Substantial Increases in Chlamydia and Gonorrhea Positivity Unexplained by Changes in Individual-Level Sexual Behaviors Among Men Who Have Sex With Men in an Australian Sexual Health Service From 2007 to 2013

Abstract

Background 

To determine the risk-adjusted temporal trend of gonorrhea and chlamydia positivity and associated risk behaviors among men who have sex with men (MSM) attending a sexual health clinic in Melbourne in Australia.

Methods 

Gonorrhea and chlamydia positivity by anatomical site adjusted for year of test, age, number of sexual partners, and condom use among MSM attending Melbourne Sexual Health Centre from 2007 to 2013 were calculated using generalized estimating equation regression models.

Results 

A total of 12,873 MSM were included with a median age of 30.0 years. The proportion with pharyngeal, urethral, and anal gonorrhea was 1.7%, 2.3%, and 2.9%, respectively. The adjusted odds of gonorrhea positivity increased by 9% (95% confidence interval [CI], 3%–15%), 11% (95% CI, 6%–17%), and 12% (95% CI, 7%–17%) per year, respectively. The proportion of MSM who were infected with anal chlamydia was 5.6%, with an average increase of 6% (95% CI, 3%–10%) per year; however, no significant change was observed in urethral chlamydia positivity (adjusted odds ratio, 1.02; 95% CI, 0.98–1.06). Increases in gonorrhea and chlamydia positivity were primarily restricted to MSM who reported more than 10 partners in 12 months. The number of partners in the last 12 months fell from 16.6 to 10.5, whereas consistent condom use with casual partners decreased from 64.6% to 58.9% over the study period.

Conclusions 

Gonorrhea and chlamydia have increased among MSM despite the decrease in the number of sexual partners and are occurring primarily in MSM with high numbers of partners and persist after adjusting for known risk factors, suggesting that unmeasured factors (e.g., more assortative mixing patterns) may explain the observed changes.

In Brief

The rapid rises of gonorrhea and chlamydia among men who have sex with men are primarily restricted to those men who have sex with men with high numbers of partners and may potentially be due to changes in sexual mixing patterns.

Neisseria gonorrhoeae (gonorrhea) and Chlamydia trachomatis (chlamydia) are significant sexually transmitted infections (STIs) in their own right but also facilitate HIV transmission.¹ Notifications of HIV, gonorrhea, and chlamydia have been increasing among men who have sex with men (MSM) in many countries including Australia in recent years.^2–4 A previous study has shown that the prevalence of chlamydia (anal: 5.4%; urethral: 3.7%) and gonorrhea (anal: 3.1%; pharyngeal: 1.8%) among MSM attending a sexual health clinic in Melbourne did not change over time from 2002 to 2009.⁵

In an effort to reduce rates of STIs in MSM, many countries have introduced STI testing guidelines. The current Australian STI guidelines⁶ recommend STI screening at least once a year, including tests for gonorrhea (pharyngeal and anal) and chlamydia (urethral and anal) and more frequently (i.e., up to 4 times a year) for MSM who are at higher risk for STIs (e.g., MSM who have engaged in any unprotected anal sex or group sex, have had >10 sexual partners in the last 6 months, or are HIV positive). The proportion of HIV-negative MSM who have had STI testing (excluding blood tests for HIV, syphilis, and hepatitis) in the last 12 months has increased from 45.8% in 2003 to 61.9% in 2013.^7,8 Screening may, if all other factors remained equal, put downward pressure on STI rates, particularly those that are asymptomatic and have a long duration. Increased screening would, however, result in an increase in notifications.⁹ Hence, using notification data to measure the underlying prevalence or incidence of STIs that are commonly asymptomatic is problematic.¹⁰

We undertook the study to investigate the substantial rises in notifications that had occurred in Victoria recently because it was uncertain whether these rises represented true rises in prevalence or changes in testing. Our aim was to determine if there has been any measureable change in the positivity of gonococcal and chlamydial infection among MSM attending a sexual health service in urban Melbourne from 2007 to 2013.

