Sample Characteristics
Sample characteristics as well as means and standard deviations of study variables can be found in the Supplementary Information (Table S1).
Depressive Symptoms
Observations were non-independent as reflected in an ICC of .22. The model including crisis-related distress fitted our data significantly better than the intercept-only model (χ2diff(3) = 759.84, p < .001). Including a random slope for Pandemic-related distress further improved model fit (χ2diff (2) = 19.87, p < .001). Variance in slopes is illustrated in Fig. 1.
--- Insert Fig. 1 here ---
Including random slopes for war-related and climate-related distress did not result in further improvements of model fit (p > .05). Including socio-demographic variables further improved model fit (χ2diff (4) = 307.65, p < .001) as did individual stressor-related distress (χ2diff (1) = 760.37, p < .001) and resilience factors (χ2diff (2) = 216.48, p < .001). Table 1provides an overview of intercepts and regression weights as well as estimated variance accounted for by fixed and random effects in each model.
Table 1
Model summaries of linear mixed model analyses for depressive symptoms
|
Intercept only
|
Baseline model
|
+ RE Distress – Pandemic+
|
+ FE Sociodemographics
|
+ FE Distress - Individual
|
+ FE Resilience factors
|
Predictors
|
B
|
CI
|
p
|
B
|
CI
|
p
|
B
|
CI
|
p
|
B
|
CI
|
p
|
B
|
CI
|
p
|
B
|
CI
|
p
|
(Intercept)
|
1.02
|
0.99–1.05
|
< 0.001
|
1.02
|
0.99–1.06
|
< 0.001
|
1.02
|
0.99–1.06
|
< 0.001
|
1.03
|
0.99–1.06
|
< 0.001
|
1.03
|
0.99–1.06
|
< 0.001
|
1.03
|
0.99–1.06
|
< 0.001
|
Distress - Pandemic
|
|
|
|
0.11
|
0.10–0.12
|
< 0.001
|
0.11
|
0.10–0.12
|
< 0.001
|
0.09
|
0.08–0.10
|
< 0.001
|
0.05
|
0.04–0.06
|
< 0.001
|
0.05
|
0.04–0.06
|
< 0.001
|
Distress - War
|
|
|
|
0.00
|
-0.01–0.02
|
0.520
|
0.00
|
-0.01–0.02
|
0.478
|
-0.00
|
-0.01–0.01
|
0.974
|
-0.01
|
-0.02–0.00
|
0.204
|
-0.00
|
-0.01–0.01
|
0.427
|
Distress - Climate
|
|
|
|
0.05
|
0.04–0.06
|
< 0.001
|
0.05
|
0.04–0.06
|
< 0.001
|
0.04
|
0.03–0.05
|
< 0.001
|
0.03
|
0.02–0.04
|
< 0.001
|
0.03
|
0.02–0.04
|
< 0.001
|
Age
|
|
|
|
|
|
|
|
|
|
0.04
|
-0.00–0.08
|
0.071
|
0.03
|
-0.01–0.06
|
0.146
|
0.03
|
-0.01–0.07
|
0.103
|
Ses
|
|
|
|
|
|
|
|
|
|
-0.05
|
-0.06 – -0.03
|
< 0.001
|
-0.03
|
-0.04 – -0.01
|
< 0.001
|
-0.02
|
-0.03 – -0.01
|
< 0.001
|
Sex - Female
|
|
|
|
|
|
|
|
|
|
0.18
|
0.15–0.20
|
< 0.001
|
0.12
|
0.10–0.15
|
< 0.001
|
0.10
|
0.08–0.12
|
< 0.001
|
Sex - Diverse
|
|
|
|
|
|
|
|
|
|
0.78
|
0.62–0.94
|
< 0.001
|
0.66
|
0.51–0.80
|
< 0.001
|
0.57
|
0.43–0.71
|
< 0.001
|
Distress - Individual
|
|
|
|
|
|
|
|
|
|
|
|
|
0.06
|
0.05–0.06
|
< 0.001
|
0.05
|
0.05–0.05
|
< 0.001
|
Self-Efficacy
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
-0.17
|
-0.19 – -0.15
|
< 0.001
|
Expressive Flexibility
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
-0.00
|
-0.01 – -0.00
|
0.018
|
Random Effects
|
σ2
|
