Abstract Major Depressive Disorder (MDD) is a significant global health concern, yet the influence of social interactions among individual neighbours, in addition to neighbourhood characteristics, remains underexplored. This study applies linear-in-means models and spatial econometric techniques to examine whether and which social interactions influence MDD among adults. The data used relate to two repeated cross-sections of 19,700 individuals aged 23 to 65 years collected via the MINI neuropsychiatric interview in two waves of the Lifelines Cohort Study in Northern Netherlands, supplemented with neighbourhood-level information from Statistics Netherlands. Three types of social interactions — peer, contextual, and correlated — are assessed, with Bayesian comparison tests identifying contextual and correlated effects as the most influential. Subsequently, a two-step regression model is employed to address the high prevalence of zero outcomes. The results highlight the influence of contextual and correlated effects among individual neighbours, in addition to individual and neighbourhood characteristics. These findings emphasise the need for further research into social interactions on MDD and support for public health interventions that take them into account.
Tort-Carrera et al. (Tue,) studied this question.