Background Neisseria gonorrhoeae can be cultured from saliva in men with pharyngeal gonorrhoea and could theoretically be transmitted from the pharynx to the urethra when saliva is used as a lubricant for masturbation. To explore this issue, we developed mathematical models for the transmission of Neisseria gonorrhoeae at each of oropharynx, urethra and anorectum among men who have sex with men (MSM).

Methods Model 1 included transmission routes (oral sex, anal sex, rimming, kissing, and three sequential sex practices) we have previously validated. In Model 2, we added masturbation to model 1. In Model 3, we included masturbation but excluded the three sequential sex practices. We calibrated our data to six international studies. We evaluated the model performance using the Root Mean Squared Error (RMSE) and Cohen’s d statistic.

Results Model 2 has significantly higher RMSE than model 1 (p-value <0.01 in five datasets, and p=0.47 in one dataset), but only p-values from two datasets revealed a substantially large effect (Cohen’s d > 0.8) compared with Model 1. This suggests performance of Model 1 and Model 2 are similar. In contrast, Model 3 has significantly higher RMSE than both Model 1 and Model 2 (p-value <0.01 for all six datasets), and p-value revealed a large effect (Cohen’s d > 0.8 for all six datasets) compared with the two models. This suggests performance of Model 3 is significantly worse than Model 1 and Model 2.

Conclusion Our findings indicate that masturbation plays a moderate role in the transmission of Neisseria gonorrhoeae. Our model also suggests that sequential sexual practices may be more important than masturbation for explaining the site-specific prevalence in men with multi-site infection. Our model predicted that about 1 in 4 cases of urethral gonorrhoea might arise from masturbation if it transmits gonorrhoea.