Abstract
Background Over the last decade, syphilis diagnoses among men-who-have-sex-with-men (MSM) have strongly increased in Europe. Understanding the drivers of the ongoing epidemic may aid to curb transmissions.
Methods and Findings We set up an epidemiological model to assess the drivers of syphilis in MSM in Switzerland between 2006 and 2017. We stratified the model by syphilis stage, HIV-diagnosis, and behavioral factors to account for syphilis infectiousness and risk for transmission. In the main model, we used ‘reported non-steady partners’ (nsP) as the main proxy for sexual risk. We parameterized the model using data from the Swiss HIV Cohort Study, Swiss Voluntary Counselling and Testing center, cross-sectional surveys among the Swiss MSM population, and published syphilis notifications from the Federal Office of Public Health.
The main model reproduced the increase in syphilis diagnoses from 168 cases in 2006 to 418 cases in 2017. It estimated that between 2006 and 2017, MSM with HIV diagnosis had 45.9 times the median syphilis incidence of MSM without HIV diagnosis. Defining risk as condomless anal intercourse with nsP decreased model accuracy (sum of squared weighted residuals, 378.8 vs 148.3). Counterfactual scenarios suggested that increasing screening of MSM without HIV diagnosis and with nsP from once every two-years to twice per year may reduce syphilis incidence (at most 12.8% reduction by 2017). Whereas, increasing screening among MSM with HIV diagnosis and with nsP from once per year to twice per year may substantially reduce syphilis incidence over time (at least 63.5% reduction by 2017).
Conclusions The model suggests that reporting nsP regardless of condom use is suitable for risk stratification when modelling syphilis transmission. More frequent screening of MSM with HIV diagnosis, particularly those with nsP may aid to curb syphilis transmission.
Introduction
Syphilis, a sexually transmitted infection caused by bacteria Treponema pallidum, remains a major public health problem. Over the last decade, a rising number of diagnoses especially in men-who-have-sex-with-men (MSM) was observed in Western Europe and Northern America[1–3]. In Switzerland, the number of syphilis diagnoses in MSM tripled between 2006 and 2017 [4]. A similar trend was observed for the incidence rate of syphilis in HIV-diagnosed MSM in the Swiss HIV Cohort Study (SHCS)[5–7]. Having oral sex, condomless anal sex, multiple partners, and sex with non-steady partners have been identified as major risk factors of syphilis transmission[7–9]. Recent studies show a decrease in condom use for anal sex in MSM with HIV diagnosis[10]. This situation may have resulted in increased incidence of syphilis.
The natural course of syphilis has four stages: primary, secondary, latent, and tertiary. Almost all syphilis transmission occurs during the primary and secondary stages and the disease remains transmissible for an average period of two to three months[11]. During the latent stage of syphilis, there are no clinical signs or symptoms of syphilis. If left untreated, the infection may reach its tertiary stage and may affect different organ systems. Hence it is essential to diagnose and treat people infected with syphilis as early as possible.
We aimed to identify possible drivers of the ongoing syphilis epidemic among MSM in Switzerland using a mathematical model (Figure 1) and to identify behavioral predictors (proxy for transmission risk of syphilis) that could best explain the changes in syphilis incidence over time. Based on this model, we evaluated the impact of different screening strategies on syphilis incidence.
Methods
We used data from the SHCS and cross-sectional surveys among the Swiss MSM population to estimate the transmission risk of syphilis. These temporal trends were used to parameterize a mathematical model, which was then used to assess the impact of different screening strategies via counterfactual scenarios (see Figure 1 for an overview of the workflow). The model’s open-source code is available on https://github.com/Kouyos-Group/Suraj_syphilis_repository.
Data
Our model population comprised all MSM in Switzerland from 2006 to 2018. We estimated the number of MSM in Switzerland between 2006 and 2018 based on the demographic dynamics of Switzerland by the Federal Statistical Office, Switzerland and Schmidt and Altpeter, 2019[12,13]. In Switzerland, all syphilis cases are anonymously reported to the Federal Office of Public Health (FOPH)[4].
