Figure 1
Summary of structural equation modelling between social isolation measures and selected characteristics at baseline (2008 to 2009) and mortality (2008 to 2017), by sex, household population aged 65 or older, Canada excluding territories

The direct and indirect impact of subjective isolation and low participation on survival were assessed using Structural equation models (SEM). Indirect paths were assessed with functional health status as a single hypothesized mediator. The SEM analyses controlled for covariates found to be significantly associated with mortality in the fully-adjusted Cox proportional hazards models (men: age, smoking status, physical activity; women: age, smoking status, physical activity, household education, living arrangement).

The figure includes two diagrams summarizing the results of SEM. In the top figure (1a) the model results are presented for men. In the bottom figure (1b) the model results are presented for women.

Four boxes are shown on the diagram; on the left had side there is a box for low social participation and one for subjective isolation; in the middle is a box for health status; on the right had side there is a box for mortality (survival time). Arrows flow between the boxes. Beta co-efficients, confidence intervals and p-values from the model are presented beside each arrow. Direct pathways are represented by arrows that flow from the low social participation and subjective isolation boxes to the mortality box. Indirect pathways flow from the social participation and subjective isolation boxes to the health status box, and then from the health status box to the mortality box. Significant pathways are represented by a solid line, non-significant pathways by a dashed line. For both men and women the direct pathway between subjective isolation and mortality is represented by a dashed line (non-significant). All other pathways are statistically significant.