Background characteristics
Of the 1136 medical care workers we invited to the study, 863 completed the questionnaire survey (response rate =76.0%). Table 1 presents the sample characteristics by adverse psychological outcomes. 29.3% were male, 77.3% were at age less than 40, 64.9% were currently married, 88.0% had a college or above educational background , 43.7% had a monthly income less than 6000 RMB (around $850),43.7% were doctors, 24.4% were nurses, 6.0% ever had chronic diseases, 7.4% were current tobacco users and 32.6% were current alcohol users. In addition, 25.6% had ever been quarantined or isolated during the outbreak, 16.8% were frontline medical workers, 74.0% were highly concerned about the epidemic. Respectively 95.0%, 4.2% and 1.3% had confirmed cases in their living city, community and relatives and friends (Table 1 and Table 2).
Table1 Association between demographics and adverse psychological outcomes of
the COVID-19 outbreak (N=863)
|
Variables
|
N(%)
|
Impact of event
|
|
Depression
|
|
Anxiety
|
|
Stress
|
β
|
p
|
|
β
|
p
|
|
β
|
p
|
|
β
|
p
|
Gender
|
|
|
|
|
|
|
|
|
|
|
|
|
Male
|
253(29.3%)
|
-0.617
|
0.091
|
|
0.520
|
0.188
|
|
-0.135
|
0.714
|
|
0.169
|
0.730
|
Female
|
610(70.7%)
|
Reference
|
|
Reference
|
|
Reference
|
|
Reference
|
Age (Years)
|
|
|
|
|
|
|
|
|
|
|
|
|
≤29
|
277(32.1%)
|
0.233
|
0.754
|
|
-0.473
|
0.557
|
|
0.954
|
0.202
|
|
-1.089
|
0.274
|
30-39
|
390(45.2%)
|
-0.410
|
0.572
|
|
-0.683
|
0.385
|
|
0·414
|
0.517
|
|
-0.968
|
0.320
|
40-49
|
145(16.8%)
|
0.480
|
0.545
|
|
0.238
|
0.782
|
|
0.531
|
0.506
|
|
-0.008
|
0.994
|
≥50
|
51(5.9%)
|
Reference
|
|
Reference
|
|
Reference
|
|
Reference
|
Marital status
|
|
|
|
|
|
|
|
|
|
|
|
|
Currently not married
|
303(35.1%)
|
-0.230
|
0.508
|
|
-0.182
|
0.630
|
|
0.475
|
0.174
|
|
-0.397
|
0.395
|
Currently married
|
560(64.9%)
|
Reference
|
|
Reference
|
|
Reference
|
|
Reference
|
Education
|
|
|
|
|
|
|
|
|
|
|
|
|
Technical secondary or below
|
104(12.0%)
|
-0.198
|
0.722
|
|
0.170
|
0.778
|
|
0.431
|
0.441
|
|
0.084
|
0.910
|
College
|
466(54.0%)
|
0.372
|
0.306
|
|
-0.134
|
0.733
|
|
0.450
|
0.219
|
|
-0.206
|
0.672
|
Advanced degree
|
293(34.0%)
|
Reference
|
|
Reference
|
|
Reference
|
|
Reference
|
Monthly income
|
|
|
|
|
|
|
|
|
|
|
|
|
<6000 yuan
|
377(43.7%)
|
-0.117
|
0.782
|
|
-0.158
|
0.728
|
|
0.442
|
0.296
|
|
0.164
|
0.771
|
6000-9999 yuan
|
277(32.1%)
|
0.391
|
0.381
|
|
-0.445
|
0.358
|
|
0.066
|
0.883
|
|
-0.061
|
0.919
|
≥10000 yuan
|
209(24.2%)
|
Reference
|
|
Reference
|
|
Reference
|
|
Reference
|
Occupation
|
|
|
|
|
|
|
|
|
|
|
|
|
Doctor
|
377(43.7%)
|
0.607
|
0.114
|
|
0.597
|
0.154
|
|
-0.046
|
0.905
|
|
0.398
|
0.443
|
Nurse
|
211(24.4%)
|
1.520
|
0.001
|
|
0.699
|
0.148
|
|
1.283
|
0.004
|
|
0·957
|
0.110
|
Other health worker
|
275(31.9%)
|
Reference
|
|
Reference
|
|
