Skip to main content

Table 4 Results of the univariable and multivariable logistic regression analyses for psychological stress

From: Associations of perceived changes in work due to digitalization and the amount of digital work with job strain among physicians: a national representative sample

Variable

Univariable model

Multivariable model

OR

95% CI

P value

OR

95% CI

P value

Perceived changes in work due to digitalization

More active role of patients

 Neutral

ref

  

ref

  

 Disagree

1.15

.97, 1.37

.10

.99

.81, 1.20

.88

 Agree

1.12

.97, 1.30

.12

1.19

1.02, 1.40

.03

Improved possibilities for preventive work

 Neutral

ref

  

ref

  

 Disagree

1.26

1.09, 1.45

 < .001

1.15

.96, 1.38

.13

 Agree

.87

.73, 1.02

.09

.92

.76, 1.11

.41

Progressed interprofessional collaboration

 Neutral

ref

  

ref

  

 Disagree

1.20

1.03, 1.40

.02

.97

.80, 1.17

.74

 Agree

1.01

.87, 1.17

.96

1.11

.94, 1.31

.22

Supported decision-making

 Neutral

ref

  

ref

  

 Disagree

1.31

1.14, 1.32

 < .001

1.26

1.06, 1.48

.007

 Agree

.99

.84, .89

.91

1.02

.85, 1.23

.83

Facilitated access to patient information

 Neutral

ref

  

ref

  

 Disagree

1.11

.93, 1.32

.24

.87

.72, 1.06

.17

 Agree

.76

.65, .89

 < .001

.84

.70, .98

.04

Accelerated clinical encounters

 Neutral

ref

  

ref

  

 Disagree

1.30

1.12, 1.51

 < .001

1.05

.89, 1.24

.58

 Agree

.65

.52, .82

 < .001

.72

.56, .92

.009

Amount of digital work

Number of HISs in daily use

  ≤ 2

ref

  

ref

  

  ≥ 3

1.39

1.22, 1.58

 < .001

1.19

1.04, 1.47

.01

Frequency of teleconsultations

 Not at all–to some extent

ref

  

ref

  

 Much–very much

1.45

1.36, 1.55

 < .001

1.47

1.22, 1.76

 < .001

  1. N = 4271, OR The Odds Ratio indicated, on average, how many times greater or lower odds of psychological stress there were for a certain group compared to a reference group; CI   Confidence interval, ref. Reference, HISs Health information systems. The univariable model measured the independent effect of each variable with psychological stress separately. The multivariable model included simultaneously the variables of perceived changes in work due to digitalization and amount of digital work, adjusted by gender, experience with the current EHR, EHR grade, and working sector. The multivariable model was significant, χ2 (23, N = 4271) = 243.74, P < .001, explaining 8% (Nagelkerke R2) of the variance in psychological stress. The goodness-of-fit of the model was good (Hosmer–Lemeshow test with P > .05), and the model correctly classified 63% of cases. Variables were based on the questionnaire of the Electronic Health Record Systems as a Tool for Physicians 2021 Study [61]