Given the persistent mental health disparities between sexual minorities and the general population, social epidemiological research should address this disparity by investigating the differential impact of neighbourhood social environments across sexual orientations. There is growing evidence that neighbourhood cohesion, conceptualized as a sense of belonging and social connection, is an important social determinant of mental health in the general population, but little is known about its impact across sexual orientations. Using data from the UK household longitudinal study (2009-2018) including waves 1, 3, 6, and 9 (n= 52,903), this paper examined the longitudinal relationship between neighbourhood cohesion and mental health (using GHQ-12) across sexual orientations. A fixed-effect regression approach was taken to model the within-person change over time in GHQ predicted by neighbourhood cohesion with disaggregated analyses by gender and sexual orientation. Across all sexual orientations and genders, individuals who experienced an increase in neighbourhood cohesion also saw an improvement in their mental health over time. Moreover, the effect of neighbourhood cohesion on mental health over time differed by sexual orientation. Each 1-point increase in neighbourhood cohesion (on a 5-point scale) lead to mental health improvements of-0.8 GHQ score (95%CI-0.89 to-0.71) for heterosexual males at the lowest end, and up to-1.71 GHQ score (95%CI-2.31 to-1.11) for homosexual men at the highest end.