Approximately 80% of all HIV-diagnosed MSM in Switzerland are enrolled in the Swiss HIV Cohort Study (SHCS)[13,14]. The SHCS is a prospective multicenter cohort study, which systematically collects epidemiological and clinical data for more than 20000 HIV-diagnosed persons of age 16 or older in Switzerland since 1988[15]. After the restart of routine syphilis testing in the SHCS in 2004, MSM are being tested for syphilis annually with treponemal and non-treponemal tests. We defined the first syphilis episode as a positive treponemal test after a negative treponemal test result. A ≥4-fold titer reduction or negativity in non-treponemal test result after an episode of syphilis indicate the absence of active syphilis. We defined subsequent episode(s) of syphilis as a subsequent ≥4-fold titer increase with a titer value of at least eight in non-treponemal test.
Patterns of transmission risk
We estimated the proportion of MSM with HIV diagnosis (MSMwHD) and with non-steady partners (nsP) based on information in the SHCS data. We estimated the yearly rates of switching from MSMwHD with nsP to MSMwHD without nsP and vice-versa using time-to-event survival analysis (Supplementary section 6). In contrast, we estimated analogous proportion for MSM without HIV diagnosis (MSMw/oHD) from published survey data (Gaysurvey)[16]. Gaysurvey was a series of anonymous internet-based self-administered surveys periodically conducted in Switzerland among MSM since 1987. It was a part of the monitoring system established by the FOPH as a tool for monitoring risk behavior among MSM in Switzerland. In the last Gaysurvey, in 2014, 834 people participated in the Gaysurvey and had a median age of 40 years. About 97% of these participants identified themselves either as gay/homosexuals (84.1%) or bisexuals (12.8%). As the data is not longitudinal at the individual level and is available for only a couple of years, a linear interpolation and a constant extrapolation was used to estimate the proportion of MSM with non-steady partners for MSMw/oHD. As the participants of Gaysurvey were not followed up over time and hence participants could be different at each time point of survey conduction, we assumed a constant rate of switching back to MSMw/oHD with nsP from MSMw/oHD without nsP. We fixed this rate to the mean rate of switching back to MSMwHD with nsP from MSMwHD without nsP for the years 2006 and 2017. Based on this assumption, we estimated that the yearly rates of switching from MSMw/oHD without nsP to MSMw/oHDwith nsP using a spline function (Supplementary section 6).
Model of syphilis transmission
To model the syphilis epidemic in MSM in Switzerland between 2006 and 2017, we set up a system of 36 coupled ordinary differential equations. We stratified the mathematical model by syphilis stage and cascade of care to account for infectiousness and disease progression in four levels: susceptible (MSM who do not have syphilis and can become infected with syphilis), infected (MSM who are infected with syphilis and can transmit; infectious), latent (MSM who are infected with syphilis but cannot transmit; non-infectious) and diagnosed (MSM diagnosed with syphilis) (Figure S1). We did not include a ‘treated’ compartment as diagnosed individuals are assumed to be immediately treated and to clear from the infection within one month[11]. We further stratified the model by reported non-steady partners to account for increased transmission risk of syphilis. Additional layers in the model include HIV status (MSMw/oHD vs. MSMwHD), the episode of syphilis: first (MSM who never had syphilis before) or subsequent (MSM who had at least one episode of syphilis before). Susceptible MSM become infected with syphilis with a force of infection (Λ) due to sexual contact with MSM infected with syphilis (Supplementary section 3.3). Infected MSM are infectious during the primary and secondary stages of syphilis that together last about 2-3 months after transmission. After this period, infected MSM enter the latent stage of syphilis during which the infection will no longer contribute to Λ. We further assume that infected and latently infected MSM would be diagnosed with syphilis if they are tested for syphilis and that all diagnosed MSM are immediately treated. Treated men would clear syphilis and become susceptible for subsequent episodes of syphilis. A complete description of the model, including model equations is available in the supplementary sections 1 and 2.
