Abstract
BACKGROUND: Loneliness and social isolation are highly prevalent public health concerns. However, their impacts in acute care settings such as the postoperative period are unclear. This study explored whether loneliness, social isolation, or both are associated with a primary composite outcome of 30-day major postoperative complication, 30-day postoperative emergency admission, and 90-day mortality. METHODS: This retrospective cohort study used the UK Biobank. Individuals with complete loneliness and social isolation data, an eligible surgery within 1 yr of baseline, and national health registry linkage were included. Loneliness and social isolation were defined with validated, self-reported measures. Participants were stratified into four groups by baseline loneliness and social isolation status. Logistic regression models assessed associations between loneliness-social isolation groups and postoperative outcomes with comprehensive covariate adjustment. RESULTS: Some 27 905 UK Biobank participants met eligibility criteria. Individuals who were socially isolated but not lonely had increased odds of the primary outcome (adjusted odds ratio 1.36, 95% confidence interval 1.10-1.68, P=0.004), mainly driven by higher odds of 30-day major postoperative complication (odds ratio 1.35, 95% confidence interval 1.04-1.72, P=0.018). In a subgroup analysis, this association remained statistically significant in males but not females. CONCLUSIONS: Individuals, particularly males, who were socially isolated but not lonely had increased odds of postoperative complications. Our results are consistent with other reports that loneliness and social isolation are weakly correlated, and that social isolation is more consistently associated with adverse physical outcomes. These findings suggest that targeted, interventional programmes addressing social isolation in surgical patients might assist in mitigating postoperative risks.