Longitudinal associations between loneliness, social isolation, and healthcare utilisation trajectories: a latent growth curve analysis

Abstract Purpose To explore the longitudinal associations between eight-year trajectories of loneliness, social isolation and healthcare utilisation (i.e. inpatient, outpatient, and nursing home care) in US older adults. Methods The study used data from the Health and Retirement Study in 2006–2018, which included a nationally representative sample of American adults aged 50 and above ( N = 6,832). We conducted latent growth curve models to assess the associations between trajectories of loneliness and isolation and healthcare utilisation over 8 years. Results Independent of sociodemographic and health-related confounders, social deficits were associated with a lower likelihood of baseline physician visits (loneliness β = -0.15, SE = 0.08; social isolation β = -0.19, SE = 0.08), but there was a positive association between loneliness and number of physician visits ( β = 0.06, SE = 0.03), while social isolation was associated with extended hospital ( β = 0.07, SE = 0.04) and nursing home stays ( β = 0.05, SE = 0.02). Longer nursing home stays also predicted better trajectories of loneliness and isolation over time. Conclusion Loneliness and social isolation are cross-sectionally related to complex patterns of different types of healthcare. There was no clear evidence that social deficits led to specific trajectories of healthcare utilisation, but nursing home stays may over time help provide social contact, supporting trajectories of isolation and potentially loneliness. Non-clinical services such as social prescribing could have the potential to address unmet social needs and further promote patients’ health-seeking profiles for improving healthcare equity.