0.55
|
0.44
|
0.43
|
0.39
|
0.30
|
0.28
|
τ00
|
0.03 School:Class
|
0.04 School:Class
|
0.04 School:Class
|
0.05 School:Class
|
0.06 School:Class
|
0.07 School:Class
|
|
0.00 School
|
0.00 School
|
0.00 School
|
0.00 School
|
0.00 School
|
0.00 School
|
τ11
|
|
|
0.00 School:Class.PANDEMIC
|
0.00 School:Class.PANDEMIC
|
0.00 School:Class.PANDEMIC
|
0.00 School:Class.PANDEMIC
|
ρ01
|
|
|
0.84 School:Class
|
0.73 School:Class
|
0.62 School:Class
|
0.56 School:Class
|
ICC
|
0.05
|
0.09
|
0.11
|
0.13
|
0.20
|
0.22
|
N
|
57 School
|
57 School
|
57 School
|
57 School
|
57 School
|
57 School
|
|
445 Class
|
445 Class
|
445 Class
|
445 Class
|
445 Class
|
445 Class
|
Observations
|
3603
|
3603
|
3603
|
3603
|
3603
|
3603
|
Marginal R2 / Conditional R2
|
0.000 / 0.049
|
0.176 / 0.248
|
0.175 / 0.264
|
0.236 / 0.338
|
0.360 / 0.485
|
0.390 / 0.523
|
Note: + Including the school-level random slope for Pandemic-related distress resulted in model non-convergence due to extremely low variance in slopes between schools. Hence, we only included the class-level random slope in all subsequent models. RE = Random effect, FE = Fixed effect(s), B = unstandardized regression weight, CI = Confidence Interval, p = significance level, SES = Socioeconomic Status, ICC = Intraclass correlation.
--- Insert Table 1 around here---
In the final model, fixed effects were estimated to account for 39% of variance in depressive symptoms. Distress related to individual stressors was found to be the strongest predictor, reflecting that participants with higher distress ratings reported more symptoms (β = .36, SE = .01, p < .001). The next highest predictor was self-efficacy, which predicted fewer symptoms (β = − .18, SE = .01, p < .001). Pandemic-related distress (β = .15, SE = .02, p < .001), female sex (β = .12, SE = .01, p < .001), and diverse sex (β = .10, SE = .01, p < .001), Climate-related distress (β = .09, SE = .01, p < .001) were found to be related to higher symptoms. Finally, higher socio-economic status (β = − .04, SE = .01, p < .001) and higher expressive flexibility (β = − .03, SE = .01, p = .018) were found to be linked to fewer symptoms.
Anxiety Symptoms
Observations were non-independent as reflected in an ICC of .20. The model including crisis-related distress fitted our data significantly better than the intercept-only model (χ2diff(3) = 998.84 p < .001). Including random slopes for pandemic-related, war-related, and climate-related distress did not result in further improvements of model fit (p > .05). Including socio-demographic variables further improved model fit (χ2diff (4) = 346.58, p < .001) as did introduced individual stressor-related distress (χ2diff (1) = 458.92, p < .001) and resilience factors (χ2diff (2) = 188.78, p < .001). Table 2provides an overview of intercepts and regression weights as well as estimated variance accounted for by fixed and random effects in each model.