METHODS

A retrospective cohort analysis was conducted to investigate gonorrhea and chlamydia positivity among MSM attending the Melbourne Sexual Health Centre (MSHC) in Victoria, Australia, between 1 January 2007 and 31 December 2013. Melbourne Sexual Health Centre is the major public sexual health clinic located in the city of Melbourne. It provides a free walk-in service, and no referrals are required. It provides approximately 35,000 consultations per year, and MSM consultations accounted for ∼37% of the total consultations.¹¹

All patients who attended MSHC were required to complete a computer-assisted self-interview in English that collected their demographic details and history of sexual behaviors.¹² Clinical diagnoses and laboratory test results of each patient were recorded in the electronic medical system, the Clinic Practice Management System, at each clinic visit.

We included all MSM who attended MSHC between 2007 and 2013 who were tested for an STI on their visit. Men who have sex with men attending HIV clinics were excluded because no sexual risk data are routinely collected in the HIV clinic and could therefore not be used in the analysis. Men who have sex with men were defined as men who had any sexual contacts (including anal, oral) with other men in the last 12 months or in the 12 months before any previous visit to MSHC. We excluded consultations within a 3-month period because they were likely to be tests of cure rather than reflecting the community incidence of infection. Screening for gonorrhea and chlamydia among MSM at MSHC was conducted according to Australian guidelines.⁶ Screening included first pass urine or a urethral swab for chlamydia, and anal swabs for gonorrhea and chlamydia. Because asymptomatic urethral gonorrhea is less common in men,¹³ and Australian guidelines do not recommend screening, urethral swabs for gonorrhea were mostly limited to symptomatic patients. Pharyngeal swabs for gonorrhea were also collected. BD ProbeTec Strand Displacement Amplification Assay (Becton, Dickinson and Company, Sparks, MD) was used for the detection of urethral and anal chlamydia.^14,15 Anal, urethral, and pharyngeal swabs for gonorrhea were plated onto GC Agar medium for culture. All MSM were usually tested for gonorrhea and chlamydia at different sites, but patients had the right to refuse having any of those tests. All MSM diagnosed as having gonorrhea or chlamydia were treated without delay. A test of cure was performed 7 to 14 days after treatment of gonorrhea¹⁶ and within 3 months after treatment of chlamydia.¹⁷

Descriptive statistics and frequency distribution of the behavioral data were calculated. Chlamydia and gonorrhea positivity at a specific site was calculated as the number of men who were positive divided by the total number of men who were tested for the infection. Because asymptomatic screening did not include urethral swabs for gonorrhea, urethral gonorrhea positivity would be only among symptomatic MSM if the number of urethral swabs for gonorrhea was used as the denominator. To investigate trends in urethral gonorrhea in both symptomatic and asymptomatic MSM at MSHC, we calculated the proportion of urethral gonorrhea among MSM who had had a test for urethral chlamydia during the study period. Nonparametric Jonckheere-Terpstra test was used to assess the trends of median number of sexual partners in the last 12 months. A χ² trend test was used to assess the outcome variables increase or decrease significance over the study period. Univariate logistic regression was used to assess associations between gonorrhea (anal, pharyngeal, and urethral) and chlamydia (anal, and urethral) positivity with a range of risk factors (e.g., year of testing, age, number of partners in the last 12 months, and condom use in the last 12 months), as suggested in a previous study.⁵ All risk factors with P ≤ 0.10 in the univariate logistic regression analyses were selected for multivariate logistic regression analyses. Five separate multivariate logistic regression analyses with generalized estimating equations were constructed for each infection. Generalized estimating equation was used to adjust for within-individual correlations related to multiple visits of the same individual. Adjusted odds ratios (aORs) with 95% confidence intervals (CI) were also calculated. Significance level of 0.05 was used for all statistical tests. All statistical analyses were performed using the STATA software (version 12.0; Stata Corp).

Ethical approval was obtained from the Ethics Committee of Alfred Hospital, Melbourne, Australia (number 484/13).

