Background Achieving adequate levels of participation and capturing high risk groups are key stones for the effectiveness of large-scale Chlamydia screening programs. We examined the determinants of (repeated) participation and Chlamydia positivity to evaluate to what extent high risk groups were reached in a large scale Internet-based screening program in the Netherlands.

Methods The Chlamydia Screening Implementation was initiated in three regions among people aged 16–29 years. Data from the first two screening rounds (2008–2010), in which approximately 280 000 persons were invited annually, were analysed. Socio-demographic and behavioural correlates of screening participation and positivity were studied in multilevel logistic regression models. Cluster was added as a second level of analysis, taking into account the effect of the neighbourhood-based invitations (to cover social or sexual networks).

Results The same socio-demographic factors associated with lower screening uptake were also associated with higher Ct-positivity such as young age, non-Dutch origin, lower education, high community risk, low SES, in round 1 as well as 2. At the same time, behavioural risk factors such as having casual partners, ≥2 partners in <6 months, concurrent partners, and a history of STI, were associated with higher participation. A small cluster effect for screening uptake was observed, independent from community risk and individual risk factors. The model for repeated participation showed that men, Turkish/Moroccans and persons ≤20 years were less likely to participate twice, while people having a short-term relationship, a non-Dutch partner or concurrent partners were more likely to participate again. Ct-positives, who did not participate in the rescreening after 6 months, were also less likely to participate in the second screening round.

Conclusions Socio-demographic factors associated with lower participation were also associated with higher Ct-positivity, showing that, high-risk demographic groups were more difficult to mobilise than low-risk groups. However, independent of this, higher behavioural risk levels were associated with higher participation rates (especially in the model for repeated participation), suggesting self-selection for screening based on the persons’ risk (perception) in both low and high community risk groups. Our study shows the complexity of the process—including individual as well as community factors and their interaction—as to whether or not be screened for chlamydia.