Table 2
Model summaries of linear mixed model analyses for anxiety symptoms
|
Intercept only+
|
Baseline model
|
+ FE Sociodem.
|
+ FE Distress
– Individual
|
+ FE Resilience factors
|
Predictors
|
B
|
CI
|
p
|
B
|
CI
|
p
|
B
|
CI
|
p
|
B
|
CI
|
p
|
B
|
CI
|
p
|
(Intercept)
|
1.03
|
1.01–1.05
|
< 0.001
|
1.03
|
1.01–1.05
|
< 0.001
|
1.03
|
1.01–1.05
|
< 0.001
|
1.03
|
1.01–1.05
|
< 0.001
|
1.03
|
1.01–1.05
|
< 0.001
|
Distress - Pandemic
|
|
|
|
0.08
|
0.07–0.08
|
< 0.001
|
0.06
|
0.06–0.07
|
< 0.001
|
0.04
|
0.03–0.05
|
< 0.001
|
0.04
|
0.03–0.04
|
< 0.001
|
Distress - War
|
|
|
|
0.02
|
0.02–0.03
|
< 0.001
|
0.02
|
0.01–0.03
|
< 0.001
|
0.02
|
0.01–0.02
|
< 0.001
|
0.02
|
0.01–0.02
|
< 0.001
|
Distress - Climate
|
|
|
|
0.04
|
0.03–0.05
|
< 0.001
|
0.03
|
0.03–0.04
|
< 0.001
|
0.03
|
0.02–0.03
|
< 0.001
|
0.03
|
0.02–0.03
|
< 0.001
|
Age
|
|
|
|
|
|
|
0.01
|
-0.02–0.04
|
0.469
|
0.01
|
-0.02–0.03
|
0.707
|
0.00
|
-0.02–0.03
|
0.726
|
Ses
|
|
|
|
|
|
|
-0.02
|
-0.03 – -0.01
|
< 0.001
|
-0.01
|
-0.02 – -0.00
|
0.039
|
-0.01
|
-0.02–0.00
|
0.155
|
Sex - Female
|
|
|
|
|
|
|
0.16
|
0.15–0.18
|
< 0.001
|
0.14
|
0.12–0.15
|
< 0.001
|
0.12
|
0.10–0.13
|
< 0.001
|
Sex - Diverse
|
|
|
|
|
|
|
0.16
|
0.05–0.27
|
0.006
|
0.09
|
-0.01–0.20
|
0.088
|
0.04
|
-0.06–0.14
|
0.446
|
Distress - Individual
|
|
|
|
|
|
|
|
|
|
0.03
|
0.03–0.03
|
< 0.001
|
0.03
|
0.02–0.03
|
< 0.001
|
Self-Efficacy
|
|
|
|
|
|
|
|
|
|
|
|
|
0.00
|
0.00–0.01
|
0.012
|
Expressive Flexibility
|
|
|
|
|
|
|
|
|
|
|
|
|
-0.12
|
-0.14 – -0.11
|
< 0.001
|
Random Effects
|
σ2
|
0.29
|
0.22
|
0.19
|
0.17
|
0.16
|
τ00
|
0.02 School:Class
|
0.03 School:Class
|
0.03 School:Class
|
0.04 School:Class
|
0.04 School:Class
|
ICC
|
0.06
|
0.12
|
0.15
|
0.18
|
0.20
|
N
|
57 School
|
57 School
|
57 School
|
57 School
|
57 School
|
|
446 Class
|
446 Class
|
446 Class
|
446 Class
|
446 Class
|
Observations
|
3624
|
3624
|
3624
|
3624
|
3624
|
Marginal R2 / Conditional R2
|
0.000 / 0.065
|
0.222 / 0.316
|
0.283 / 0.387
|
0.355 / 0.472
|
0.380 / 0.503
|
Note: + Including the school-level random intercept resulted in non-convergence for some models due to extremely low variance in intercepts between schools. Hence, we only included the class-level random intercept in all models. FE = Fixed effect(s), B = unstandardized regression weight, CI = Confidence Interval, p = significance level, SES = Socioeconomic Status, ICC = Intraclass correlation.
