Shared predictors of social participation in psychosis and the general population: The central relevance of emotional and relational wellbeing

Introduction: Social participation (SP) lowers mortality rate and risk for depression, dementia (Wanchai & Phrompayak, 2019) and functional disability (Ashida et al., 2016) as well as being a factor influencing recovery from severe psychiatric illnesses such as psychosis (Harrison et al., 2001). This is due to complex combinations of biological, psychological and behavioural mechanisms. Demographic factors such as socioeconomic status have been shown to influence social participation (Kung et al., 2022) as well as more specifically focused studies implicate life factors such as urbanism, community connection, life satisfaction and anxiety. However, the relative strength of association of these factors on social participation remains unclear. Using a large-scale prospective population cohort, I aim to identify significant correlates of social participation and, furthermore, to examine whether these associations differ in psychosis. Methods: Social participation was derived from the frequency of leisure and friend/family activities (range 0-10) in the UK Biobank. The relative importance of a broad range of 514 demographic, psychosocial, health and lifestyle variables on social participation was established using XGBoost on a data sample, with predictive regression models developed to quantify the associations completed on an independent sample. The moderating effect of psychosis on these relationships was then analysed. Results: Regression analysis of the dataset (n=492,875, age mean years ± SD 56.5±8.1) revealed that, when combined with age and sex, satisfaction metrics including friendship satisfaction (R²=0.067), perceived life meaning (R²=0.051), ability to confide (R²=0.047), happiness (R²=0.046), and family relationship satisfaction (R²=0.046) were among the strongest correlates of social participation, along with employment status (R²=0.043). An eight-parameter model was generated to predict 11.5% of the variance in social participation (3.8% explained by sex/age). Factor analysis identified a single latent emotional and relational wellbeing factor that accounted for 8.35% of the variance in social participation (3.7% explained by sex/age). These relationships did not differ significantly between individuals with and without psychosis. Conclusions: Perceived emotional and relational wellbeing, along with employment status and access to transport are statistical predictors of the extent of social participation, both in a healthy population and for those with psychosis. These findings highlight factors that are related to better social integration for individuals in the general population and those with psychiatric disorders. These links can therefore help inform policy to provide more effective interventions, thereby reducing associated health and economic burdens.

Funder

Taighde Eireann

Publisher

University of Galway

Rights

CC BY-NC-ND

cbn

Collections

University of Galway Theses (Research Masters Theses)