This study aimed to assess the prevalence and determinants of self-reported loneliness and to examine whether loneliness is associated with subsequent mortality among adults aged 65 years and older in Latin America, China, and India. Data were drawn from population-based cross-sectional surveys conducted between 2003 and 2007, with longitudinal follow-up from 2007 to 2010, as part of the 10/66 Dementia Research Group study. Poisson regression models were used to identify factors associated with loneliness, while Cox proportional hazards models were applied to evaluate the relationship between loneliness and mortality. Age-standardised estimates indicated that the prevalence of loneliness ranged from 25.3% to 32.4% across Latin American sites and was 18.3% in India. In contrast, loneliness was uncommon in China, with a prevalence of 3.8%. Meta-analyses combining results across countries provided strong evidence of an association between loneliness and increased mortality risk in Latin America (hazard ratio [HR] = 1.13, 95% confidence interval [CI] 1.01–1.26; I² = 10.1%) and in China (HR = 1.58, 95% CI 1.03–2.41). No significant association between loneliness and mortality was observed in India. These findings indicate that cultural differences may influence how loneliness is experienced or reported in later life. While loneliness was consistently linked to higher mortality risk across most cultural contexts examined, this relationship was not evident in India. Consequently, loneliness should be recognised as an important public health concern among older populations.