--- Insert Table 2 around here---
In the final model, fixed effects were estimated to account for 38% of variance in anxiety symptoms. Distress related to individual stressors was found to be the strongest predictor, reflecting that participants with higher distress ratings reported more symptoms (β = .27, SE = .01, p < .001). The next highest predictors were female sex (β = .18, SE = .01, p < .001) and self-efficacy (β =-.18, SE = .01, p < .001), which predicted more and fewer symptoms respectively. Pandemic-related distress (β = .15, SE = .01, p < .001), climate-related distress (β = .10, SE = .01, p < .001), and war-related distress (β = .07, SE = .01, p < .001) were found to be related to higher symptoms. Finally, contrary to our assumption, higher expressive flexibility was found to be linked to higher symptoms (β = .03, SE = .01, p < .001).
Health-related quality of life (HRQoL)
Observations were non-independent as reflected in an ICC of .20. The model including crisis-related distress fitted our data significantly better than the intercept-only model (χ2diff(3) = 467.20, p < .001). Including a random slope for Pandemic-related distress further improved model fit (χ2diff (4) = 11.46, p = .022). Variance in slopes is illustrated in Fig. 2.
--- Insert Fig. 2 around here ---
Including random slopes for war-related and climate-related distress did not result in further improvements of model fit (p > .05). Including socio-demographic variables further improved model fit (χ2diff (4) = 271.50, p < .001) as did introduced individual stressor-related distress (χ2diff (1) = 575.19, p < .001) and resilience factors (χ2diff (2) = 496.99, p < .001). Table 3provides an overview of intercepts and regression weights as well as estimated variance accounted for by fixed and random effects in each model.
Table 3
Model summaries of linear mixed model analyses for Health-related quality of life
|
Intercept only
|
Baseline model
|
+ RE Distress - Pandemic
|
+ FE Sociodemographics
|
+ FE Distress – Individual+
|
+ FE Resilience factors
|
Predictors
|
B
|
CI
|
p
|
B
|
CI
|
p
|
B
|
CI
|
p
|
B
|
CI
|
p
|
B
|
CI
|
p
|
B
|
CI
|
p
|
(Intercept)
|
45.34
|
44.77–45.92
|
< 0.001
|
45.33
|
44.75–45.90
|
< 0.001
|
45.34
|
44.76–45.91
|
< 0.001
|
45.34
|
44.78–45.90
|
< 0.001
|
45.31
|
44.74–45.88
|
< 0.001
|
45.31
|
44.74–45.88
|
< 0.001
|
Distress - Pandemic
|
|
|
|
-1.26
|
-1.42 – -1.10
|
< 0.001
|
-1.27
|
-1.44 – -1.09
|
< 0.001
|
-1.05
|
-1.22 – -0.88
|
< 0.001
|
-0.49
|
-0.65 – -0.33
|
< 0.001
|
-0.36
|
-0.51 – -0.20
|
< 0.001
|
Distress - War
|
|
|
|
-0.04
|
-0.21–0.13
|
0.653
|
-0.04
|
-0.22–0.13
|
0.624
|
0.03
|
-0.14–0.19
|
0.746
|
0.10
|
-0.05–0.25
|
0.182
|
0.04
|
-0.10–0.18
|
0.593
|
Distress - Climate
|
|
|
|
-0.59
|
-0.77 – -0.42
|
< 0.001
|
-0.59
|
-0.76 – -0.41
|
< 0.001
|
-0.48
|
-0.65 – -0.31
|
< 0.001
|
-0.31
|
-0.46 – -0.16
|
< 0.001
|
-0.31
|
-0.46 – -0.17
|
< 0.001
|
Age
|
|
|
|
|
|
|
|
|
|
-0.42
|
-1.04–0.20
|
0.186
|
-0.25
|
-0.82–0.33
|
0.398
|
-0.32
|
-0.85–0.21
|
0.237
|
Ses
|
|
|
|
|
