August 15, 2020
The group density effect, where a group member's psychiatric risk is associated with the proportion of the local population their group comprises, demonstrates the importance of minority group status to mental health. Previous research, focusing on ethnicity, has been correlational, but newly-formed identities provide opportunities for natural experiments, with greater scope for causal inference. This study examines whether such a group density effect can be found for the novel Brexit identities of ‘leaver’ and ‘remainer’ following the UK's divisive 2016 referendum on EU membership. Mixed effects models were fitted to the Understanding Society panel survey series (N = 25,555, 19,767 for analyses controlling for pre-referendum mental health data), predicting mental health as a function of individual opinion on EU membership and local referendum results. These interacted such that those holding the local majority opinion had better mental health (Odds ratio (OR):875 [0.766- 0.9995]), compared to those in the minority. This result survived adjustment for individual and area-level economic circumstances (OR:866 [0.758-0.989]), and, strikingly, pre-referendum mental health (OR: 0.841 [0.709-0.998]), as well as a number of other potential confounding variables. The results provide evidence for rapidly forming group density effects based on de novo identities, and suggest that identity may be a causal mechanism for group density effects more broadly. They also speak to the extent of polarisation in the Brexit-era UK, and its public health consequences.
Social Science and Medicine