Few behavioural interventions on condom use for sex workers have been evaluated on their effects for 5 years or more [1–4]. Interventions for brothel-based sex workers in Africa have increased consistent condom use to 78%  and 80.7% , with a concomitant decline in HIV and sexually transmitted infections (STI). The intervention in Thailand increased condom use among brothel-based sex workers from 14% in 1989 to 94% in 1993 [1,4], with a concomitant decline in HIV incidence , but subsequent evaluations reported lower levels of condom use [5–7] and high HIV rates [8,9] in some provinces. Condom use among sex workers should be maintained close to 100%, as transmission can occur even with a low level of non-condom use because of their high number of partners and high HIV infection rates .
It is unclear why these programmes did not achieve higher condom use levels as programme activities were not evaluated. Given the increasing evidence of HIV spread by oral sex [11,12], condom promotion programmes for oral sex should also be evaluated. More research is needed to identify interventions that would sustain condom use for all modes of intercourse to high levels among sex workers.
In Singapore, condom promotion programmes for vaginal [13–15] and oral [16,17] sex among female brothel-based sex workers have produced sustained increases in condom use, with corresponding declines in cervical and pharyngeal gonorrhoea incidence. The effects of both programmes from 1990 to 2002 are reported here.
In 2003, approximately 1100 female brothel-based sex workers worked in Singapore. They are required to undergo regular screening for gonorrhoea, chlamydial infections, HIV and syphilis. The condom promotion programme for vaginal sex, implemented as a pilot project in 1994 [13,14], was expanded to all sex workers in 1995. In mid-1996, the condom promotion programme for oral sex [16,17] was incorporated into the programme because of the marked rise in oral sex ; since then, all newly recruited sex workers were enrolled monthly in the combined interventions.
Programme development was based on (i) Green's PRECEDE PROCEED framework  to change sex workers’ risk behaviour and create a supportive work environment; (ii) Bandura's self-efficacy theory  to develop sex workers’ condom negotiation skills; (iii) continuous quality-improvement principles; and (iv) findings from our studies on condom use for vaginal and oral sex among sex workers [21–25]. Interventions for sex workers consisted of talks on STI and AIDS, problem-solving sessions and video demonstrations of condom use and negotiation techniques, followed by reinforcement sessions held over 2 years. Interventions for brothels included talks and administrative measures, with temporary closure of brothels with high STI rates. Programme activities were monitored and regular meetings were held with health staff and peer leaders of sex workers to provide feedback and involve them in solving problems.
A pre-test/post-test comparison group followed by a time series design was used to compare trends in 6-month post-intervention condom use for vaginal sex and cervical gonorrhoea incidence from 1990 to 2002 across cross-sectional samples of sex workers before and after programme implementation in 1995. The subsequent condom promotion programme for oral sex implemented in 1996 was evaluated using an interrupted time series design with a retrospective pre-test/post-test matched control group design. Serial independent cross-sectional surveys rather than repeat surveys within cohorts were used to assess long-term programme effects because three-quarters quit sex work by 1.5 years.
After informed consent was obtained, participants completed a simple questionnaire on demographics and condom use just before and 6 months after the educational sessions. To avoid over-reporting of condom use, health staff stressed the confidentiality of the information. They also explained that the results would be used to plan a better STI prevention programme for them and that ‘100% condom use’ answers were not expected because they were aware of difficulties faced by sex workers in negotiating with clients.
Consistent condom use was defined as using a condom with every act of oral or vaginal sex with clients in the last working week. Monthly endocervial and pharyngeal swabs were taken from the sex workers for culture for Neisseria gonorrhoeae. Gonorrhoea incidence was used as an objective indicator of the impact of intervention on condom use because of its short incubation period, short duration of the infection, high specificity and sensitivity of the diagnostic test, and rapid and effective response to treatment. In contrast, other STI such as chlamydial infections may become chronic and are prone to relapses, hence might not be influenced by recent condom use. HIV was not used as an outcome measure because of its low incidence (0.02–0.04/100 person-months) among the sex workers.
The 6-month post-intervention cervical and pharyngeal gonorrhoea incidence rate was calculated by dividing the number of new cases of gonorrhoea by the sum of the time periods of observation for all sex workers (person-months) who completed follow-up.
Sex workers also rated the health education methods and messages. In-depth interviews were held with 22 randomly selected participants to explore the process leading to their sustained condom use.
The chi-square test for trend was used to compare proportions in condom use and gonorrhoea incidence across independent cross-sectional samples over time. Cox's proportional hazards regression model  was used to assess the independent effect of the time period of the survey on gonorrhoea incidence rates, controlling for confounders. The Cox regression model modified for cross-sectional data [27,28] was used to estimate the adjusted prevalence ratios of condom use.