|
|
|
|
|
0.94
|
0.73–1.15
|
< 0.001
|
0.63
|
0.43–0.82
|
< 0.001
|
0.53
|
0.35–0.71
|
< 0.001
|
Sex - Female
|
|
|
|
|
|
|
|
|
|
-2.70
|
-3.08 – -2.32
|
< 0.001
|
-2.01
|
-2.36 – -1.66
|
< 0.001
|
-1.50
|
-1.83 – -1.17
|
< 0.001
|
Sex - Diverse
|
|
|
|
|
|
|
|
|
|
-5.95
|
-8.42 – -3.47
|
< 0.001
|
-4.19
|
-6.46 – -1.93
|
< 0.001
|
-2.23
|
-4.33 – -0.13
|
0.037
|
Distress - Individual
|
|
|
|
|
|
|
|
|
|
|
|
|
-0.76
|
-0.82 – -0.70
|
< 0.001
|
-0.61
|
-0.67 – -0.56
|
< 0.001
|
Self-Efficacy
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3.84
|
3.48–4.19
|
< 0.001
|
Expressive Flexibility
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
0.16
|
0.10–0.22
|
< 0.001
|
Random Effects
|
σ2
|
116.39
|
100.67
|
98.54
|
90.34
|
75.70
|
64.33
|
τ00
|
4.91 School:Class
|
6.71 School:Class
|
7.14 School:Class
|
8.31 School:Class
|
10.52 School:Class
|
12.76 School:Class
|
|
1.80 School
|
1.86 School
|
1.82 School
|
1.62 School
|
1.69 School
|
1.61 School
|
τ11
|
|
|
0.33 School:Class.PANDEMIC
|
0.36 School:Class.PANDEMIC
|
0.23 School:Class.PANDEMIC
|
0.20 School:Class.PANDEMIC
|
|
|
|
0.01 School.PANDEMIC
|
0.00 School.PANDEMIC
|
|
0.03 School.PANDEMIC
|
ρ01
|
|
|
-0.56 School:Class
|
-0.47 School:Class
|
-0.61 School:Class
|
-0.56 School:Class
|
|
|
|
0.30 School
|
0.75 School
|
|
0.30 School
|
ICC
|
0.05
|
0.08
|
0.10
|
0.12
|
0.15
|
0.20
|
N
|
57 School
|
57 School
|
57 School
|
57 School
|
57 School
|
57 School
|
|
447 Class
|
447 Class
|
447 Class
|
447 Class
|
447 Class
|
447 Class
|
Observations
|
3595
|
3595
|
3595
|
3595
|
3595
|
3595
|
Marginal R2 / Conditional R2
|
0.000 / 0.054
|
0.114 / 0.184
|
0.115 / 0.202
|
0.173 / 0.270
|
0.281 / 0.390
|
0.359 / 0.484
|
Note: +Including the school-level random slope for Pandemic-related distress resulted in non-convergence due to extremely low variance in slopes between schools. Since this issue occured exclusively in this specific submodel we excluded the school-level slope only in this model. RE = Random effect, FE = Fixed effect(s), B = unstandardized regression weight, CI = Confidence Interval, p = significance level, SES = Socioeconomic Status, ICC = Intraclass correlation.
--- Insert Table 3 around here---
In the final model, fixed effects were estimated to account for 35.9% of variance in HRQoL. Distress related to individual stressors was found to be the strongest predictor, reflecting that participants with higher distress ratings reported lower HRQoL (β = − .30, SE = .01, p < .001). The next highest predictor was self-efficacy, which predicted higher HRQoL (β = .28, SE = .01, p < .001). Female sex (β = − .12, SE = .01, p < .001) and Pandemic-related distress (β = − .08, SE = .02, p < .001) were found to be related to higher symptoms. Higher socio-economic status (β = .07, SE = .01, p < .001) and expressive flexibility (β = .07, SE = .01, p < .001) were linked to higher HRQoL whereas higher Climate-related distress (β = − .06, SE = .01, p < .001) and diverse sex (β = − .03, SE = .01, p = .037) were associated with lower HRQoL.