Introduction Transmission of Chlamydia trachomatis (chlamydia) is influenced by both sexual behaviour and psychological determinants (i.e., risk perception). However, mathematical models describing chlamydia transmission often consider only sexual behaviour. We explored the influence of incorporating risk perception in a mathematical model that estimated the impact of different testing scenarios on chlamydia prevalence.

Methods We developed a pair compartmental model with a susceptible-infected-susceptible structure representing heterosexuals aged 16–26 years. Model parameters were informed by an online pilot study (n=173) on sexual behaviour, psychological determinants, and chlamydia (re-) infections and will be updated with data from an ongoing longitudinal cohort study. The model population was divided in two sexual risk behaviour groups (based on the number of partners in the last year) and further divided in two risk perception groups (based on perceived risk for chlamydia). We compared the impact of an overall testing uptake of 20% per year on population chlamydia prevalence with different testing scenarios: 1) differential uptake among sexual behaviour groups (higher uptake in high sexual behaviour group) and additionally 2) differential uptake among risk perception groups (higher uptake in high risk perception groups).

Results Respondents with high sexual risk behaviour (SB) and low risk perception (RP) had the highest mean number of partners (high SB/low RP=5.3, compared to high SB/high RP=5.0, low SB/high RP=1.3 and low SB/low RP=1.1, p<0.001), shortest mean duration of partnerships (high SB/low RP=153 days, compared to high SB/high RP=233, low SB/high RP=512 and low SB/low RP=607 days, p<0.001) and the highest percentage of self-reported chlamydia infections in the past year (high SB/low RP=22.2%, compared to high SB/high RP=14.3%, low SB/high RP=3.1% and low SB/low RP=0%, p=0.02). Models that did not consider differential testing uptake among risk perception groups overestimated the impact of testing on chlamydia prevalence. This effect was largest in the high sexual behaviour/low risk perception group.

Conclusion Mathematical models incorporating risk perception could improve the estimation of the impact of testing interventions on the prevalence of chlamydia in specific subgroups.