A cohort of 490,114 adults aged 40 to 69 was followed from 2006 to 2022 to test whether the association between baseline social isolation and mortality differed by usual non-commute transport mode. Social isolation carried a hazard ratio of 1.34 for all-cause mortality and 1.49 for cardiovascular mortality. Interaction estimates were small and their intervals reached the null, at 1.17 for public transport against car only and 0.93 for mixed modes including a car.
Why it is interesting: It tests whether an everyday, modifiable feature of daily life alters the isolation and mortality association, which is the sort of handle an intervention would need.