Table 2 Multilevel linear regression results of the interrelation between GP and practice characteristics and GP job satisfaction (N_country = 34; N_GPs = 7,379)

Fixed coefficients

 Constant

2.505 (.034)***

2.370 (.059)***

Stimulation

Breadth of service profile:

 First contact care

 Management of disease

 Technical procedures

 Health promotion

.009 (.009)

.000 (.009)

.028 (.008)***

.061 (.024)**

Medical equipment

.001 (.001)

Other paid activities (ref = no other paid activities)

.003 (.008)

Practice location (ref = big inner city)

 Suburbs or small town

 Urban–rural or rural

− .000 (.009)

− .005 (.009)

Comfort

 Working hours

− .001 (.000)***

 Administrative work

− .000 (.000)

 Hours spent in out-of-hours work

.000 (.000)

 ICT use

.000 (.002)

 Vacation

.007 (.002)**

Practice building (ref = not clean and no privacy)

 Clean or private

 Clean and private

.000 (.022)

.001 (.022)

Status

 Self-employed (ref = salaried)

.000 (.011)

Behavioural confirmation

 Feedback from colleagues (ref = no feedback)

.027 (.008)**

 Patient satisfaction

.095 (.032)**

 Shared practice (ref = solo)

.001 (.008)

Control variables

 Age

.000 (.000)**

 Gender (ref = male)

− .013 (.007)

Random coefficients

 Level: country variance

.038 (.009)***

.033 (.008)***

 Level: GP/practice variance

.078 (.001)***

.077 (.001)***

 N: country

34

34

 N: practice

7379

7379

 ICC

.329

.302

 − 2*log-likelihood

2294.18

2139.39

 Change in − 2LL (df)

− 154.79 (28)***