Article Text

P3.128 Development of a risk calculator for the 3-month prediction of incident syphilis infection among high-risk men who have sex with men and transgender women presenting to a std clinic in lima, peru
  1. Lao-Tzu Allan-Blitz1,
  2. Kelika A Konda2,
  3. Silver K Vargas3,
  4. Eddy R Segura3,
  5. Gino M Calvo3,
  6. Carlos F Caceres3,
  7. Jeffrey D Klausner2
  1. 1David Geffen School of Medicine, UCLA, And Saphir, Los Angeles, USA
  2. 2Department of Medicine, David Geffen School of Medicine, University of California Los Angeles Los Angeles, USA
  3. 3Universidad Peruana Cayetano Heredia, Lima – Peru


Introduction Syphilis incidence worldwide has rebounded since 2000, especially among men who have sex with men (MSM). A predictive model for incident syphilis infection may inform prevention counselling and use of chemoprophylaxis.

Methods We analysed data from a longitudinal study of a STD clinic-based cohort of MSM and transgender women reporting a history of HIV or syphilis infection and/or high-risk sexual behaviour, followed quarterly for two years. Incident infection was defined as a four-fold increase in RPR titers or new RPR reactivity if two prior titers were non-reactive. We used generalised estimating equations with a Poisson regression to develop a predictive model of syphilis incidence in one-half of the data set, and verified the model in the second half, calculating an area under the curve (AUC), summarising specificity and sensitivity. We then applied the final model to the full baseline dataset. Finally we developed an online risk calculator from our model.

Results Among 401 participants enrolled, 22% were transgender women and 31% were HIV-infected at baseline. Syphilis incidence was 19.9 cases per 100-person years (95% CI 16.3–24.3). Predictors of syphilis incidence were HIV infection, high number of male sex partners (categorised as: 0, 1, 2–3, 4–9,>10), history of syphilis infection, receptive and versatile anal sex role and condomless receptive anal sex. The AUC was 71% (95% CI 64%–78%) in the validation dataset for incident syphilis infection in the next 3 months. Those at highest risk had a 1-in-7 likelihood of syphilis infection in the next 3 months. When applied to the baseline dataset the AUC was 71% (95% CI 65%–77%) for predicting recent syphilis infection. The online syphilis risk calculator is available at: (English), (Spanish).

Conclusion Our results show that the likelihood of syphilis infection among a high-risk STD clinic-based cohort can be estimated accurately. Our calculator may guide STD clinical management by directing risk behaviour counselling and potential use of doxycycline chemoprophylaxis.

Statistics from

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.