Cox proportional hazards regression on purpose in life at 2004/2005 predicting T2D incidence from 2006/2007 to 2016/2017—potential mediators and confounders (N=4134)
| HR | 95% CI | P value | PPAE | |
|---|---|---|---|---|
| Unadjusted model | 0.85 | 0.74 to 0.99 | 0.032 | |
| +Sociodemographic variables | 0.90 | 0.77 to 1.04 | 0.154 | 33% |
| +Behavioural variables | 0.87 | 0.75 to 1.01 | 0.066 | 13% |
| +Clinical variables | 0.93 | 0.80 to 1.09 | 0.389 | 53% |
| +Sociodemographic, behavioural and clinical variables | 0.92 | 0.79 to 1.08 | 0.322 | 47% |
CI, confidence interval; HR, hazards ratio, n, number; PPAE, percentage of protective association explained; T2D, type 2 diabetes.