Reference
|
|
Reference
|
Ever had chronic disease(s)
|
|
|
|
|
|
|
|
|
|
|
|
|
Yes
|
52(6.0%)
|
1.800
|
0.010
|
|
2.324
|
0.002
|
|
1.782
|
0.011
|
|
2.797
|
0.003
|
No
|
811(94.0%)
|
Reference
|
|
Reference
|
|
Reference
|
|
Reference
|
Current tobacco user
|
|
|
|
|
|
|
|
|
|
|
|
|
Yes
|
64(7.4%)
|
-1.426
|
0.024
|
|
-0.324
|
0.637
|
|
-0.733
|
0.250
|
|
-0.711
|
0.403
|
No
|
799(92.6%)
|
Reference
|
|
Reference
|
|
Reference
|
|
Reference
|
Current alcohol user
|
|
|
|
|
|
|
|
|
|
|
|
|
Yes
|
281(32.6%)
|
0.262
|
0.460
|
|
0.791
|
0.039
|
|
0.475
|
0.182
|
|
0.646
|
0.174
|
No
|
582(67.4%)
|
Reference
|
|
Reference
|
|
Reference
|
|
Reference
|
Table2 Association between variables related to the COVID-19 outbreak and
adverse psychological outcomes (N=863)
|
Variables
|
N(%)
|
Impact of event
|
|
Depression
|
|
Anxiety
|
|
Stress
|
β
|
p
|
|
β
|
p
|
|
β
|
p
|
|
β
|
p
|
Ever been quarantined or isolated
|
Yes
|
221(25.6%)
|
0.407
|
0.285
|
|
0.776
|
0.059
|
|
0.308
|
0.421
|
|
0.610
|
0.232
|
No
|
642(74.4%)
|
Reference
|
|
Reference
|
|
Reference
|
|
Reference
|
Duty during the epidemic
|
Front line workers
|
145(16.8%)
|
0.156
|
0.725
|
|
0.012
|
0.980
|
|
0.817
|
0.067
|
|
0.938
|
0.115
|
Second line or others
|
718(83.2%)
|
Reference
|
|
Reference
|
|
Reference
|
|
Reference
|
Levels of concern
|
High concern
|
639(74.0%)
|
1.704
|
0.000
|
|
0.218
|
0.595
|
|
0.163
|
0.669
|
|
1.232
|
0.015
|
Less concern
|
224(26.0%)
|
Reference
|
|
Reference
|
|
Reference
|
|
Reference
|
Confirmed cases in the living city
|
Yes
|
820(95.0%)
|
0.983
|
0.197
|
|
-0.014
|
0.986
|
|
0.147
|
0.848
|
|
1.206
|
0.239
|
No or not sure
|
43(5.0%)
|
Reference
|
|
Reference
|
|
Reference
|
|
Reference
|
Confirmed cases in the living community
|
Yes
|
36(4.2%)
|
0.249
|
0.764
|
|
2.082
|
0.020
|
|
2.854
|
0.001
|
|
2.542
|
0.022
|
No or not sure
|
827(95.8%)
|
Reference
|
|
Reference
|
|
Reference
|
|
Reference
|
Confirmed cases among relatives and friends
|
Yes
|
11(1.3%)
|
3.045
|
0.039
|
|
5.265
|
0.001
|
|
3.563
|
0.017
|
|
7.808
|
0.000
|
No
|
852(98.7%)
|
Reference
|
|
Reference
|
|
Reference
|
|
Reference
|
Factor loadings of perceived threat items and their psychometric properties
EFA on the 8 items of perceived threat yielded 3 factors (explaining 76.9% of the total variance; KMO=0.75), with satisfactory eigenvalue and factor loadings (eigenvalue>1 and factor loading>0.45) [23]. The three factors identified from the EFA assess respondents’ level of perceived threat with respect to stigmatization/distancing, fears of infection and high-risk job, with Chronbach’s alpha of 0.81, 0.78 and 0.79 (Table 3). The overall score of perceived threat and the extracted factors were calculated by summing up all item scores and were proved all significantly correlated with the scores of IES-6 and three subscales of DASS (Table 4).
