995 than in 1992. Pinhole breaks, common manufacturing defects, did not differ across the studies, suggesting that condom use itself had improved, not condom quality. Several factors may be involved, including experience among men with condom use, greater ability among CSWs to use condoms effectively, and more appropriate lubricant use.
Behavioral and condom use factors associated with condom breakage were related to client behavior and sex act characteristics. These included men having penile foreskin implants, younger client age, longer than average duration of intercourse, sexual services late in the night, and repeated acts of intercourse with the same clients. None of these would seem likely intervention targets in brothel settings, although double condom use might reduce the risk of condom breakage for higher-than-average intensity of sex and multiple acts per client. Indeed, our data support this, given the very low rate of breaks leading to exposure in sex acts in which clients wore two or more condoms. These low rates of breakage and slippage are encouraging, as is the very high rate of condom use. Taken together, these findings suggest that consistent use of male condoms in Thai brothels is now normative behavior. A recent study of declines in HIV infection and risk behaviors, and increasing condom use, among young Thai men suggests that these high condom rates may be affecting the Thai HIV epidemic(14). The use of multiple condoms further suggests that many Thai men might be willing to accept multiple condom use to increase protection from sexually transmitted pathogens. Male condoms that offered greater protection are likely to be well accepted in this motivated population, despite potential decreases in male sexual pleasure.
Acknowledgement: Supported, in part, by funds from the National Institutes of Health (R21 AI33862), and the Royal Thai Government. We acknowledge the assistance of Philip Guest and Anthony Pramulratana for their input into the design of the study, Melanie de Boer and Linda Sussman of Johns Hopkins University for assistance with data management, Tasanai Vongchak and Yupadee Yutabootr of RIHES for assistance with field coordination, and the Lamphun Provincial STD Clinic staff for assistance with outreach.
1. De Vincenzi I. The European Study Group on Heterosexual Transmission of HIV. A longitudinal study of human immunodeficiency virus transmission by heterosexual partners. N Engl J Med 1994;331:341-6.
2. Mann J, Quinn TC, Piot P. Condom use and HIV infection among prostitutes in Zaire. N Engl J Med 1986;316:345.
3. Ngugi EN, Plummer FA, Simonsen JN, et al. Prevention of transmission of human immunodeficiency virus in Africa: effectiveness of condom promotion and health education among prostitutes. Lancet 1988;2:887-890.
4. Allen S, Serufilira A, Bogaerts J, et al. Confidential HIV testing and condom promotion in Africa: impact on HIV and gonorrhea rates. JAMA 1992;268:3338-43.
5. Estebanez P, Fitch K, Najera R. HIV and female sex workers. Bull World Health Organ 1993;71:397-412.
6. Lazzarin A, Saracco A, Musicco M, Nicolosi A. Manto-woman sexual transmission of the human immunodeficiency virus: risk factors related to sexual behavior, man's infectiousness, and woman's susceptibility. Arch Intern Med 1991;151(12):2411-6.
7. Liskin L, Wharton C, Blackburn R, Kestelman P. Condoms—now more than ever. Popul Rep 1990; series H, 8:8-12.
8. Albert AE, Hatcher RA, Graves W. Condom use and breakage among women in a municipal hospital family planning clinic. Contraception 1991;43:167-75.
9. Albert AE, Warner DL, Hatcher RA, Trussell J, Bennett C. Condom use among female commercial sex workers in Nevada's legal brothels. Am J Public Health 1995;85:1514-20.
10. Richters J, Donovan B, Gerofi J. How often do condoms break or slip off in use? Intern J Sex Transm Dis AIDS 1993;4:90-4.
11. Thompson JLP, Yager TJ, Martin JL. Estimated condom failure and frequency of condom use among gay men. Am J Public Health 1993;83:1409-13.
12. Rugpao S, Pruithithada N, Yutabootr Y, Prasertvitayakit W, Tovanabutra S. Condom breakage during commercial sex in Chiang Mai. Thailand. Contraception 1993;48:537-47.
13. Hanenberg R, Rojanapithayakorn W, Kunasol P, Sokal D. Impact of Thailand's HIV-control programme as indicated by the decline of sexually transmitted diseases. Lancet 1994;344:243-5.
14. Nelson KE, Celentano DD, Eiumtrakul S, Hoover D, Beyrer C, Suprasert S, Kuntolbutra S, Khamboonruang C. Changes in sexual behavior and a decline in HIV infection among young men in Thailand. New Engl J Med 1996;335:297-303.
© Lippincott-Raven Publishers.