Skip to main content
. 2023 Jul 25;13:12042. doi: 10.1038/s41598-023-39113-6

Table 4.

Results of the hierarchical model with presenteeism as the outcome variable.

Predictors Estimate (%Δ) Estimate (log) SE t value p value* CI (2.5%)* CI (97.5%)* VIF R2
(Intercept) 0.941 0.093 10.174  < 0.001 0.760 1.123 0.25
Setting: psychiatry1 14.9 0.139 0.034 4.046  < 0.001 0.075 0.207 1.39
Setting: nursing home1 18.3 0.168 0.033 5.098 0.105 0.233
Setting: home care organization1 9.6 0.092 0.039 2.357 0.012 0.163
French-speaking language region2 − 26.2 − 0.304 0.033 − 9.112  < 0.001 − 0.368 − 0.241 1.12
Italian-speaking language region2 − 14.3 − 0.154 0.075 − 2.051 − 0.305 − 0.011
Employment level (working hours per week) 0.3 0.003 0.000 5.989  < 0.001 0.002 0.004 1.14
Compliance with legal break times (yes = 1) − 10.0 − 0.105 0.024 − 4.431  < 0.001 − 0.154 − 0.061 1.07
Profession: physician − 13.5 − 0.145 0.040 − 3.639  < 0.001 − 0.223 − 0.067 1.10
Profession: administration & research 20.8 0.189 0.066 2.846  < 0.007 0.052 0.319 1.02
Quantitative demands at work 0.4 0.004 0.001 6.283  < 0.001 0.003 0.005 1.29
Emotional demands at work 0.4 0.004 0.001 5.602  < 0.001 0.002 0.005 1.27
Demands to hide emotions 0.3 0.003 0.000 5.860  < 0.001 0.002 0.004 1.22
Perceived reward − 0.20 − 0.002 0.000 − 5.167  < 0.001 − 0.003 − 0.001 1.25
Insecurity of the working environment 0.3 0.003 0.000 6.097  < 0.001 0.002 0.004 1.34
Work-private life conflict 0.5 0.005 0.001 9.115  < 0.001 0.004 0.006 1.65
Demanding work environment 0.3 0.003 0.001 5.112  < 0.001 0.002 0.004 1.30
General health status − 1 − 0.010 0.001 − 15.822  < 0.001 − 0.011 − 0.008 1.18
 Random effects
Std.Dev. 0.06 (0.00–0.09)
Residuals 0.70 (0.68–0.72)

*Based on bootstrap, 1settings was used as a categorial variable with acute care/rehabilitation hospitals as an indicator, 2language regions was used as a categorical variable with the German-speaking region as an indicator, SE = standard error, VIF = variance inflation factor, estimate (%Δ): percentage increase (or decrease) in the response for every one-unit increase in the predictor variable (exp(β1)− 1)⋅100.