Table 3 Factor loadings of perceived threat items among medical care workers (N=863)
|
Perceived threat (Items)
|
Factor1
|
Factor 2
|
Factor 3
|
Stigmatization
/distancing
|
Fears of infection
|
High-risk job
|
Afraid of being infected by COVID-19
|
-0.057
|
0.704
|
0.390
|
Anxious to be shifted to the ward for COVID-19
|
0.213
|
0.848
|
0.057
|
Worried about being quarantined or isolated
|
0.138
|
0.860
|
0.122
|
My job puts me at a high risk of being infected
|
0.119
|
0.297
|
0.820
|
My close contacts are at high risk of being infected
|
0.182
|
0.090
|
0.888
|
Friends and relatives are worried to be infected by me
|
0.627
|
0.110
|
0.498
|
I’m distanced by others due to my job
|
0.916
|
0.142
|
0.113
|
I’m stigmatized by others due to my job
|
0.901
|
0.087
|
0.079
|
Eigenvalue
|
2.158
|
2.091
|
1.899
|
Cumulative % of variance explained
|
26.971
|
53.111
|
76.845
|
Cronbach's alpha
|
0.812
|
0.780
|
0.793
|
Exploratory factor analysis, using principle component analysis for factor extraction(with varimax rotation).
|
Factor 1-3 addressed threat due to stigmatization/distancing, fears of infection and high-risk job.
|
Table 4 Pearson correlation between perceived threat, perceived social support,
coping style and adverse psychological outcomes (N=863)
|
|
|
Variables
|
Impact of event
|
|
Depression
|
|
Anxiety
|
|
Stress
|
|
β
|
p
|
|
β
|
p
|
|
β
|
p
|
|
β
|
p
|
|
Perceived threat
|
0.342
|
0.000
|
|
0.284
|
0.000
|
|
0.234
|
0.000
|
|
0.353
|
0.000
|
|
Stigmatization/distancing
|
0.517
|
0.000
|
|
0.674
|
0.000
|
|
0.548
|
0.000
|
|
0.761
|
0.000
|
|
Fears of infection
|
0.455
|
0.000
|
|
0.339
|
0.000
|
|
0.260
|
0.000
|
|
0.377
|
0.000
|
|
High-risk job
|
0.666
|
0.000
|
|
0.560
|
0.000
|
|
0.529
|
0.000
|
|
0.772
|
0.000
|
|
Perceived social support
|
-0.003
|
0.856
|
|
-0.109
|
0.000
|
|
-0.106
|
0.000
|
|
-0.130
|
0.000
|
|
Active coping (AC)
|
0.063
|
0.048
|
|
-0.127
|
0.000
|
|
-0.134
|
0.000
|
|
-0.172
|
0.000
|
|
Passive coping (PC)
|
0.234
|
0.000
|
|
0.309
|
0.000
|
|
0.247
|
0.000
|
|
0.318
|
0.000
|
|
Prevalence of psychological symptoms and perceived threat
IES-6 scale was used to measure the posttraumatic stress of COVID-19 outbreak, which revealed a sample mean score of 8.54 (SD=4.87) (data not tabulated). The IES-6 with a cutoff of 10 was used as a proximate measure of PTS, which is considered to have the best overall efficiency [12], and 347 (40.2%) were considered to meet the clinical concern of PTSD. 97.9% of the respondents had one or more PTS symptoms. The most severe PTS domain among the respondents was intrusion: “I thought about it when I did not mean to”(93.7%) and “Other things kept making me think about it”(76.4%) , then the hyperarousal and avoidance domain (Figure 1 and Figure 2). Figure 1 presents DASS severity ratings of the respondents. The proportion of having mild to extremely severe symptoms of depression, anxiety and stress were 13.6%, 13.9% and 8.6%, respectively.
