Table 2.
Hazard ratios of work-related factors and fatigue on partial return to work
| Univariate (step 1) | Adjusted (step 2)a | Multivariate, adjusted (step 3)a, b | |||||
|---|---|---|---|---|---|---|---|
| HR | 95% CI | HR | 95% CI | HR | 95% CI | Χ2(p value)c | |
| Confounding factors | |||||||
| Sole breadwinner (reference: no or shared breadwinner) | 1.812 | 1.013–2.977 | – | – | – | – | – |
| Covariates | |||||||
| Low work ability (reference: “adequate” work ability) | 0.540 | 0.324–0.900 | 0.501 | 0.296–0.848 | – | – | – |
| Higher job self-efficacy (increase in one point) | 1.446 | 0.999–2.094 | 1.536 | 1.034–2.282 | 1.329d | 0.877–2.014 | 2.034 (.174) |
| Low value of work (reference: “adequate” value of work) | 0.570 | 0.345–0.939 | 0.567 | 0.338–0.950 | 0.624 | 0.369–1.055 | 3.494 (.076) |
| Low fatigue (reference: considerable fatigue) | 1.753 | 1.044–2.943 | 1.631 | 0.948–2.870 | 1.443 | 0.823–2.529 | 1.824 (.200) |
An HR (hazard ratio) > 1 reflects a shorter time to return to work. CI confidence interval
aAdjustments were stratification for time since first chemo, and including breadwinner status as predictor
bEffect over perceived work ability
cImprovement of the model by adding the covariate mean of 20 imputations
dHR of work ability was not statistically significant in this model