AimsWe aimed to investigate the association of household crowding in childhood with trajectories of depressive symptoms in middle-aged and older adults. MethodsWe studied 47,010 participants (56 % women, 63 years at baseline) from SHARE. Using multinomial logistic regression, we estimated odds ratio (OR) with 95 % confidence interval (CI) for the association of household crowding in childhood (number of household members/number of rooms at the age of 10) with trajectories of depressive symptoms (EURO-D scale), which were generated with growth mixture modeling. We adjusted for resources in childhood, sociodemographic and health-related characteristics in mid-life and older age and tested effect modification by sex. ResultsWe identified four trajectories of depressive symptoms: constantly low (n = 33,969), decreasing (n = 5595), increasing (n = 5574) and constantly high (n = 1872). When compared to the those with constantly low depressive symptoms and adjusting for all covariates, household crowding in childhood was associated with greater odds of constantly high (OR 1.12; 95 % CI 1.08–1.17), decreasing (OR 1.11; 95 % CI 1.07–1.15) and increasing (OR 1.09; 95 % CI 1.06–1.13) depressive symptoms. The associations were stronger in women than in men. ConclusionsPrevention of household crowding in childhood may ameliorate the development of constant as well as transient depressive symptoms during ageing. The effect can be stronger in women than in men.