Changes in transmission risk such as MSM with non-steady partners (nsP), as well as syphilis diagnosis and treatment rates among the MSMwHD were derived from the SHCS data. Changes in sexual risk behaviour among MSMw/oHD were derived from the Gaysurvey. Diagnosis rate and treatment rate of syphilis among the MSMw/oHD were derived from the literature, and Voluntary Counselling and Testing (VCT) data (Table S2).
Parameters governing transmission of syphilis and the syphilis incidence rate among the MSMwHD not enrolled in the SHCS were assumed to be same as the MSMwHD enrolled in the SHCS. Parameters which could not be obtained from databases and literature such as transmission rate of syphilis, transmission rate of HIV, risk-sorting behaviour, and initial number of MSM infected with syphilis in 2006 were derived by fitting the model to data on syphilis cases among the MSMwHD and MSMw/oHD in Switzerland. This fit was done by minimizing the sum of squared weighted residuals between each observed and predicted datapoint assuming normally distributed errors (Supplementary section 7). We further used simplified Markov chain Monte Carlo (MCMC), based on an adaptive Metropolis algorithm and including a delayed rejection procedure, with prior values specified by the distribution of the parameters to obtain their posterior distributions using the package ‘A Flexible Modelling Environment for Inverse Modelling, Sensitivity, Identifiability and Monte Carlo Analysis (FME)’ in R version 4.0.0[17,18]. We used independent and uniform priors within ad hoc bounds for the free parameters (Table S1). In a sensitivity analysis, we also fitted the model through maximization of the likelihood by assuming Poisson distributed incident cases of syphilis and compared the model fits.
Counterfactual scenarios
We compared the incidence of syphilis obtained with the main model fit with that simulated when assuming the following two alternative scenarios. Firstly, we considered the impact of change in transmission risk (reported nsP in MSM) between the years 2006 and 2018 by assuming after 2006 no change in the proportion of MSM switching from with nsP to without nsP and vice-versa. Secondly, we considered the impact of change in screening frequency for syphilis between the years 2006 and 2018. We estimated the influence of this frequency by assuming an increase (by a given factor) of the screening frequency among MSMwHD and MSMw/oHD with non-steady partners.
Sensitivity analysis
To test the robustness of the model and possible inconsistencies in data, we recalibrated the model in the following five ways: Firstly, we used ‘reported condomless anal intercourse with non-steady partners’ (nsCAI) as the proxy for sexual risk instead of nsP. Secondly, we assessed the impact of assuming the same transmission rate for MSMwHD and MSMw/oHD but taking into account a potential underreporting of syphilis cases among MSM without HIV diagnosis. Thirdly, we investigated the impact of a possible overestimation of transmission risk among MSM without HIV diagnosis. Fourthly, we assessed the impact of a potential infectiousness of syphilis in the latent stage on the dependency between screening frequency and syphilis incidence. Finally, we tested the effect of model fitting through maximization of the likelihood by assuming Poisson distributed incident cases of syphilis.
Results
Data
Our mathematical model simulates transmission dynamics of syphilis in the MSM population in Switzerland from 2006 to 2018. Based on the Federal Statistical Office, Switzerland and Schmidt and Altpeter, 2019, we estimated the number of MSM in Switzerland in 2006 and 2018 to be 75016 and 85325, respectively (Figure S2)[12,13]. There were 113 and 456 cases of syphilis reported to the FOPH in 2006 and 2018, respectively[4].
To estimate the number of MSM with HIV diagnosis (MSMwHD) in Switzerland from 2006 and 2018 and the syphilis cases among them, we analyzed 5382 MSM enrolled in the Swiss HIV cohort study (SHCS) with a syphilis test between 2006 and 2018. The median registration year was 2007 [IQR (1999, 2013)] and the median year of birth was 1967 [IQR (1960, 1975)]. The key demographic characteristics of MSMwHD in the SHCS is summarized in Table 1. There were 2665 and 4366 MSMwHD enrolled in the SHCS in 2006 and 2018, respectively. By assuming that 80% of MSMwHD in Switzerland are enrolled in the SHCS, we estimated the number of MSMwHD in Switzerland in 2006 and 2018 to be 3031 and 5257[13,14]. We also estimated syphilis cases reported among MSMwHD in Switzerland in 2006 and 2018 to be 106 and 312, respectively.
