Introduction In the United States reported cases of chlamydia have increased since reporting began, due in part to increased screening. However, the implication of these trends for the population prevalence remains unclear. We aimed to understand and reconcile the epidemiological trends, and examine counterfactuals.

Methods We developed a deterministic heterosexual pair formation model to simulate chlamydia epidemiology in the US heterosexual population aged 15-54y. The pair formation model accounts explicitly for sexual partnership dynamics, such as re-infection within the partnership, and the model is stratified by age, risk and relationship type (long-term v. casual). We used a Bayesian approach to calibrate model parameters (including time-varying screening, reporting and test sensitivity) to age- and sex-specific national case report rates from 2000–2015 (ages 15-54y), lab-measured population prevalence estimates from NHANES 1999–2014 (15-39y), and sexual behaviour data from the Youth Risk Behaviour Survey (15-18y).

Results Model estimates were able to reproduce both chlamydia prevalence and reported case rates. Results indicate an increase in chlamydia screening in women.

Conclusion This analysis is the first to fit a chlamydia transmission model to national sex- and age- specific prevalence and case report time trends. The results suggest screening would have to achieve a higher coverage, or we should investigate novel strategies to reduce chlamydia prevalence further. This model could be used to investigate the impact of novel prevention interventions, such as improved partner notification strategies and targeted screening programs.