Background Estimating the per sex act transmission probability of Chlamydia trachomatis is necessary to assess an individual's risk of infection and re-infection after sexual contacts with chlamydia-positive partners. Katz (1992, Stat Med) made one of the first attempts to estimate the probability of chlamydial transmission based on data about the proportions of concordant (both partners infected or uninfected) and discordant (only one partner infected) couples tested for chlamydia. This provides a cross-sectional per partnership transmission probability but did not address the duration of sexual partnerships and the spontaneous clearance of chlamydia. Mathematical models can be used to take these complexities into account.

Methods We developed a simple mathematical model of chlamydia transmission that incorporates the formation and dissolution of sexual partnerships. Chlamydia can be transmitted from an infected to an uninfected partner throughout the duration of the partnership. Infected partners can also clear chlamydia spontaneously. The model was fitted to published data from a cross-sectional study of chlamydia infection status in heterosexual couples at a US sexually transmitted diseases clinic. The model was further parameterised with study data about overall chlamydia positivity and the number of sexual partners in the last 6 months.

Results The data showed similar proportions of infected female partners of infected men and infected male partners of infected women, suggesting similar transmission probabilities. Assuming an average infectious duration of 12 months for women and men, we obtained best-fit estimates for the transmission rate (0.0064 per day) and average duration of partnerships (6.0 months) in the study. The transmission rate per day corresponds to a per sex act transmission probability of 4.5%, assuming one unprotected episode of coitus per week in a partnership. Our analysis also shows that these estimates depend on the assumed infectious duration of chlamydia.

Conclusions Our approach shows how the dynamics of sexual partnership formation and disease transmission can be taken into account to derive the daily rate of chlamydia transmission within a partnership. The per sex act transmission probability can then be inferred. These refined estimates will help to parameterise future mathematical models of disease transmission and assess the risk of infection and re-infection after sexual contacts with chlamydia-positive partners.