Patterns of transmission risk
Information about reported non-steady partners and condomless anal intercourse was available for 5027 out of the analyzed 5382 MSMwHD (93.4%) enrolled in the SHCS during the observation period. 3768 out of 5027 MSM (75.0%) reported having a non-steady partner (nsP) at least once and 1942 (51.5%) among them reported condomless anal intercourse at least once. We estimated the proportion of MSMwHD with nsP to be 0.48 and 0.52 for the years 2006 and 2018, respectively (Figure S3). We estimated that among MSMwHD the yearly rates of switching from without nsP to with nsP has decreased only slightly from 0.49 in 2006 to 0.41 in 2017. For the same period, we estimated that the yearly rates of switching back from with nsP to without nsP has decreased from 0.74 in 2006 to 0.45 in 2017 (Figure S4). We estimated the proportion of MSMw/oHD with nsP to be 0.74 and 0.73 for the years 2006 and 2018 respectively using data published in Gaysurvey (Figure S5)[16]. Our results were similar to the proportion of MSMw/oHD with nsP (0.75) published in European Men-who-have-sex-with-men Internet Survey (EMIS) 2017[19]. As the Gaysurvey data is not longitudinal, we assumed a constant rate of switching for MSMw/oHD from without nsP to with nsP. It equaled the mean rate of switching for MSMwHD from without nsP to with nsP for the years 2006 and 2017 and estimated it to be 0.54. Based on the Gaysurvey data, we then estimated that the yearly rate of switching for MSMw/oHD from without nsP to with nsP has increased by 2.7% (1.44 in 2006 vs 1.48 in 2017) (Figure S6). Detailed information on the data used in our model is provided in supplementary material (Table S1-S2)
Observed and modelled data
The model could accurately reproduce the number of MSMwHD in Switzerland and the syphilis cases among MSMwHD and MSMw/oHD (Figure 2). The model estimated that the number of new syphilis infections among MSM for the years 2006 to 2017 ranged between 168 and 418 diagnosed cases per year, which corresponds to an overall incidence rate of 2.24 and 4.91 syphilis cases per 1000 person-years (median syphilis incidence rate of 3.77 cases per 1000 person-years).
In addition, the model provided an estimate for the incidence rate of syphilis among MSMwHDand MSMw/oHDstratified by reported non-steady partners (Figure 3). MSMwHD had 45.93 [IQR (45.28, 46.32)] times the incidence of MSMw/oHD and MSM with non-steady partners had 1.63 [IQR (1.60, 1.67)] times the incidence of MSM without non-steady partners. Characterization of model optimization, model performance, and sampling for uncertainty estimates are depicted in the supplementary material (Figure S7-S11, Table S3)
Counterfactual scenarios
We compared the incidence rate of syphilis obtained with the model fitted to the observed number of syphilis cases with simulations assuming the following alternative scenarios: Firstly, we considered the impact of change in transmission risk between the years 2006 and 2018. With no change in the proportion of MSM switching from MSM with nsP to MSM without nsP and vice-versa between the years 2006 and 2018, we estimated a reduction of 17.56% (4.04 vs 4.91 cases per 1000 person-years) in incidence rate of syphilis in 2017 compared to incidence rate obtained by the model fit in 2017 (Figure S12). Secondly, we considered the impact of change in screening frequency for syphilis between the years 2006 and 2018. Increasing the frequency of screening for syphilis among MSM with HIV diagnosis and with nsP from once per year to twice per year during the observation period, we estimated a reduction of 99.19% (0.04 vs 4.91 cases per 1000 person-years) in incidence rate of syphilis in 2017 compared to incidence rate obtained by the model fit in 2017. By contrast, we estimated only a reduction of 12.75% (4.28 vs 4.91 cases per 1000 person-years) in incidence rate of syphilis in 2017 by increasing the frequency of screening for syphilis among MSM without HIV diagnosis and with nsP from once per two years to twice per year during the observation period (Figure 4).
