Table 3. Association between stress and factors that might influence stress among 74 students measured at 21 different time points, every third week (estimates of mean difference compared with a reference or for one-point increase for the continuous variables, confidence interval (CI) from linear mixed regression analysis)
Independent Factor Category Adjusted for Time1 p-value2 Final Model3 p-value
Estimate (95% CI) Estimate (95% CI)
Intercept                                                                                                     3.47 (2.89-4.05) <0.001
Time of data collectiona Continuous 0.08 (0.03-0.12) 0.002 –0.03 (-0.09-0.02) 0.23
Squared timea Continuous –0.003 (-0.01-–0.00) 0.004 0.002 (-0.00–0.01) 0.08
Gender Male Reference Reference
Female 0.41 (-0.02–0.86) 0.06 0.36 (-0.08–0.8) 0.11
Age group <27 years 0.12 (-0.43-0.66) 0.91 Not available
27–30 years 0.11 (-0.52–0.74)
>30 years Reference
Presence of children No Reference 0.88 Not available
Yes 0.05 (-0.59–0.69)
Previous university degree No Reference
Yes –0.14 (-0.72–0.44) 0.23 Not available
Missing
Family origin Swedish Reference 0.95 Not available
European 0.11 (-0.62–0.84)
Non-European 0.12 (-0.72–0.96)
Missing N/A -
CanMEDS rolesb Total        
  Medical expert –0.64 (-0.82–0.45) <0.001   –0.16, (-0.43–0.12) 0.27
  All other roles Reference   Reference  
  Communicator 0.17 (0.00–0.35) 0.06 0.40 (0.15–0.65) 0.002
  All other roles Reference Reference  
  Collaborator 0.02 (-0.22–0.18 0.04 0.43 (0.17–0.70) 0.002
  All other roles Reference   Reference  
  Scholar 0.35 (0.18–0.52) <0.001 0.51 (0.25–0.76) <0.001
  All other roles Reference   Reference  
  Manager 0.63 (0.41–0.84) <0.001 0.72 (0.44–1.00) <0.001
  All other roles Reference   Reference  
  Health advocate –0.56 (-0.90-–0.22) 0.001 –0.39 (-0.75–0.03) 0.03
  All other roles Reference Reference  
  Professional 0.03 (-0.18–0.24) 0.08 0.33 (0.07–0.59) 0.01
  All other roles Reference   Reference  
Important life eventc Yes 0.40 (0.19–0.61) <0.001 0.37 (0.16–0.58) <0.001
  No Reference   Reference  

1) Adjusted for a random intercept, time, squared time and each of the listed factors (one at a time). Estimates for time and squared time are reported when a random intercept, but no other factors, are included. 2) p-value is reported as a fixed effect test, and all factors with p <0.1 were included in the model adjusted for all factors, indicated in italics. 3) Adjusted for a random intercept, time-dependent factor, gender, all CanMEDS roles, and life events. The mean stress for the final estimated model was calculated as: Stress = 3.47 - 0.03*Time + 0.002*Time*Time + 0.36*Gender – 0.16 * Medical Expert + …+ 0.08*Professionals + 0.37 *Life Event, where Time = 0,1,…,21, and all other factors are 1 or 0 (reference). *Estimates of mean difference compared with a reference or for one-point increase for the continuous variables, presented with confidence interval [CI] a) Time was used a continuous variable (coded as 0,1,…,20) where each step corresponds to 3 weeks. Squared time is included to account for a changing effect of time. b) The total number of learning activities is based on all individuals’ accumulated responses, i.e., n = 1390, and the percentages are calculated relative to this number. Each CanMEDS role was treated as an independent factor in the mixed model. c) Important life event, defined as something that had an impact on the student’s daily life during the last 3-week period

Int J Med Educ. 2019; 10:68-74; doi: 10.5116/ijme.5c94.9391