Article Text

Download PDFPDF

Refocusing health promotion for syphilis prevention: results of a case-control study of men who have sex with men on England’s south coast
  1. J Imrie1,2,
  2. N Lambert1,
  3. C H Mercer2,
  4. A J Copas2,
  5. A Phillips1,
  6. G Dean1,
  7. R Watson3,
  8. M Fisher1
  1. 1HIV/GUM Research Department, Residence Block, Brighton General Hospital, Elm Grove, Brighton BN2 3EW, UK
  2. 2Centre for Sexual Health and HIV Research, Royal Free and University College Medical School, University College London, Mortimer Market Centre, off Capper Street, London WC1E 6AU, UK
  3. 3Terrence Higgins Trust South, 61 Ship Street, Brighton BN1 1AE, UK
  1. Correspondence to:
 John Imrie
 PhD, Centre for Sexual Health and HIV Research, Mortimer Market Centre, off Capper Street, London WC1E 6AU, UK; jimrie{at}gum.ucl.ac.uk

Abstract

Background: Prevention interventions have had little impact on syphilis outbreaks among men who have sex with men (MSM) and diagnosis rates continue to rise rapidly. Detailed studies of the determinants of syphilis transmission are needed to inform new prevention interventions.

Objective: To investigate factors associated with recent syphilis diagnosis and recommend strategies for improved prevention interventions.

Methods: A case-control study of MSM attending genitourinary medicine (GUM) and HIV outpatient clinics. Cases were MSM testing positive for early syphilis, based on either laboratory or microscopy results, with those testing negative being controls. All participants completed the same anonymous questionnaire covering demographics, lifestyle, sexual behaviour, and sexual partnerships.

Results: 50 cases and 108 controls returned questionnaires. Syphilis diagnosis was significantly associated with sexually transmitted infection history, recent recreational drug use, receptive anal sex practices, but not insertive ones, higher numbers of oral sex partners, but not specific oral sex practices. Overall, there were no differences between HIV positive and HIV negative/untested men in reporting of sexual behaviours or partnership combinations. The use of public sex settings (backrooms, saunas, “cruising grounds,” etc) was reported by 68% of all participants and not significantly associated with syphilis diagnosis.

Conclusions: Many key behavioural and partnership risk factors for syphilis are also risk behaviours for HIV transmission and point to the need for integrated strategies that tackle the two infections together. Simplified procedures for syphilis screening should be available in GUM clinical settings, along with targeted community outreach prevention interventions.

  • GUM, genitourinary medicine
  • MSM, men who have sex with men
  • STI, sexually transmitted infections
  • UAI, unprotected anal intercourse
  • syphilis
  • prevention
  • case-control study
  • homosexual men
View Full Text

Statistics from Altmetric.com

Syphilis rates among British men who have sex with men (MSM) have increased dramatically.1 The re-emergence of syphilis in this population is a particular public health concern because of high levels of HIV co-infection that may contribute to increased HIV transmission.2–4 Brighton, on England’s south coast (population 250 000) has an ongoing syphilis outbreak concentrated in MSM.5 Like other outbreak sites, a high proportion of syphilis cases are HIV co-infected (40%) and most report predominantly casual or anonymous sexual contacts (86%).6–9 To better understand the determinants of syphilis transmission and inform new prevention measures, we undertook a case-control study of factors associated with recent syphilis diagnosis.

METHODS

Participants were recruited from the local genitourinary medicine (GUM) and HIV outpatient clinics between October 2002 and February 2004. Eligible participants were MSM, having a syphilis test, aged 18 or over, reporting sexual contact with a male partner (<90 days), and able to give informed consent. Our intention was to enrol cases and controls in a 1:2 ratio from the same source clinic but, overall, slightly more controls were recruited (50 cases and 108 controls). All participants completed the same paper questionnaire. For cases, this was within 60 days of an early syphilis diagnosis based on laboratory testing and/or dark ground microscopy, while for controls, it followed their initial clinical assessment. The questionnaire covered demographics, lifestyle, general and sexual health, social and sexual mixing, sexual behaviours, sexual dysfunction, and partnerships. Completed questionnaires were double entered into SPSS. Univariate comparison of variables between cases and controls was undertaken using χ2, Mann-Whitney and t tests. Crude odds ratios (OR) are given with corresponding 95% confidence intervals (95% CI).

