1. Study Population Characteristics
A total of 1600 questionnaires were distributed and received 1493 questionnaires yielded valid data in this survey. Demographic information was presented in Table 1.
Table 1: Baseline characteristics of interviewees
Characteristics
|
Sample size
|
No (%)
|
Age, Range, y
|
|
|
<18
|
13
|
0.87
|
18-25
|
572
|
38.31
|
26-35
|
467
|
31.28
|
36-45
|
296
|
19.83
|
46-55
|
119
|
7.97
|
56-65
|
25
|
1.67
|
>66
|
1
|
0.07
|
Gender
|
|
|
Male
|
827
|
55.39
|
Female
|
666
|
44.61
|
Education
|
|
|
Junior middle school or below
|
41
|
2.75
|
Technical secondary school/High school/Vocational high school
|
288
|
19.29
|
Junior college
|
841
|
56.33
|
Master's/Doctor's degree
|
323
|
21.63
|
Occupation
|
|
|
Medical staff
|
422
|
28.27
|
No-medical staff
|
1071
|
71.73
|
If there are any suspected or confirmed COVID-19 cases
|
|
|
Yes
|
327
|
21.90
|
No
|
1166
|
78.10
|
History of mental illness
|
|
|
Yes
|
8
|
0.54
|
No
|
1485
|
99.46
|
Still working during the epidemic
|
|
|
Yes
|
866
|
58.00
|
No
|
627
|
42.00
|
Participate in the frontline response
|
|
|
Yes
|
301
|
20.16
|
No
|
1192
|
79.84
|
2. The relationship between basic information and psychological changes
To determine the relationship between demographic characteristics and scores on the IES-R, PSS–10 and related factors of the SDS score, we conducted linear regression analyses. Specifically, we examined the demographic variables of sex, age, education, history of mental illness, the presence of suspected or confirmed COVID–19 cases, occupation, whether the interviewee is still working during the epidemic, and whether the interviewee participated in the frontline response. The regression coefficients were shown in Table 2.
Table 2: Linear regression analysis of baseline characteristics and questionnaire scores
|
IES-R
|
PSS-10
|
SDS
|
|
B (Std. Error) a
|
Beta b
|
B (Std. Error)
|
Beta
|
B (Std. Error)
|
Beta
|
Sex
|
4.35 (6.39)**
|
0.17
|
0.76 (2.57)*
|
0.07
|
3.18 (4.98)**
|
0.14
|
Age
|
1.56 (4.71)**
|
0.13
|
0.98 (6.82)**
|
0.19
|
-1.03 (-3.31)**
|
-0.09
|
Education
|
0.79 (1.64)
|
0.04
|
1.18 (5.65)**
|
0.16
|
-1.93 (-4.25)**
|
-0.12
|
History of mental illness
|
17.36 (3.88)**
|
0.10
|
4.38 (2.26)*
|
0.06
|
18.96 (4.52)**
|
0.118
|
Suspected/confirmed cases around
|
-4.28 (-4.61)**
|
-0.14
|
-1.38 (-3.42)**
|
-0.10
|
-4.38 (-5.04)**
|
-0.08
|
Occupation
|
-0.04 (-0.54)
|
-0.02
|
-0.04 (-1.07)
|
-0.03
|
0.02 (0.31)
|
0.01
|
Still working during epidemic
|
3.36 (4.51)**
|
0.13
|
0.49 (1.53)
|
0.04
|
2.36 (3.38)**
|
0.10
|
Participate in the frontline response
|
-4.15 (-4.32)**
|
-0.13
|
-0.33 (-0.80)
|
-0.02
|
-2.95 (-3.27)**
|
-0.10
|
R²
|
0.19
|
0.15
|
0.13
|
F-value
|
13.54**
|
10.04**
|
10.04**
|
*P < 0.05, ** P < 0.01
a Unstandardized Coefficients (Standard Errors)
b Standardized Coefficients
The relationship between related factors and the three scales was shown in Fig. 1. In brief, gender and history of mental illness were positively correlated to IES-R, PSS–10 and SDS scores. There was a negative correlation between IES-R, PSS–10 and SDS scores in the presence of closing to COVID–19 patients or suspected patients. Age was positively correlated with IES-R and PSS–10, but negatively correlated with SDS. Education had no effect on IES-R score, which was positively correlated with PSS–10 score, while negatively correlated with SDS score. Working during the epidemic showed a positive correlation between IES-R and SDS score, and a statistical difference with PSS–10 score. There was a negative correlation between participate in the frontline response. However, there was no statistical difference on PSS–10. Other factors showed no statistical difference in the scores of IES-R, PSS–10, and SDS.
3. Differences between questionnaire scores of medical staff and the general public
The Wilcoxon rank sum test was conducted in order to determine difference in psychological scores between medical staff and members of the general public. There was no significant difference in SDS score between medical staff 44.52(12.36) and the general public 43.08(11.42), P = 0.08. The PSS–10 score of medical staff 16.81(4.87) was higher than that of general public 14.80(5.60), P < 0.01. Meanwhile, the IES-R score of medical staff 22.40(12.12) was also higher than that of general public 17.89(13.08), P < 0.01. The results were presented in Table 3.
Table 3: Wilcoxon rank sum test results for questionnaire score differences between medical staff and the general public
|
Medical staff, mean (SD)
|
General public, mean (SD)
|
P-value
|
SDS
|
44.52 (12.36)
|
43.08 (11.42)
|
0.08
|
PSS-10
|
16.81 (4.87)
|
14.80 (5.60)
|
< 0.01
|
IES-R
|
22.40 (12.12)
|
17.89 (13.08)
|
< 0.01
|
Abbreviation: SDS: Self-rating Depression Scale; PSS-10, Perceived Stress Scale; IES-R, Impact of Event Scale-Revised scales.
4. Attitudes and needs of medical staff and the general public towards psychological counseling during the epidemic
Of the 1493 respondents, only 428 (28.67%) had access to psychological training and 761 (50.97%) needed psychological counseling. The relative risk ratio (RR) was defined as 1. In this study, it was found that medical staff (40 [9.47%]) had access to psychological training only 0.27 (95%CI 0.23 to 0.31; P < 0.01) times that of the general public (388 [36.23%]), while the demand for psychological counseling of medical staff (278 [65.87%]) was 1.46 (95%CI 1.33 to 1.61; P < 0.01) times that of the general public (483 [45.10%]). The results were presented in Table 4.
Table 4: Comparison of the needs of medical staff and the general public for psychological counseling
|
N (%)
|
Medical staff (n=422), (%)
|
General public (n=1071), (%)
|
RR (95%CI)
|
P-value
|
Access to psychological training
|
428 (28.67)
|
40 (9.47)
|
388 (36.23)
|
0.27 (0.23~0.31)
|
< 0.01
|
Demand for psychological counseling
|
761 (50.97)
|
278 (65.87)
|
483 (45.10)
|
1.46 (1.33~1.61)
|
< 0.01
|