Abstract
BACKGROUND: Social isolation and loneliness have emerged as important modifiable risk factors for mental disorders, posing significant public health challenges. However, they have not been comprehensively investigated in relation to multisystemic diseases and their temporal trajectories. We aimed to systematically identify the health outcomes associated with social isolation and loneliness and characterize their disease trajectories and comorbidity networks. METHODS: A total of 466,547 participants (mean age: 56.5 ± 8.1, 54.6% females) with available information on social isolation and loneliness were included from the UK Biobank between 2006 and 2010 and followed up until 2022. Social isolation was measured using a composite score derived from three questions on number in household, frequency of friend and family visits, as well as participating in leisure and social activities. Loneliness was assessed by the subjective perception of feeling lonely and the willingness to confide in others. A total of 246 medical conditions were included in phenome-wide association analyses. Disease trajectory and comorbidity network analyses were performed to identify sequential patterns and visualize disease clusters. RESULTS: During an average medium follow-up of 11.7 years, social isolation was found to be significantly associated with increased risks of 28 medical conditions, and loneliness with 80. The strongest association for both exposures was with personality disorders, with respective hazard ratios of 2.12 (95% CI, 1.59-2.82) for social isolation and 2.62 (95% CI, 1.90-3.61) for loneliness. The conditions covered a broad spectrum, including respiratory, neurological, digestive, musculoskeletal, and genitourinary diseases and mental disorders. Three main disease clusters were identified in relation to social isolation and eight disease clusters were related to loneliness. Septicemia and alcohol-related disorders as the initial condition was notably observed in the disease cluster trajectories of both social isolation and loneliness. Subsequently, the comorbidity network revealed three and five distinct comorbidity modules associated with social isolation and loneliness, respectively. The disease nodes within these modules exhibited structural consistency with those within the disease trajectory clusters. CONCLUSIONS: These findings highlight the importance of integrating screening, interventions, and referrals for social isolation and loneliness into health care system to prevent the adverse health conditions.