RESULTS

Of 223 MSM invited to participate, 63% (50/79) of those with early syphilis returned a questionnaire (cases), as did 75% (108/144) of those testing negative (controls). Among controls, responders and non-responders were similar with respect to age. For cases, responders and non-responders were similar with respect to both age and number of sexual partners in the last 90 days, based on self report and enhanced surveillance data. Response rates were similar in the GUM and HIV clinics, as was the distribution of disease stage (for cases). Overall, 36% (18/50) of cases were HIV co-infected, which compares to 40% in Brighton generally, 46% in London, and 26% in Manchester.1,5,9

Comparison of sociodemographic, social, and lifestyle characteristics showed cases had generally lower educational attainment, and were significantly more likely to report a previous STI and recent recreational drugs use (table 1). The majority reported use of local gay venues and higher proportions than in other studies reported use of “public sex settings” (backrooms, saunas, “cruising grounds,” “cottages”) during the last year (74% (37/49) of cases and 65% (68/105) of controls; p = 0.250).1,6–9

Table 1

 Sociodemographic, social, and lifestyle characteristics of study participants

Overall, more than 90% reported ⩾2 sexual contacts in the last 6 months (98% (49/50) of cases and 90% (97/108) of controls; p = 0.071), and nearly three fifths reported both regular and casual/anonymous partners (59% (26/44) of cases and 60% (60/97) of controls; p = 0.937). There were no significant differences between HIV positive and HIV negative/untested men in reported sexual behaviours or partnership combinations—for example, any UAI in the last 6 months (HIV+ 69% (53/77) versus HIV−/untested 63% (49/78); p = 0.430), or UAI with a casual/anonymous partner in the last 6 months (HIV+ 50% (37/74) versus HIV−/untested 50% (49/78); p = 1.00). There was some evidence of an association between syphilis and higher numbers of casual/anonymous partners (p = 0.057) (table 2). Men diagnosed with syphilis were also somewhat more likely to report more regular partners, either serially or concurrently, than controls (OR 2.21, 95% CI 0.98 to 5.00; p = 0.058). This borderline significant effect remained even after controlling for numbers of casual/anonymous partners (data not shown).

Table 2

 Associations between recent syphilis diagnosis and reported partnerships and sexual behaviours in the last 6 months

Reporting “any oral sex” was very high (94%) and condom use low (8% for insertive and 5% for receptive oral sex). There was no significant association between syphilis and oral sex practices or between syphilis, oral sex, and partnership type (data not shown). However, like previous studies, we found a significant association between syphilis and higher numbers of oral sex partners (p = 0.034) (table 2).

Syphilis was significantly associated with receptive, but not insertive anal sex, overall and for both regular partnerships (OR 6.20, 95% CI 1.74 to 22.0; p = 0.005) and casual/anonymous partnerships (OR 3.25, 95% CI 1.31 to 8.07; p = 0.019). It was also significantly associated with unprotected receptive anal sex, but not unprotected insertive, and “inconsistent condom use with all partners” in relation to receptive (table 2), though not insertive anal sex (OR 1.21, 95% CI 0.49 to 2.94; p = 0.554). Overall, two thirds of all participants reported at least one episode of unprotected anal intercourse (UAI) in the last 6 months (76% (38/50) of cases and 61% (64/105) of controls; p = 0.065) and of these, approximately 36% were HIV positive.