From 1995 till June 2002, 2737 currently working and newly recruited sex workers (mean age, 25.1 years) were enrolled in the intervention. The majority (> 90%) were non-locals, being Malaysian Chinese and Thai. From July 1996 to June 2002, 1986 newly recruited sex workers were enrolled in the programme, with a mean of 27 per month. About one-third (32.2%) quit work before the 6-month follow-up. Of those still in sex work, 93.5% (1259) completed follow-up. They did not differ significantly from those who quit or were lost to follow-up in age (25.1 versus 24.9 years) and number of clients per day (10.7 versus 10.2).
Condom use for vaginal sex and pharyngeal gonorrhoea
Consistent condom use among newly recruited sex workers (worked < 6 months) increased by more than twofold from < 45% before programme implementation in 1995 to > 90% after 1996 (P < 0.001), reaching 95.2% in 2002 (Fig. 1). Cervical gonorrhoea incidence declined from 30–42/1000 person-months pre-intervention to < 5/1000 person-months since 2000 (P < 0.001). Adjustment for temporal changes in sociodemographic characteristics did not materially alter the trends.
As reported previously , the intervention group at 5-month follow-up was twice as likely as the comparison group to use condoms always (adjusted prevalence ratio 1.90; 95% confidence interval, 1.22–2.94). Cervical gonorrhoea incidence declined significantly in the intervention group compared with a non-significant decline in the comparison group.
Condom use for oral sex and pharyngeal gonorrhoea
Consistent condom use for oral sex increased from < 50% pre-intervention to 87% in 1999 (P < 0.001), following implementation of the condom promotion programme for oral sex (Fig. 2). When measures targeting brothels were subsequently withheld, condom use decreased to 79.7% (P = 0.081). After intensification of brothel measures in 2000, condom use increased to 97.1% in 2002 (P < 0.01). Pharyngeal gonorrhoea incidence declined from 12.5 to 16.6/1000 person-months pre-intervention to 4.7/1000 person months in 2002 (P < 0.01), with a peak corresponding to the decline in condom use when brothel interventions were withheld.
As reported previously , sex workers exposed to brothel interventions showed a significant increase in oral condom use to 92.5%, compared with a decrease among matched controls without brothel interventions. The pharyngeal gonorrhoea incidence rate was also lower than the control group (adjusted risk ratio, 0.22; 95% confidence interval, 0.06–0.78).
Video demonstrations of condom negotiation strategies for vaginal and oral sex were rated very useful by 98.5% of sex workers, compared with < 40% for other educational methods. They attributed their sustained condom use to the skills learnt from the interventions and support from the health staff.
Consistent condom use of > 90% for vaginal and oral sex was sustained, with corresponding declines in cervical and pharyngeal gonorrhoea incidence, among female brothel-based sex workers in Singapore after implementation of condom promotion programmes for vaginal and oral sex over 8 and 7 years respectively.
As there was no control group, maturation, increased awareness of condom use over time or public education could have explained the observed increase. However, changes in condom use over the 5-year pre-intervention period were negligible compared with the significant increase post-intervention, suggesting that increased awareness could not have contributed solely to the increase. There was no public education targeting clients or sex workers because of local sensitivities. Self-reported condom use was also corroborated by high self-reported condom use for vaginal and oral sex (98.2 and 92.6%, respectively) in a recent survey on 400 clients of brothel-based sex workers.
Continuous quality improvement principles were incorporated to monitor and improve the programme activities that have not been reported elsewhere. This led to early detection of (i) condom-use-related problems; (ii) inadequate measures targeting brothels; and (iii) the unexpected marked increase in oral sex and pharyngeal gonorrhoea. The last two findings prompted the strengthening of brothel measures and the implementation of a specific condom promotion programme for oral sex. We, therefore, recommend that continuous quality improvement principles be incorporated into on-going programmes for brothel-based sex workers in other countries to sustain condom use.
We thank the health advisors of the Department of STI Control for conducting the educational sessions and assisting in the survey. We are grateful to Ms Sharon Wee and Dr Jayabaskar for contribution to data entry and maintenance of the database for the past 12 years.
Sponsorship: This work was supported by grants from the National Medical Research Council, Singapore (NMRC/R186000036213/1997-2000; NMRC/R186000047213/2001-2002) and Action for AIDS, Singapore: 1993–1994.