RESULTS

There were 29,670 clinical consultations among 12,873 MSM included in the analysis from 2007 to 2013. The median age of the participants was 30.0 years (range, 14–85 years). There were 51,184 tests for gonorrhea (anal: 22,746; urethral: 3,205; and pharyngeal: 25,233) and 48,477 tests for chlamydia (anal: 22,728; and urethral: 25,749). The proportion of individual MSM who attended in a given year and were tested at least once for gonorrhea or chlamydia increased over the study period, from 75% to 90% for gonorrhea and 78% to 93% for chlamydia.

Gonorrhea in MSM

The positivity for pharyngeal gonorrhea increased significantly from 1.5% to 2.4% (P_trend = 0.001) for anal gonorrhea increased from 2.1% to 3.6% (P_trend < 0.001) and for urethral gonorrhea increased from 2.1% to 2.8% (P_trend < 0.001) between 2007 and 2013.

Rises in gonorrhea were primarily restricted to those reporting higher numbers of partners and did not occur in those reporting fewer partners in the last 12 months. Figure 1 shows that pharyngeal, anal, and urethral gonorrhea positivity did not change over time among MSM with 4 or less male partners (Fig. 1). However, MSM with 5 to 10 partners had a 2-fold increase in anal gonorrhea (1.6% in 2007 to 3.5% in 2013; P_trend = 0.004) and a 1.6-fold increase in urethral gonorrhea (2.5% in 2007 to 3.9% in 2013; P_trend < 0.001) but no observable change in pharyngeal gonorrhea (P_trend = 0.153). Similarly, MSM with more than 10 partners had a 2-fold increase in both anal (2.7% in 2007 to 5.7% in 2013; P_trend < 0.001) and pharyngeal gonorrhea (1.7% in 2007 to 3.3% in 2013; P_trend < 0.001) but no observable change in urethral gonorrhea (P_trend = 0.148).

Figure 1: Gonorrhea positivity among MSM, 2007 to 2013, by number of male partners in the last 12 months.

In the multivariate analysis, there was a significant annual increase in odds of men testing positive for anal (aOR, 1.12; 95% CI, 1.07–1.17), pharyngeal (aOR, 1.09; 95% CI, 1.03–1.15), and urethral gonorrhea (aOR, 1.11; 95% CI, 1.06–1.17), after adjusting for age, partner number, and condom use in the last 12 months (Table 1). There was a significant increase in odds of men testing positive for anal and pharyngeal gonorrhea who were younger than 25 years, had more than 10 sexual partners, and did not use condoms consistently in the last 12 months.

TABLE 1: Gonorrhea Positivity and Associations in MSM Attending MSHC, 2007 to 2013

Chlamydia in MSM

There was a 2-fold increase in anal chlamydia (from 2.9% to 6.1%; P_trend = 0.001) over the study period, but no significant changes in urethral chlamydia (from 3.1% in 2007 to 3.0% in 2013; P_trend = 0.770).

Rises in chlamydia were primarily restricted to those reporting higher numbers of partners and did not occur in those reporting fewer partners in the last 12 months. Figure 2 shows that anal and urethral chlamydia positivity remained stable among MSM with 4 or less or 5 to 10 male partners (Fig. 2). However, MSM with more than 10 partners had a 3-fold increase in anal chlamydia (from 2.8% in 2007 to 8.4% in 2013; P_trend = 0.002) and a 2-fold increase in urethral chlamydia (from 1.9% in 2007 to 3.5% in 2013; P_trend = 0.045).

Figure 2: Chlamydia positivity among MSM, 2007 to 2013, by number of male partners in the last 12 months.

In the multivariate analysis, there was a significant increase in odds per year of men testing positive for anal chlamydia (aOR, 1.06; 95% CI, 1.03–1.10) but no change in urethral chlamydia (aOR, 1.02; 95% CI, 0.98–1.06) over time after adjusting for age, number of partners, and condom use in the last 12 months (Table 2). Similar to the predictors in gonorrhea infection, there was strong evidence to support MSM who were younger than 25 years, had more than 10 sexual partners, and did not use condoms consistently in the last 12 months being at greater risk of anal or urethral chlamydia.

