Loneliness and long-term health-related quality of life: A latent class analysis among middle-aged and older Australian women

Abstract Background Loneliness has been long associated with poor health outcomes including health-related quality of life (HRQoL). However, few studies used longitudinal data and none to our knowledge considered the dynamic nature of loneliness over time. Objective To identify longitudinal patterns of loneliness over 18 years and its association with physical and mental HRQoL of middle-aged and older women. Methods Data were from the Australian Longitudinal Study of Women's Health. Latent Class Analysis was conducted to identify long-term loneliness patterns. Multinomial logistic regression was used to examine the prospective associations between baseline predictors of loneliness and loneliness trajectories. Linear regression was used to examine the association between loneliness trajectories and HRQoL. Results Baseline predictors such as current smoking, depression, anxiety, stress and low social engagement were associated with higher odds of increasing, stable medium and stable high loneliness. Compared to stable low loneliness, increasing [B = -3.73 (95%CI = -5.42, -2.04)], medium [B = -3.12 (95%CI = -5.08, -1.15)] and high loneliness [B = -5.67 (95%CI = -6.84, -4.49)] were associated with lower mental HRQoL. The increasing loneliness class was also associated with lower physical HRQoL [B = -1.06 (95%CI = -2.11, -0.02)]. Out of all HRQoL sub-scales, emotional role, social functioning and physical role were the most strongly associated with loneliness. Conclusions Long-term loneliness patterns are consistently associated with prospective physical and mental HRQoL. This finding highlights the importance of addressing loneliness among women to promote their health and well-being.