Of 6,551 Australians aged 50 and over in a national household panel survey, 1,529 died over 19 waves. Five measures were modelled jointly: loneliness, lack of emotional support, infrequent social contact (social isolation), club membership and living alone. Cox models counted each measure only if reported in two consecutive years, adjusting for health, income and deprivation. The hazard ratio for loneliness was 1.41, at 1.55 in men and 1.24 in women, against 1.19 for social isolation. Active club membership was linked with lower risk (hazard ratio 0.69), but living alone was not (1.03). The design is observational with self-reported measures, so reverse causation and residual confounding cannot be excluded.
Why it is interesting: It estimates each aspect of social relationships separately rather than pooling them in one isolation index. Loneliness was linked to more risk than past smoking but much less than current smoking.
Cite this item
BibTeX
Free sign-up. A digest of the new papers twice a week, grouped by domain.
Other drivers in Behavioural science
-
Diet, exercise and sleep
What people eat and how much they move: what the measured evidence says each does to mortality, and what has been shown to change either.
1 this week -
Smoking and vaping
How far smoking has fallen and how many deaths it still causes, what vaping is doing to the people who take it up, and what generational bans will show.
None this week Earlier → -
Alcohol and drug use
What is happening to deaths from alcohol, drugs and suicide, whether they move together, and how much of it is behaviour and how much circumstance.
4 this week -
Social connection
Isolation, loneliness and mortality: what the evidence shows, how far the relationship is causal, and what interventions have shown.
1 this week -
Infection and pandemics
Infection and pandemics: the risks, and how well prepared we are
5 this week -
Adherence and persistence
Whether a treatment is actually taken, for how long, and what changes either.
4 this week