Sensitivity analysis
To test the robustness of the model and possible inconsistencies in data, we recalibrated the model and determined the corresponding change in the incidence rate of syphilis, the fitted parameters, and the goodness of fit (sum of squared weighted residuals) (Table S4).
Firstly, we considered alternative criteria to stratify transmission risk (Figure S13). When assuming reported non-steady condomless anal intercourse (nsCAI) as the transmission risk instead of nsP, we estimated that MSM with nsCAI had 1.71 [IQR (1.65, 1.81)] times the incidence of MSM without nsCAI. However, the goodness of fit for this model was considerably worse than for the main model (378.8 vs 148.3). Secondly, we assessed the impact of the assumption of different transmission rates of syphilis among MSMwHD and MSMw/oHD. In the main model we had to assume different transmission rates, because of an apparent discrepancy in the data: despite similar frequency of reported nsP among MSMwHD and MSMw/oHD, we observed much higher incidence of syphilis among MSMwHD than MSMw/oHD even when taking serosorting into account (Table S2). To resolve this contradiction, in the main model, we assumed different transmission rates of syphilis (beta) in MSMwHD and MSMw/oHD in our model (Table S1). In the absence of this assumption, the model was unable to reproduce the syphilis epidemic in Switzerland and led to either large overestimation of syphilis incidence in MSMw/oHD or large underestimation in MSMwHD (Figure S14). As an alternative approach to capture the discrepancy in the data, we fitted a model assuming that there is no difference in transmission rate of syphilis among MSMwHD and MSMw/oHD but instead including a parameter to account for possible underreporting of syphilis cases to FOPH among MSMw/oHD (Figure S15). However, the goodness of fit for this model was worse than for the main model (sum of squared weighted residuals, 206.6 vs 148.3) and required the assumption of an unrealistically high underreporting factor (40.8). Thirdly, we investigated the impact of a possible overestimation of transmission risk among MSMw/oHD. We fitted models assuming the proportion of MSMw/oHD with nsP to be 25%, 50%, and 75% lower than the proportion of MSMw/oHD with nsP estimated from data published in the Gaysurvey (Figure S16). This corresponds to proportion of MSMw/oHD with nsP in 2017 to be 0.54, 0.36, and 0.18 vs 0.73, respectively. The estimated goodness of fits for these models were slightly better than the goodness of fit for the main model (136.4, 124.4, and 123.6 vs 148.3 respectively) suggesting that model is robust to the hypothesized overestimation.
Fourthly, we assessed the impact of the duration of syphilis infectiousness on the dependency between the screening frequency and the syphilis incidence. Instead of assuming zero infectiousness in the latent stage of syphilis (>3 months since infection), we varied the infectiousness of syphilis in the latent stage to be 1% and 10% of that in the primary and/or secondary stage of syphilis (≤ 3 months since infection) (Figure S17). The goodness of fits for these models were worse than for the main model (154.3 and 397.6 vs 148.3 respectively). As depicted above in the counterfactual scenario, increase in the frequency of screening for syphilis among MSMwHD with nsP led to reduced simulated syphilis incidence in 2017. Increased infectiousness during the latent stage led to a smaller estimated reduction in syphilis incidence in 2017 in the counterfactual scenarios (93.59% and 63.50% vs 99.19% by assuming the infectiousness in latent stage of syphilis to be 1% and 10% vs 0% of that in primary and secondary stage of syphilis respectively) (Figure 5).
Finally, we tested the effect of using an alternative fitting procedure. Instead of fitting the model through minimization of the sum of squared weighted residuals, we fitted the model through maximization of the likelihood by assuming Poisson distributed incident cases of syphilis. We obtained similar median incidence rate and model fit as obtained by the main approach (Figure S18).