In total, ‘Fears of infection’ and ‘Doing high-risk job’ were the mostly perceived threat by the participants. While 525 participants (60.8%) reported ‘I am afraid of being infected by COVID-19’, only 159 (18.4%) responded that ‘I’m anxious to be shifted to the ward for COVID-19 patients. In addition, 420 (48.7%) agreed ‘My close contacts are at higher risk of being infected due to my job’, and 417 participants (48.3%) agreed ‘My job puts me at a high risk of being infected by COVID-19’. 78.5% of the participants reported have at least one out of the 8 perceived threat items (Figure 3)
Associations between background variables and adverse psychological outcomes
Table 1 and table 2 presented the bivariate correlates of having PTS and mild to extremely severe symptoms of depression, anxiety and stress among the study participants. Those who were nurses (β=1.52, p=0.001), ever had chronic diseases (β=1.80, p=0.010) , had high concern to the outbreak (β=1.70, p=0.000), and had confirmed cases among their relatives and friends (β=3.05, p=0.039) were more likely to have PTS symptoms. Those who were current tobacco user were less likely to have PTS symptoms (β=-1.43, p=0.024)
Those who ever had chronic diseases (β=2.32, p=0.002), who were current alcohol user (β=0.79, p=0.039), had confirmed cases in their living community (β=2.08, p=0.020) and had confirmed cases among their relatives and friends (β=5.27, p=0.001) were more likely to have depression symptoms. Those who were nurses (β=1.28, p=0.004), ever had chronic diseases (β=1.78, p=0.011) and had confirmed cases in their living community (β=2.85, p=0.001) and had confirmed cases among their relatives and friends (β=3.56, p=0.017) were more likely to have anxiety symptoms. Those who ever had chronic diseases (β=2.80, p=0.003), had high concern to the outbreak (β=1.23, p=0.015), had confirmed cases in their living community (β=2.54, p=0.022) and had confirmed cases among their relatives and friends (β=7.81, p=0.000) were more likely to have stress symptoms.
Pearson correlation between adverse psychological symptoms and perceived threat, perceived social support and coping style
The IES-6 and DASS score were correlated with almost all proposed psychosocial variables including overall score and scores of the three domains of perceived threat, perceived social support, and two forms of coping strategies in anticipated directions. The adverse psychological symptoms were positively associated with perceived threat and passive coping strategies, while negatively associated with perceived social support and active coping strategies, except the correlation of IES-6 with perceived social support ((β=-0.00, p=0.856) and with active coping (β=0.06, p=0.048) (Table 4).
Multivariate correlates of psychological symptoms
Table 5 presented the multivariate correlates of PTS and DASS symptoms. Compare to other groups, nurses were more likely to have anxiety symptoms (β=0.93, p=0.026). Participants who were current tobacco user were less likely to have PTS symptoms (β= -1.55, p=0.015). Level of concern to the outbreak was positively correlated with PTS (β=1.49, p=0.000) and stress symptoms (β=1.22, p=0.009). Those who had confirmed cases in their living community were more likely to be anxious (β=1.82, p=0.018), and who have confirmed cases among relatives and friends were more likely to have depression (β=3.70, p=0.011) and stress symptoms (β=5.75, p=0.002). Apart from the ‘Fears of infection’, other two dimension of perceived threat were positively associated with the PTS and DASS (value of β and p see Table 5), those with more fears of infection were more likely to have PTS (β=0.25, p=0.000). Three dimensions of Perceived threat and passive coping strategies were positively related to both PTS (β=0.28, 0.25, 0.29 and 0.31, p=0.000) and DASS symptoms (value of β and p see Table 5) . Perceived social support and active coping were negatively related to DASS symptoms (value of β and p see Table 5) . Those adopted passive coping strategies were more likely to have PTS and DASS.