TABLE 2: Chlamydia Positivity and Associations in MSM Attending MSHC, 2007 to 2013

Sexual Behaviors in MSM

A small change in sexual risk among MSM attending the clinic was observed over the last 7 years (Table 3). The median number of sexual partners in the last 12 months over the period slightly dropped from 6 men in 2007 to 5 men in 2008 and remained stable till 2013, although the overall downward trend was statistically significant throughout the study period (P_trend < 0.001). The proportion of MSM who reported having a casual sexual partner in the last 12 months had a minimal increase from 81.2% to 85.1%, consistent condom use for receptive anal sex with casual partners had a narrow decrease from 64.6% to 58.9%, and consistent condom use for insertive anal sex with casual partners had fallen from 62.8% to 57.1%.

TABLE 3: Risk Behaviors Among MSM Attending MSHC, 2007 to 2013

DISCUSSION

In this study, we found substantial and significant increases in STIs unexplained by changes in individual-level sexual risk behavior in MSM. The question then arises as to why this happened and what factors are associated with this rise. In the last 7 years (2007–2013), the proportion of MSM with pharyngeal and anal gonorrhea, anal chlamydia rose after adjusting for other risk factors. These latest data from 2007 to 2013 contrast with stable trends of gonorrhea (anal and pharyngeal) and chlamydia (urethral and anal) in the previous 7 years (2002–2009) both from our center and more widely in the State of Victoria in Australia.⁵ Given that both studies used the proportion positive and adjusted for risk, and that the magnitude of the recent change is substantial, it is likely that the recent increases are real. What is surprising is that these increases have occurred in the context of substantial reductions in the number of sexual partners in the last 12 months, a minimal decrease in condom use with casual partners, and only minor changes in sexual risk in Melbourne over this time. It is possible that these substantial rises may have occurred because of changes in mixing patterns among sexual partners, with higher-risk partners now mixing together more commonly, possibly related to new methods of finding partners.^18,19 Our data showing that the change in positivity is primarily restricted to those with a high number of sexual partners support this hypothesis.

Similar changes in STI rates have been reported in Sydney with stable notifications of gonorrhea in men from 2000 to 2009 but a remarkable increase from 2010 to 2012, including a 4-fold increase in pharyngeal gonorrhea,²⁰ with most infections in MSM.²¹ Similarly, Sydney has also seen little change in condom use over the last 3 years, with 65.5% in 2010 and 66.0% in 2012 reporting no unprotected anal sex with MSM with casual sexual partners.²² It should be noted that the increase in gonorrhea notifications in Australia might also be due to the widespread use of dual nucleic acid amplification tests (NAATs) for the detection of chlamydia and gonorrhea in Australia since 2007. Gonococcal NAAT has a reduced specificity when testing pharyngeal and rectal specimens.²³ This cannot explain our findings because a major strength of our study is the use of bacterial culture for gonorrhea detection at MSHC throughout the period. Culture has a specificity of 100%,²⁴ and hence, the data are not undermined by any potential NAAT specificity issues.

Our study has several limitations that are important to consider in interpreting the results. First, this is a retrospective analysis of records from a single urban sexual health clinic, and the findings may not be representative of changes in prevalence and behaviors among MSM on a community level in Victoria or Australia. However, the magnitude of the change since 2009 is marked, and similar changes have been reported elsewhere (such as in Sydney).²⁰ Second, self-reporting and recall bias may have occurred in this retrospective study that requires recall of sexual activities in the last 12 months. Patients might be inclined to provide socially desirable responses, although in Australia, both computer answers and answers to a clinician are very similar.²⁵ Third, the concurrency of sexual partnerships is not examined in this study but may be an important factor associated with the sexual networks and risk of infection,²⁶ as approximately 37% of Australian MSM have had concurrent sexual partners in the last 12 months.²⁷ Fourth, the positivity of gonorrhea may be underestimated. Culture is less sensitive than NAAT for the detection of gonorrhea, particularly in extragenital sites,^23,28 and a true infection could be missed. However, we used the same detection method over time, so this should not have affected our analysis of changes in proportions with gonorrhea by site over time. Fifth, the positivity of each infection by anatomical site does not take into account multiple sites of infection because this is not common among MSM in Australia. Our data showed that only 0.6% and 0.9% of MSM were infected at multiple sites for chlamydia and gonorrhea, respectively, at the population level. However, among the positive cases, 7.8% and 19.1% of MSM were infected at 2 or more sites of chlamydia and gonorrhea. Sixth, the trends of chlamydia and gonorrhea may be different among HIV-positive and HIV-negative MSM. Although most HIV-positive MSM were excluded, a small proportion of MSM in this study might be HIV positive, but we were not able to identify them. Seventh, the number of MSM consultations at MSHC has doubled from 5946 in 2007 to 10,715 in 2013,²⁹ which may have influenced our results if there had been significant changes in the characteristics of the MSM seen. It is important to note, however, that the number of sexual partners has fallen significantly over the period, which would have biased against seeing the rise in the proportion of positive test results, which we observed. Finally, the proportion of MSM who attended in a given year and were tested for gonorrhea or chlamydia has increased over the study period, which may have influenced our results. However, in general when the screening coverage rises, it is usually associated with a fall in positivity so this potential bias in our study would have worked against our finding of a rising positivity.