Discussion
The transmission model presented here could accurately reproduce the syphilis epidemic in Switzerland. Our study outlines the much higher incidence of syphilis in MSM with HIV diagnosis compared to MSM without HIV. In our simulations, MSM with HIV diagnosis have 46 times the syphilis incidence of MSM without HIV diagnosis. On average, MSM with non-steady partners (nsP) have 1.6 times the syphilis incidence in MSM without nsP. Also, constant yearly rates of switching from MSM without nsP to MSM with nsP (our measure of transmission risk) and vice-versa during the observation period would have led to lower incidence of syphilis in 2017 by 17.6%. Finally, increasing the screening frequency of syphilis to twice per year in MSM with HIV diagnosis and with nsP would have contributed to a much higher reduction in syphilis incidence than increasing the screening frequency to twice per year in MSM without HIV diagnosis and with nsP (99.2% vs 12.8%).
A key challenge in simultaneously modelling syphilis transmission in MSMwHD and MSMw/oHD population is the apparent discrepancy in the data: When comparing the data from the FOPH and the SHCS, MSMwHD exhibited 46 times the syphilis incidence of MSMw/oHD, even though proportions of MSMw/oHD with nsP obtained from the data published in Gaysurvey was higher than the proportion of MSMwHD with nsP estimated from the SHCS. To resolve this contradiction, we had to assume different transmission rates of syphilis in MSMwHD and MSMw/oHD in our model. Based on fitting the model, we estimated >20-fold in transmission rate of syphilis in MSMwHD compared to MSMw/oHD, even after adjusting for contact probabilities and difference in testing frequency among MSMwHD and MSMw/oHD. Because the majority of the MSMwHD in Switzerland are virally suppressed, it is unclear whether a biological explanation for a higher transmission rate of syphilis in MSMwHD is plausible. Alternatively, this difference may also be due to unadjusted transmission risk of syphilis such as number of partners and frequency of contacts, which we did not model explicitly.
To explore alternative distributions of risk, we stratified the structure of transmission risk in our model based on reported condomless anal intercourse with non-steady partners (versus just reported non-steady partners). The model fit was worse than that in the main analysis. This might be because syphilis could be transmitted through oral sex, and using condoms for anal intercourses is not completely protective against syphilis[20]. In an alternative approach, we assumed no difference in transmission rate of syphilis among MSMwHD and MSMw/oHD and allowed the model to account for possible underreporting of syphilis cases to FOPH among MSMw/oHD to explain the apparent difference in the incidence rate among MSMwHD and MSMw/oHD. We estimated the incidence of syphilis in MSMw/oHD to be 40.8 times that reported by FOPH. This hypothesized underreporting of syphilis cases, although not in the same magnitude, seems to be supported by the European Men-who-have-sex-with-men Internet Survey (EMIS) 2017[19]. Based on the EMIS-2017 survey, 3.4% of all European MSMw/oHD and 13.9% of all European MSMwHD (overall 4.4%) were diagnosed syphilis in the previous twelve months. If the EMIS-2017 was representative of MSM in Switzerland, the number of syphilis cases would have been about 6.3 times those reported to the FOPH. However, about 70% of the participants of EMIS-2017 were recruited through geosocial dating apps such as Planet-Romeo, Grindr, and Hornet. Several studies have shown that MSM who use geosocial dating apps have higher number of partners, frequency of unprotected sex, and incidence of sexually transmitted diseases compared to MSM who do not use such apps[21–24]. Hence, incidence of syphilis among participants of EMIS 2017 is likely higher than that in the overall MSM population in Switzerland.
Persons infected with syphilis are usually symptomatic during the primary and secondary stage of syphilis that lasts about 2-3 months since infection. After this period, individuals enter latent stage of syphilis during which they become asymptomatic, yet still detectable through syphilis screening. If left untreated, individuals can develop late/tertiary syphilis disease any time from 1 to 30 years after primary infection. About 25-40% of individuals with untreated syphilis can develop such diseases[25]. Hence, MSMw/oHD may be infected with syphilis but are not diagnosed due to lower frequency of syphilis screening and eventually may have non-syphilis related deaths, resulting in an underreporting of syphilis cases to FOPH. Moreover, despite syphilis being a notifiable disease, it is possible that not all cases are reported to the FOPH resulting in underreporting. However, the extent of underreporting is not clear and warrants further investigation.
