The sample consisted of 1625 patients with a CIDI-confirmed diagnosis of MDD and 1698 screened controls from the Netherlands. Four MDD outcomes were studied as dependent variables: one main MDD prevalence-outcome (all MDD), two more severe MDD prevalence-outcomes (suicidal and chronic MDD), and one MDD course outcome (chronic versus non-chronic MDD). Because SNP rs25531 modifies the effect of 5-HTTLPR, haplotypes of 5-HTTLPR and rs25531 were measured. For the four MDD outcome measures, we examined the direct effects of 5-HTTLPR/rs25531-haplotypes, five environmental factors (lifetime and recent stressful life-events, sexual abuse, low educational attainment, and childhood trauma) and their interaction in logistic regression models.
The environmental factors had large and consistent effects on all four MDD outcomes, including course of MDD. The 5-HTTLPR/rs25531-haplotype had a suggestive effect on course of MDD, but not on presence of MDD. Gene-by-environment interaction was significant (<0.05) for one of the 20 tests performed, which is not more than expected by chance.
Environmental factors were not assessed before the onset of MDD.
Environmental factors had a strong impact on the presence and course of MDD, but no evidence for gene-by-environment interaction was found.