Risk factors associated with chlamydia and gonorrhoea infection among female sex workers in two Mexico–USA border cities

INTRODUCTION

Tijuana, Baja California and Ciudad Juarez, Chihuahua lie on Mexico's border with the USA and are characterized by large populations of migrants and female sex workers (FSWs). Both cities have ‘zones of tolerance’ where sex work is quasilegal. Clients include men from both sides of the USA–Mexico border and from other countries, ranging in age from 18–80 years old. ^1,2 In 2006, it was reported that 4850 Tijuana FSWs were registered with the Municipal Health Service, while thousands of others are thought to work without a permit. ³ Approximately 4000 FSWs work in zones of tolerance of Ciudad Juarez, where sex work is not regulated by the Health Department. ³

Prevalence estimates for chlamydia and gonorrhoea among FSWs in Mexico have been reported in a few studies. ^4,5 However, data are sparse among FSWs who reside near the Mexico–USA border. Among 354 registered FSWs in three northern non-border Mexican cities (Durango, Durango; Zacatecas, Zacatecas; and Torreón, Coahuila), the prevalence of cervical Chlamydia trachomatis infection was 12.4%. ⁶ In a 1993 study among 826 sex workers in Mexico City, 3.7% tested positive for gonorrhoea. ⁴

Chlamydia and gonorrhoea share some risk factors, such as female gender, younger age at sexual debut, low contraceptive use, greater number of sexual partners, engaging in commercial sex work and drug use. ^7–12 Recent studies have also identified partner-level risk factors; for example, in a sample of 15–24 year olds, partners who had a sexually transmitted infection (STI) in the past year and partnerships with an age difference of five years or more were more predictive of STIs than were the individuals' own sexual behaviours. ¹³ STI risks may also differ by sexual partner type. Among 553 high-risk heterosexual women in Houston, TX, condom use was higher with casual partners than with steady partners; however, this advantage was reversed when alcohol, drugs or both were used. ¹⁴

Although the number of studies has been fewer, there has been some research on partner-level characteristics that may influence STI acquisition in countries or communities that are specifically low- or middle-income. For example, a study of 4000 married or co-habiting men living in Mexico City found that those having extra-relational sex in the last year were more likely to use condoms with their secondary partners (e.g. FSWs) than with their primary partners. ¹⁵

In Mexico, certain population subgroups show high prevalence of chlamydia and gonorrhoea varying by region. ^16,17 Of particular interest for our study was that women in communities along the border between Arizona and the Mexican state of Sonora were at significantly increased risk for chlamydia infection compared with women attending clinics in non-border locations (9.0% versus 5.4%). Testing positive for chlamydia was associated with younger age, a history of new sexual partner(s) in the previous three months and proximity to the international border. ¹⁸

Major drug trafficking routes for heroin, cocaine and methamphetamine pass through both Tijuana and Ciudad Juarez. We previously reported that FSWs in these cities who injected drugs had higher overall STI prevalence and engaged in more unsafe sex with clients compared with FSWs who did not inject drugs. ¹⁹ In addition, FSWs with USA clients were significantly more likely to inject drugs, have unprotected sex for more money and to have recently tested positive for one of four STIs (gonorrhoea, chlamydia, HIV, active syphilis) compared with other FSWs. ²⁰ However, in that study we did not examine individual- or partner-level risk factors for STI acquisition, nor did we examine risk factors for gonorrhoea and chlamydia separately. In the present analysis, we examine whether FSWs testing positive for gonorrhoea and chlamydia report higher rates of drug use before or during sex and whether partner-level characteristics are closely associated with these incident STIs. In a resource-limited setting such as Mexico, identifying correlates of STIs is important for targeting funding for STI screening and treatment.