Also, there is no clear evidence on the infectiousness of syphilis in the latent stage of syphilis (>3 months since infection) [11]. We tested the influence of infectiousness in the latent stage of syphilis. Our model could reproduce the syphilis epidemic for these scenarios and estimated similar incidence rates compared to the main model fit. By means of counterfactual scenarios, we further assessed the impact of infectiousness of syphilis in latent stage on the dependency between the screening frequency and the syphilis incidence. We found that the estimated reduction in syphilis incidence due to increase in screening frequency decreases when assuming residual infectiousness during the latent stage. Infectiousness during the latent stage of syphilis warrants further investigation as it may have an impact on the effectiveness of different screening strategies. However, even when we assumed infectiousness during the latent stage to be as high as 10% of that in primary and secondary stage, we observed 63.5% reduction in syphilis incidence in 2017 when screening of syphilis was performed twice per year instead of once per year in MSMwHD with nsP. Our study results thus robustly indicate that frequent screening of MSM with HIV diagnosis and with nsP is required to curb syphilis transmission. This supports the EACS guidelines which recommend to “consider more frequent than annual screening if at risk”[26].
As mentioned before, we observed that the proportion of MSMw/oHD with non-steady partners is higher than the proportion of MSMwHD with non-steady partners. On one hand, MSMwHD in the SHCS answer the questionnaires on sexual behaviour during consultation or interviews with physicians or nurses they are likely to have recurrent contact with, while enrolled in the cohort. This may cause social desirability bias, thereby resulting in potential underestimation of transmission risk among MSMwHD. On the other hand, MSM participating in Gaysurvey may not be representative of MSM in Switzerland as MSM who engage in practices such as MSM with nsP may more likely enroll in online surveys such as Gaysurvey compared to MSM without nsP, thereby resulting in an overestimation in transmission risk among MSMw/oHD. We tested the influence of overestimation in transmission risk among MSMw/oHD estimated from data published from Gaysurvey. Our model could reproduce the syphilis epidemic for these scenarios and estimated similar incidence rates and similar goodness of fit (sum of squared weighted residuals) compared to the main model fit indicating the robustness of our model. Finally, we fitted the model using maximization of likelihood by assuming Poisson distributed incident cases of syphilis and found similar results to that obtained by main model fit.
Our transmission model enabled us to integrate data from different sources such as published data from the FOPH and the literature, and the results estimated from the SHCS and cross-sectional studies to assess the drivers of the syphilis epidemic in Switzerland. Previous studies have shown the potential impact of more frequent screening in MSM for syphilis on reducing syphilis infections[27–29]. An agent-based, network model of syphilis transmission, representing a core population of 2000 Canadian MSM at particular high risk for syphilis, explicitly estimated the effect of different screening frequencies on the projected annual rates of syphilis in this population. Our model differs from this study, in that we not only focused on a core group of MSM by means of transmission risk of syphilis but also stratified the model based on HIV status. To the best of our knowledge, this is the first study to link the syphilis epidemic in MSMwHD with that in MSMw/oHD and to estimate the effect of screening strategies implemented in target groups of the population on the overall epidemic.
One of the main limitations of our study is the uncertainty regarding the representativeness of the data used for MSMw/oHD. This resulted in model assumptions which required critical sensitivity analyses to assess the robustness of our findings. Our study has however a substantial strength with regard to HIVwHD: the use of routine serological screening in a representative population over many years. Estimation of transmission risk of syphilis in MSMwHD were based on the SHCS data. As the SHCS is highly representative for MSMwHD in care in Switzerland, we expect these data to provide accurate estimates. In contrast, estimation of transmission risk in MSMw/oHD was based on cross-sectional studies which might have led to more uncertainty in results. The degree of mixing between MSMwHD and MSMw/oHD were based on the estimated frequency of HIV discordant status in the data collected from VCT checkpoints across Switzerland where people can get spontaneously tested for sexually transmitted infections in a fully anonymized way. As bias towards testing upon transmission risk is expected, representativeness cannot be warranted. We stratified based on reported non-steady partners and condomless anal intercourse with non-steady partners as the transmission risk of syphilis. Although the available data in the SHCS did not allow us to account for the effect of other risk factors such as number of partners and frequency of sexual acts, the goodness of fit suggests reported non-steady partner to be a good proxy for risk stratification to model syphilis transmission.