DISCUSSION

Extending our understanding of the determinants of syphilis transmission is an essential part of reinvigorating our syphilis prevention efforts.11 The risk factors identified in this study, lower educational attainment, previous STI history, recent recreational drug use, receptive anal sex and number of oral sex partners largely confirm what is already known.1–3,6–9,12 However, this study for the first time enables us to look in greater detail at behaviour and partnership combinations, and how these contribute to individual risk. In particular, the high proportions reporting a previous STI, and reporting UAI (66% in last 6 months), the similar distribution of risks between HIV positive and HIV negative/untested men, and inconsistent condom use for receptive anal sex with all partners conferring the greatest risk, are all factors that have important implications for syphilis and HIV transmission, and require prevention strategies that tackle both infections simultaneously.2,4,11 The high proportion overall reporting a previous STI suggests that an important opportunity to offer intensive health promotion, aimed specifically at reducing future risk, may be being missed.11

Our failure to find an association between syphilis and use of public sex settings runs counter to other research.1,6–9,11–13 However, we would argue it reflects rapid evolution of MSM’s social and sexual environments, proliferation of new venues for men seeking partners, and increasing use of such venues, along with locally specific factors; not that these venues are becoming less risky.14

The association between syphilis diagnosis and higher numbers of oral sex partners adds depth to what we already know about syphilis acquisition and oral sex.1,12 We believe the association and the low levels of reported condom use for oral sex may partially explain the rapid transmission of syphilis, particularly in some higher risk sexual networks or situations where men have adopted oral sex practices to reduce HIV risk.12 If this is correct, it underscores the need for syphilis prevention strategies to embrace the often difficult to deliver, partner reduction messages.15 It stands to reason that where population risk behaviours are increasing, and the increase includes more partnerships, targeting settings where risk occurs for the most intensive interventions is sensible. These should specifically include outreach which can deliver partner reduction messages sensitively and one to one.11,14,15

Finally, we recruited two controls from the same clinic as each case to ensure our study was inclusive and locally representative of MSM at risk of syphilis and attending services. Although this approach precluded investigating HIV as an independent risk factor, we nevertheless found little variation between HIV positive and negative/untested men in their self reported risk behaviours or partnership combinations.1,3,9 The high proportion reporting key risk behaviours indicate that many MSM GUM attenders are also at elevated risk of syphilis.1,13 Syphilis screening is part of routine clinical HIV care in the United Kingdom,16 but there are more limited screening opportunities for other MSM. Simplified, innovative approaches that make syphilis screening more accessible, such as “online” services and clinical use of oral tests, warrant further investigation.13

Key messages

  • To date, prevention initiatives have had little impact on syphilis outbreaks affecting mainly MSM

  • Improved understanding of factors associated with syphilis transmission is essential in order to refocus prevention

  • Syphilis diagnosis appears to be strongly associated with recreational drug use, receptive anal sex, and higher numbers of oral sex partners

  • Key risk factors for syphilis are also risks for HIV and point to the need for integrated strategies that tackle the two infections together

  • Outreach work that can deliver partner reduction messages sensitively and innovative approaches that simplify screening and access warrant further investigation

Acknowledgments

The authors wish to acknowledge and thank all the study participants, the Department of Health for providing funding, and the staff of the Claude Nicol Centre and the Lawson Unit, Royal Sussex Country Hospital, Brighton for their patience, commitment, and support of this study.

CONTRIBUTORS
 JI, NL, and AC were primarily responsible for the study design and development of the questionnaire; NL and AP recruited the study participants and co-ordinated data collection and management; AC, CM, NL, and JI designed and carried out the main analysis; RW coordinated the community components and local commercial gay venues’ participation in the Brighton Syphilis Outbreak Project; MF and GD were overall principal investigators on the Brighton Syphilis Outbreak Project and contributed to all stages of the research and paper writing; JI and NL co-authored the original paper and all the authors reviewed successive drafts in advance of submission to the journal.

REFERENCES

View Abstract

Footnotes

  • Funding: UK Department of Health.

  • Conflict of interest: none.

  • Ethical approval: The study was approved by the Brighton and Mid-Sussex Local Research Ethics Committee.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.