Effects of loneliness on healthcare utilization and medical expenditures in US cancer survivors.

368 Background: Loneliness is associated with increased mortality risk in cancer survivors. However, effects of loneliness on healthcare utilization are understudied. This study examined associations of loneliness with having emergency room (ER) visit, usual source of care, and hospitalization in US adults with and without cancer history. Methods: We identified 1,844 cancer survivors and 9,517 adults without a cancer history from the 2021 Medical Expenditure Panel Survey. Loneliness was measured by 3 questions on feeling lack of companionship, left out, and isolated from others. Scores for each question were summed to create a total loneliness score, which was categorized into low/no loneliness (3-5) and high loneliness (6-9) based on previous literature. Separate logistic regression models were used to assess association of loneliness and whether having ER visit, usual source of care, and hospitalization. All models adjusted for age, region, sex, race, non-cancer health conditions, activity limitation, marital status, education, health insurance, and cancer diagnosis. We used a cancer history by loneliness interaction term to examine the differential associations of loneliness and healthcare utilization by cancer history. Results: 2,367 (23%) adults without a cancer history and 450 (22%) cancer survivors reported loneliness. All adults averagely aged over 50, with more female (non-cancer: 57.1 vs 49.9%, cancer: 68.0 vs 55.9%) and those with non-cancer health conditions ≥3 (non-cancer: 20.6 vs 15.4%, cancer: 54.0 vs 44.4%) in the loneliness group. Cancer history modified the association between loneliness and ER visit (interaction term p-value 0.04), but didn’t modify the association with having usual source of care (p=0.38) or hospitalization (p=0.87). Loneliness was associated with having ER visit (OR 1.49, 95% CI 1.11-2.00) in cancer survivors while not in non-cancer patients (OR 1.08, 95% CI 0.92-1.27). No significant association was observed between loneliness and having usual source of care or hospitalization. Conclusions: Loneliness is associated with ER visit in US cancer survivors. Findings warrant further study of impact of loneliness on clinical outcomes in cancer survivors, with the aim of enhancing strategies to mitigate loneliness and optimize healthcare utilization. Association of loneliness and healthcare utilization. No Cancer History (9,517, 85%) Cancer History (1,844, 15%) Loneliness No (7,150, 77%) Yes (2,367, 23%) No (1,394, 78%) Yes (450, 22%) Odds ratio Having ER visit Ref 1.08 (0.92-1.27) Ref 1.49 (1.11-2.00) Having usual source of care 1.08 (0.89-1.30) 0.90 (0.62-1.30) Having hospitalization 0.93 (0.79-1.09) 0.96 (0.62-1.48)