Objectives: This study estimates health utility decrements associated with social isolation and loneliness (SIL) using nationally representative longitudinal data from Australia.
Methods: Four waves of the Household, Income and Labour Dynamics in Australia survey were analyzed. Health utility was measured using the Short Form 6-Dimension with the Australian weighting algorithm. Social isolation was defined as living alone with infrequent social contact, and loneliness as a score ≥5 on a 1 to 7 scale. Respondents were classified as neither, socially isolated only, lonely only, or both. Individual fixed-effects panel regressions were applied to estimate within-person associations, with adjustment for time-varying health and sociodemographic factors.
Results: A total of 53 108 observations from 21 965 individuals (mean age 44.9; 53% male) were included, with 20% experiencing SIL (3% socially isolated only, 15% lonely only, 2% both). Compared with no SIL, adjusted models showed utility decrements of 0.020 for social isolation, 0.061 for loneliness, and 0.102 for both (all P < .001). The combined decrements significantly exceeded the sum of individual estimates (P < .05), suggesting interaction effects. Loneliness had the greatest impact on young adults (15-24 years), whereas social isolation affected middle-aged females (25-44 years) the most. Results were robust across sensitivity analyses, except that applying the UK weighting algorithm yielded smaller estimates and no significant interaction effects.
Conclusions: SIL is independently associated with reduced health utility, with some variation by age and sex. Potential interaction effects of SIL on health utility should be considered in economic evaluation, while recognizing their sensitivity to the choice of weighting algorithm.