Abstract Despite the increasing evidence of association between chlamydia infection and depression, currently there is a paucity of research with limited scope to understand better the temporal nature of the relationship between them. We consider this problem in adolescence and young adulthood by pooled analysis of 7250 participants from the Avon Longitudinal Study of Parents and Children and the third National Survey of Sexual Attitudes and Lifestyles. We propose a latent variable modelling approach which can handle harmonization of categorical variables including ordinal measures from the two studies as well as measurement error and time trends.