1. Hanenberg RS, Rojanapithayakorn W, Kunasol P, Sokal DC. Impact of Thailand's HIV control program as indicated by the decline of sexually transmitted diseases.Lancet
2. Ghys PD, Diallo MO, Ettiegne-Traore V, et al
. Increase in condom use and decline in HIV and sexually transmitted diseases among female sex workers in Abidjan, Cote d'lvoire, 1991–1998.AIDS
3. Alary M, Mukenge-Tshibaka L, Bernier F, et al.Decline in the prevalence of HIV and sexually transmitted diseases among female sex workers in Cotonou, Benin, 1993–1999.AIDS
4. Rojanapithayakorn W, Hanenberg R. The 100% condom program in Thailand.AIDS
5. van Griensven GJP, Limanonda B, Ngaokeow S, Isarankura na Ayuthaya S, Poshyachinda V. Evaluation of a targeted HIV prevention program among female commercial sex workers in the south of Thailand.Sex Transm Infect
6. Visrutaratna S, Lindan CP, Sirhorachai A, Mandel JS. `Superstar’ and ‘model’ brothels: developing and evaluating a condom promotion program for sex establishments in Chiang Mai, Thailand.AIDS
(suppl 1): 569–575.
7. Ford N, Koetsawang S. A pragmatic intervention to promote condom use by female sex workers in Thailand.Bull WHO
8. Kilmarx PH, Palanuvej T, Limpakarnjanarat K, Chitvarakorn A, St Louis ME, Mastro TD. Seroprevalence of HIV among female sex workers in Bangkok: evidence of ongoing infection risk after the ‘100% condom program’ was implemented.J Acquir Immune Defic Syndr
9. Ungchusak K, Rujiviput V, Saengwonloey O. HIV serosurveillance in Thailand. Results of the 15th round, June 1997.Wkly Epidemiol Surveil Rep
(Thai) 1997, 28
10. Celentano DD, Nelson KE, Suprasert S, Eiumtrakul S, Tulvatana S, Kuntolbutra S, et al.Risk factors for HIV-1 seroconversion among young men in Northern Thailand.JAMA
11. Rothenberg RB, Scarlett M, del Rio C, Reznik D, O'Daniels C. Oral transmission of HIV.AIDS
12. Stephenson J. HIV risk from oral sex higher than many realize.JAMA
13. Wong ML, Chan R, Koh D, Wong CM. Theory and action for effective condom promotion. Illustrations from a behavior intervention project for sex workers in Singapore.Int Q Commun Health Educ
14. Wong ML, Chan Roy, Lee J, Koh D, Wong C. Controlled evaluation of a behavioral intervention program on condom use and gonorrhea incidence among sex workers in Singapore.Health Educ Res
15. Wong ML, Chan Roy KW, Koh D. A sustainable behavioral intervention to increase condom use and reduce gonorrhea among sex workers in Singapore: 2-year follow-up.Prev Med
16. Wong ML, Chan RKW, Koh DSQ, Wong S, Wee S, Lee E. Designing an oral sex condom promotion program for female sex workers in Singapore.Int Q Commun Health Educ
17. Wong ML, Chan RKW, Koh D. Promoting condoms for oral sex: impact on pharyngeal gonorrhea among female brothel-based sex workers.Sex Transm Dis
18. Wong ML, Chan RKW, Koh D, Wee S. Increase in oral sex and pharyngeal gonorrhea: an unintended effect of a successful condom program for vaginal sex.AIDS
19. Green LW, Kreuter M. Health Promotion Planning: An Educational and Environmental Approach.
Palo Alto: Mayfield; 1991:125–149.
20. Bandura A. Social Foundations of Thought and Action
. Englewood Cliffs, NJ: Prentice Hall; 1986.
21. Wong ML, Tan TC, Ho ML, Lim JY, Wan S, Chan R. Factors associated with sexually transmitted diseases among prostitutes in Singapore.Int J STD AIDS
22. Wong ML, Tan TC, Ho ML, Lim JY, Lim R, Wan S, et al.Knowledge and sexual behavior related to HIV and AIDS of female sex workers in Singapore.Health Educ J
23. Wong ML, Archibald C, Chan R, Goh A, Tan TC, Goh CL. Condom use negotiation among sex workers in Singapore: findings from qualitative research.Health Educ Res
24. Wong ML, Chan RKW, Wee S. Sex workers’ perspectives on condom use for oral sex: a qualitative study.Health Educ Behav
25. Wong ML, Chan RKW, Koh D, Wee S. Factors associated with condom use for oral sex among female brothel-based sex workers in Singapore.Sex Transm Dis
26. Cox DR. Regression models and life tables.J R Stat Soc Ser
27. Lee J. Odds ratio or relative risk for cross-sectional data.Int J Epidemiol
28. Axelson O, Fredrikson M, Ekberg K. Use of the prevalence ratio versus the prevalence odds ratio as a measure of risk in cross sectional studies.Occup Environ Med