Chlamydia and gonorrhea positivity has increased substantially in our study despite a small decrease in consistent condom use between MSM and their casual sex partners and an increase in STI testing in the state of Victoria (from 34.4% in 2007 to 38.2% in 2012).⁸ There are 3 possible explanations for this observation. First, consistent condom use may not necessarily refer to correct use. Incorrect condom use among heterosexual men has been reported not only in Australia,³⁰ but also internationally^31–33; however, errors of condom use are less well understood in MSM populations. A previous Australian study showed that approximately 6% of MSM have experienced partly or briefly inserting their penis into their partners’ anus without a condom then followed by immediate withdrawal.³⁴ In addition, a study conducted in the United States showed that some MSM have also experienced condom breakage (7%), slippage (6%), and incomplete use of a condom throughout anal intercourse (20%).³⁵ Other condom use errors such as not using lubricant with condoms (16%–26%), not leaving space at the tip (24%–46%), not squeezing the air from the tip before use (42%), and not checking for condom damage (75%) are commonly reported among all men, particular at younger ages.³³

The second possible reason for the rises in positivity could be changes in other types of sexual behavior and practices in MSM that were not measured in our study. For example, a substantial proportion of MSM in Australia have engaged in other anal activities such as rimming (19%–27%), nudging (20%–27%), and anal fingering (33%–39%),³⁴ which have been associated with gonorrhea and chlamydia infections, although we do not have evidence of changes in these practices over time.³⁶ In addition, 29% of MSM have engaged in group sex,³⁴ but there have been no studies investigating the association between group sex and condom use among MSM in Australia or studies investigating changes in group sex over time.

A third factor that may influence changes in positivity is screening. Generally, as screening increases in a population, it puts downward pressure on STI positivity because lower-risk individuals tend to be attracted into screening programs later than do higher-risk ones. However, if increasing screening identified higher-risk individuals, STI positivity may rise. Over the study period, the proportion of individual MSM attending MSHC who were screened at least once rose somewhat, and this may have influenced our results. We compared the changes in the positivity by site for gonorrhea and chlamydia, for new patients and returning patients, and they both showed similar trends over time (data not shown). Also, within the state of Victoria, the proportion of MSM tested rose from 2008 to 2010,^8,37 and then remained stable or declined slightly after 2010.⁸ It has been suggested that the rise in STI testing between 2008 and 2010 might be caused from the launch of a large social marketing campaign “Drama Downunder” in the state of Victoria in February 2008.³⁷ There are unfortunately no studies in MSM that have documented whether STI screening actually affects STI rates, but rising rates of STI screening in the early stage of our observation period could potentially influenced our results.

Overall, both gonorrhea and chlamydia positivities are increasing among MSM in Australia in recent years, particularly among those who had more than 10 sexual partners in the last 12 months. Other changes may have explained the increase in STI rates in MSM, but a better understanding of the patterns of finding sexual partners and their sexual practices is an important part of exploring the reasons for the large rises in STI.