TY - JOUR AB - Background Neighbourhood has risen as a relevant determinant of health. While there is substantial evidence that environmental factors affect health, far less evidence of the role of social mechanisms in the causal chain between neighbourhood characteristics and health is available. Method To evaluate the role of social cohesion as a mediator between four different neighbourhood characteristics and health using data from German Socio-Economic-Panel (SOEP), a longitudinal mediation analysis was performed. Multilevel linear regression models adjusted for socio-economic variables involved three time points and two measures of physical and mental health (physical and mental component scores (PCS and MCS) of the SF12 Questionnaire. Participants were followed-up for 4 and 10 year starting in 2004. Results A total of 15,518 measures of MCS and PCS on 10,013 participants living in 4985 households were included. After adjusting for values of MCS and PCS at baseline and demographic/socio-economic variables, social cohesion was a significant positive predictor of both MCS and PCS (β-coefficient MCS: 1.57 (0.27); PCS: 1.50 (0.24)). Interaction between social cohesion and follow-up were significant for PCS. The effect of environmental and built characteristics on health was consistently mediated by social cohesion with proportion varying between 10 and 23%. Discussion We show that social cohesion is part of the causal chain between environmental and built characteristics of a neighbourhood and health, with increasing mediation effect over time for physical health. Social mechanisms should be considered when studying the effect of neighbourhood characteristics on health inequalities making social cohesion as a legitimate target of public health interventions at neighbourhood level. DA - 2020 DO - 10.1186/s12889-020-09149-8 KW - Public Health KW - Environmental and Occupational Health LA - eng PY - 2020 T2 - BMC Public Health TI - Does social cohesion mediate neighbourhood effects on mental and physical health? Longitudinal analysis using German Socio-Economic Panel data UR - https://nbn-resolving.org/urn:nbn:de:0070-pub-29445374 Y2 - 2024-11-22T09:22:43 ER -