Table 5 Results of multiple linear regression analysis on IES-6 and DASS(N=863)
|
|
Variables
|
Impact of event
|
|
Depression
|
|
Anxiety
|
|
Stress
|
|
β
|
p
|
|
β
|
p
|
|
β
|
p
|
|
β
|
p
|
|
Gender
|
|
Male
|
-0.140
|
0.721
|
|
-
|
|
-
|
|
-
|
|
Female
|
Reference
|
|
|
|
|
Occupation
|
|
Doctor
|
-0.206
|
0.565
|
|
-
|
|
-0.473
|
0.187
|
|
-
|
|
Nurse
|
0.549
|
0.201
|
|
|
0.930
|
0.026
|
|
|
Other health worker
|
Reference
|
|
|
Reference
|
|
|
Ever had chronic disease(s)
|
|
Yes
|
1.174
|
0.068
|
|
1.198
|
0.080
|
|
1.067
|
0.095
|
|
1.357
|
0.111
|
|
No
|
Reference
|
|
Reference
|
|
Reference
|
|
Reference
|
|
Current tobacco user
|
|
Yes
|
-1.549
|
0.015
|
|
-
|
|
-
|
|
-
|
|
No
|
Reference
|
|
|
|
|
Current alcohol user
|
|
Yes
|
-
|
|
0.344
|
0.323
|
|
-
|
|
-
|
|
No
|
|
Reference
|
|
|
|
Ever been quarantined or isolated
|
|
Yes
|
-
|
|
0.588
|
0.111
|
|
-
|
|
-
|
|
No
|
|
Reference
|
|
|
|
Levels of concern
|
|
Highly concerned
|
1.488
|
0.000
|
|
-
|
|
-
|
|
1.224
|
0.009
|
|
Less concerned
|
Reference
|
|
|
|
Reference
|
|
Confirmed cases in the living community
|
|
Yes
|
-
|
|
0.639
|
0.436
|
|
1.822
|
0.018
|
|
0.666
|
0.515
|
|
No or not sure
|
|
Reference
|
|
Reference
|
|
Reference
|
|
Confirmed cases among relatives and friends
|
|
Yes
|
2.071
|
0.125
|
|
3.701
|
0.011
|
|
1.792
|
0.190
|
|
5.747
|
0.002
|
|
No
|
Reference
|
|
Reference
|
|
Reference
|
|
Reference
|
|
Perceived threat
|
|
Stigmatization/distancing
|
0.282
|
0.000
|
|
0.416
|
0.000
|
|
0.306
|
0.000
|
|
0.433
|
0.000
|
|
Fears of infection
|
0.251
|
0.000
|
|
0.099
|
0.116
|
|
0.044
|
0.448
|
|
0.052
|
0.495
|
|
High-risk job
|
0.285
|
0.001
|
|
0.231
|
0.010
|
|
0.272
|
0.001
|
|
0.428
|
0.000
|
|
Perceived social support
|
-
|
|
-0.064
|
0.000
|
|
-0.072
|
0.000
|
|
-0.083
|
0.000
|
|
Active coping (AC)
|
0.031
|
0.303
|
|
-0.106
|
0.003
|
|
-0.106
|
0.001
|
|
-0.152
|
0.001
|
|
Passive coping (PC)
|
0.172
|
0.000
|
|
0.274
|
0.000
|
|
0.220
|
0.000
|
|
0.286
|
0.000
|
|
# Impact of event: R-Squared (R2) = 0.193 , Adjusted R-Squared (AR2) = 0.182, p=0.000; Depression: R2 = 0.217, AR2 =0.206, p=0.000; Anxiety: R2 =0.199 , AR2 =0.189, p=0.000; Stress: R2 =0.200 , AR2 =0.191, p=0.000.
|
|
|