*3.2.4 Multivariate analysis of 2016 cohort*

The binary logistic regression showed an explanatory model of the moral harassment at work for 2016 Cohort consisting of job position (p = 0.008), specialty (p = 0.031), deliberate choice of medical studies (p < 10−3) and Medical practice dissatisfaction (p < 10−3) (**Table 4**).

#### *3.2.5 Comparison of both cohorts' scores*

Comparison of the NAQ-R scores between 2009 and 2016 showed that it was significantly higher in 2009 score, while no difference was noticed in bullying scores between both cohorts (**Table 5**).


#### **Table 3.**

*Multivariate model of moral harassment at work in 2009 cohort.*


#### **Table 4.**

*Multivariate model explaining workplace bullying in 2016 cohort.*


**Table 5.**

*Comparison of both cohorts' NAQ-R scores.*

### **3.3 Impact of MHW on the quality of life**

#### *3.3.1 Mental health score (MCS)*

The median MCS was 40.3 ± 12 in 2009 and 40.61 ± 13.48 in 2016. No significant difference was recorded for the MCS. However, the majority of respondents had a mental health score below the standard of 50 in 2009 Cohort (86.1%) as well in 2016 Cohort (84.8%).

Univariate regression showed that workplace bullying influenced the mental quality of mental life significantly in 2009 (p < 10−3, SD = [2.63; 7.06]) as well as in 2016 (p < 10−3, SD = [3.14, 7.04]).

## *3.3.2 Physical health score (PCS)*

The median score for the PSC was 45.6- ± 11.72009 and 45.36 ± 11.09 in 2016 with no statistically significant difference between both cohorts.

In 2009 Cohort as well as in 2016 one, the majority of participants had comparable PSC < 50 (69.8%/72%) with no significant difference.

Univariate regression showed that workplace bullying had significantly influenced the physical quality of life in 2009 (p = 0.013, SD = [0.49; 4.2]) and in 2016 (p = 0.004, SD = [0.86, 4.37]).