In conclusion, our model reconstructed the syphilis epidemic among MSM in Switzerland by capturing changes in sexual risk behaviour and interactions between HIV MSM with and without HIV diagnosis. It suggests that non-steady partnership (nsP) regardless of condomless anal intercourse would be a suitable criterion for risk stratification when modelling syphilis transmission. More frequent screening and treating MSM with HIV diagnosis, particularly those with nsP may aid to curb syphilis transmission.
Data Availability
The individual level datasets generated or analyzed during the current study do not fulfill the requirements for open data access: 1) The SHCS informed consent states that sharing data outside the SHCS network is only permitted for specific studies on HIV infection and its complications, and to researchers who have signed an agreement detailing the use of the data and biological samples; and 2) the data is too dense and comprehensive to preserve patient privacy in persons living with HIV. According to the Swiss law, data cannot be shared if data subjects have not agreed or data is too sensitive to share. Investigators with a request for selected data should send a proposal to the respective SHCS address (www.shcs.ch/contact). The provision of data will be considered by the Scientific Board of the SHCS and the study team and is subject to Swiss legal and ethical regulations and is outlined in a material and data transfer agreement. However, the numerical data underlying the study are presented in the main manuscript and supplementary information.
Supporting information
S1: Supplementary material
Acknowledgements
We thank the patients for participating in the SHCS, the study nurses, physicians, data managers, and the administrative assistants.
The members of the SHCS are Anagnostopoulos A, Battegay M, Bernasconi E, Böni J, Braun DL, Bucher HC, Calmy A, Cavassini M, Ciuffi A, Dollenmaier G, Egger M, Elzi L, Fehr J, Fellay J, Furrer H, Fux CA, Günthard H (President of the SHCS), Haerry D (deputy of “Positive Council”), Hasse B, Hirsch HH, Hoffmann M, Hösli I, Huber M, Kahlert CR (Chairman of the Mother and Child Substudy), Kaiser L, Keiser O, Klimkait T, Kouyos RD, Kovari H, Ledergerber B, Martinetti G, Martinez de Tejada B, Marzolini C, Metzner KJ, Müller N, Nicca D, Paioni P, Pantaleo G, Perreau M, Rauch A (Chairman of the Scientific Board), Rudin C, Scherrer AU (Head of Data Centre), Schmid P, Speck R, Stöckle M (Chairman of the Clinical and Laboratory Committee), Tarr P, Trkola A, Vernazza P, Wandeler G, Weber R, and Yerly S.
Footnotes
Author Contributions: SB, LSV, RDK, and AR participated in study conception and design, data acquisition and interpretation, and critical revision of the manuscript. SB drafted the first manuscript. SB and LSV performed the statistical analyses. Remaining authors critically reviewed the study protocol and the manuscript. All authors contributed to the design of the study and approved the final version of the manuscript.
Abbreviations
- EMIS
- European Men-Who-Have-Sex-With-Men Internet Survey
- FOPH
- Federal Office of Public Health
- HIV
- Human immunodeficiency virus
- IQR
- Interquartile range
- MSM
- Men-who-have-sex-with-men
- MSMwHD
- MSM with HIV diagnosis
- MSMw/oHD
- MSM without HIV diagnosis
- nsCAI
- Condomless anal intercourse with non-steady partners
- nsP
- non-steady partnership
- SHCS
- Swiss HIV Cohort Study
- VCT
- Voluntary Testing and Counselling