Background Partner notification (PN) is an essential part of the case management of sexually transmitted infections STI), including Chlamydia trachomatis. Failure to notify current partners might cause re-infection of the index case, whilst failure to notify previous partners could result in ongoing transmission in the population. The impact of PN at both the individual and population level is, however, unclear.

Methods We developed an individual-based modelling framework called Rstisim, which can simulate transmission of any STI through a dynamic sexual network and track the history of an individual's partnerships. The effect of different PN strategies for C trachomatis was investigated in three models with increasing levels of complexity of the sexual partnership dynamics: a) an instantaneous contact model which is based on the widely used assumption that sexual contacts happen instantaneously; b) a pair model where sexual partnerships last for a certain period; c) a triple model in which individuals can have up to two concurrent partnerships. We used data from the National Survey of Sexual Attitudes and Lifestyles (Natsal) 2000 for 16–25-year-old women and men to parameterise the sexual behaviour of young adults.

Results The models all have a baseline chlamydia prevalence of 3%. In the triple model, chains of contacts can be seen at cross-section, whereas there are, by definition, no ongoing partnerships in the instantaneous contact model. In all three models, we find that a substantial proportion of partners (>10%) from partnerships that ended as far back as 18 months is infected with C trachomatis. We then investigated the population level effect of PN (with 50% success) as a complementary strategy to screening (at a rate of 0.1 per year). Increasing both the number of notified partners and the PN period results in lower levels of C trachomatis. Under the most realistic assumptions of the sexual partnership dynamic, most of the effect of PN results from notifying the current partner.

Conclusions We found that extended PN periods can efficiently identify new chlamydia-infected cases. At low screening levels, the additional benefit of PN in decreasing chlamydia prevalence is minor and primarily derives from notifying the current partners in order to prevent re-infection. This study exemplifies the differences between individual and population level outcomes of PN as an intervention for the management of C trachomatis infections.