MATERIALS AND METHODS

RESULTS

Of 924 participants, 126 had samples that were spoilt due to extreme heat and border crossing delays; their chlamydia and gonorrhoea test results were thus excluded. Women who were excluded for this reason were significantly older, more likely to be from Ciudad Juarez and to be street workers, were less likely to be migrants, reported significantly less unprotected vaginal sex with any type of client and earned less for sex with and without a condom. Of the 798 remaining, 466 (58.4%) worked in Tijuana and 332 (41.6%) in Ciudad Juarez. The median age was 32 years (interquartile range [IQR]: 26, 39); 17.8% had ever injected drugs; and the median number of clients in the past six months was 240 (IQR: 72, 480). Overall, 104 (13.0%) tested positive for chlamydia, and 51 (6.4%) tested positive for gonorrhoea.

DISCUSSION

In this study of high-risk FSWs residing in two Mexico–USA border cities, chlamydia and gonorrhoea were more closely associated with drug use among FSWs and their sexual partners than with FSWs' sexual behaviours. Specifically, use of illegal drugs before or during sex and having male IDU clients were independently associated with gonorrhoea infection, and recent methamphetamine injection was independently associated with chlamydia infection. Odds of chlamydia and gonorrhoea infection were two to four times higher in Tijuana than in Ciudad Juarez. These findings have implications for targeting STI prevention and treatment to subgroups of FSWs who use illicit drugs in these resource-limited settings, especially in Tijuana.

We found that methamphetamine injection was an independent predictor of chlamydia infection. Use of methamphetamine has been associated with increased libido and numbers of sexual partners, as well as with high-risk sexual behaviour. ^23–26

Although methamphetamine use is becoming more common among heterosexual women in countries such as the USA, the high prevalence of methamphetamine injection in our sample (21%) is unusual.

While gonorrhoea infection was not associated with the use of methamphetamine or other drug types, it was independently associated with using an illegal drug before or during sex. A critical review of 16 epidemiological studies associated drug use with increasing rates of STIs, including gonorrhoea. ²⁷ Other studies have documented an association between substance use and lower rates of condom use or impaired ability to use condoms. ^14,28 Some of the authors have initiated a new intervention study that aims to promote skills in negotiating condom use among FSWs in Tijuana and Ciudad Juarez who inject drugs, and we hope this will lead to a significant reduction in HIV/STI incidence.

Having an IDU sex partner in the past six months was the only partner-level variable to be associated with incident STIs in our study, and this relationship was observed only for gonorrhoea. The lack of independent association of chlamydia and gonorrhoea with other partner characteristics may be due to limited study power, and the fact that women were asked to report their partner characteristics. Future studies should consider obtaining such data directly from FSWs' clients and intimate partners, since FSWs' reports of their partners may not be a suitable proxy.

Our results support the existing literature linking younger age with chlamydia infection. In a systematic review of risk factors for chlamydia among women in the UK and Ireland, younger age was one of the most important determinants. ²⁹ Our findings reinforce the need for increased screening for chlamydia among young FSWs, especially those who inject methamphetamine.

Interpretation of these findings should consider a number of limitations. First, we relied on the accuracy of self-report, which could result in under-reporting, especially for variables relating to partner characteristics. Further, given that data were gathered as part of an intervention study to reduce sexual risk, only women reporting recent unsafe sex were included, which may have biased associations with condom use towards the null. Specimen spoilage led us to exclude samples from 126 FSWs, some of whom appear to have been at higher risk. This may have caused some selection bias that influenced our results in unanticipated ways; however, this would tend to affect external validity and not the internal validity of our findings. Finally, since this analysis was cross-sectional, no causal inferences can be drawn; however, since nucleic acid amplification detected active infection, the variables we found to be associated with chlamydia and gonorrhoea may be true risk factors.

Given that both chlamydia and gonorrhoea are co-factors of HIV transmission, ³⁰ there is a need to scale up interventions to increase testing, diagnosis and treatment for STIs, especially among drug-using FSWs and their partners in these cities. Since Tijuana and Ciudad Juarez are experiencing rising rates of HIV infection among FSWs, ^31,32 identifying and intervening on risk factors for chlamydia and gonorrhoea have implications not only for STI control, but also for containing the burgeoning HIV epidemic.