Social isolation and loneliness and their association with all-cause mortality. A population-based longitudinal study in Norway: The Tromsø Study 1994–2023

Abstract

Objectives

Loneliness and social isolation are associated with increased mortality, but few studies have assessed this association over long time in young adults.

Methods

The study sample comprised 9061 women and 8735 men aged 25 to 69 years who participated in the Tromsø4 survey (1994–95, baseline) of the Tromsø Study, Norway. A subset of the study sample also attended the Tromsø5 (2001), Tromsø6 (2007–08), and Tromsø7 (2015–16) surveys. Participants were followed up for all-cause mortality until November 2023; with 1630 women and 2099 men deceased. Information on social isolation (least isolated, modestly isolated, and most isolated) and loneliness (yes, no) were taken from self-administered questionnaires. Sex-specific, time-varying Cox models were employed, updating exposures and covariates from Tromsø5.

Results

Most-isolated versus least-isolated women and men had hazard ratios of 1.37 (95 % confidence interval 1.18–1.59) and 1.41 (1.25–1.60), respectively, after adjustment for covariates. These hazard ratios were higher in younger adults (HR = 1.55 in women and HR = 1.76 in men aged <50 years at baseline), though the age-isolation interaction was not statistically significant in women (P = 0.26), but in men (P = 0.01). For loneliness, the adjusted hazard ratios were 1.51 (1.23–1.87) and 1.46 (1.16–1.84). Over time, 51 % and 47 % of participants remained most isolated at Tromsø5 and Tromsø7, respectively; 25 % of those initially lonely remained so at Tromsø5, while only 2.6 % of those initially non-lonely became lonely at Tromsø5.

Conclusion

Both social isolation and loneliness are strongly associated with all-cause mortality, particularly among younger adults, underscoring their importance as public health concerns.