A total of 124 out of 143 participants returned the questionnaires with a response rate of 86.7%. More than two-thirds of the participants belonged to the 25–34 years age group, and there was a slight male preponderance (57.2% versus 42.8%).
The majority of the respondents were between the ages of 20 and 34 years (n = 88 [71.0%]) and married (n = 79 [63.7%]). Most of the respondents were nurses (n = 117 [94.4%]), and 79.8% (n = 99) had previous ICU experience. The majority of participants (n = 103 [83.1%]) were assigned to the ICU due to the COVID-19 pandemic. Only 4.0% (n = 5) of participants had previous mental health concerns, and 7.3% (n = 9) had a family history of mental illness. Nearly half of the respondents (n = 58 [46.8%]) endorsed secondary stressors such as financial problems. The sociodemographic characteristics of the respondents are presented in Table 1.
3.1. Perceived stress among ICU staff
Although 25.8% (n = 32) perceived working in the ICU with COVID-19 patients as a low-stress activity, 67.7% (n = 84) and 6.5% (n = 8), perceived it as moderate and high stress activity, respectively. Moderate and severe perceived stress responses were grouped together (74.2%; n = 92) to assess their possible association with sociodemographic variables.
3.2. Logistic regression of sociodemographic variables and moderate to high perceived stress
Three-quarters of the respondents (n = 92 [74.2%]) perceived working in the ICU during the COVID-19 pandemic as moderate-to-high-stress activity. A total of 78.4% (n = 69) of the 92 respondents in the 20‒34-year age group experienced moderate to severe psychological distress, and this proportion was 63.9% (n = 23) among respondents in the 35‒64-year age group (p = 0.093). Overall, 70.4% (n = 50) of men and 79.2% (n = 42) of women experienced moderate to severe perceived stress (p = 0.267). Approximately 80.0% of the respondents (n = 36) who were single and 70.9% (n = 56) who were married experienced moderate to severe perceived stress (p = 0.265).
In addition, 71.4% (n = 5) of doctors and 74.4% (n = 87) of nurses experienced moderate to severe perceived stress (p = 0.863). Eighty percent (n = 20) of staff without previous ICU experience and 72.7% (n = 72) of staff with previous ICU experience reported moderate to high perceived stress (p = 0.458). Three-quarters of the respondents that were newly assigned to the ICU due to the COVID-19 pandemic (n = 78 [75.7%]) and 66.7% of staff who were already working in the ICU (n = 14) experienced moderate to high perceived stress (p = 0.387).
All staff with a history of mental health concerns (n = 5) reported moderate to severe stress when working in the ICU, while such levels of stress were reported by 73.1% (n = 87) of those who did not have a history of mental health difficulties (p = 0.178). Moreover, 77.8% (n = 7) of respondents with a family history of mental illness reported moderate to severe perceived stress, while such high levels were reported by 73.9% (n = 85) of those without a family history of these illnesses (p = 0.799). Furthermore, 86.2% (n = 50) of respondents who had secondary stressors, such as financial problems, experienced moderate to high stress levels, and this proportion was only 63.6% (n = 42) among those who did not have any secondary stressors (p = 0.004).
Univariate logistic regression was performed to assess the effect of sociodemographic variables and other clinical characteristics on the outcome variable of the moderate to high perceived stress group, as presented in Table 2. In the multivariate logistic regression, age and secondary stressors such as financial problems were found to remain significantly associated with moderate to severe perceived stress. Multivariate logistic regression indicated that respondents in the 20–34-year age group were more likely to experience moderate to severe perceived stress compared to the 35–64-year age group (adjusted OR: 3.72; 95% CI: 1.10–12.60; p = 0.035). Moreover, respondents who had secondary stressors such as financial difficulties were more likely to experience moderate to severe perceived stress compared with those who did not have these stressors (adjusted OR: 4.49; 95% CI: 1.67–12.12; p = 0.003).
Table 2. Effect of sociodemographic variables on moderate to high perceived stress
Sociodemographic variables
|
PSS ≥ 14; n = 92
n (%)
|
Unadjusted odds ratio
|
P value
|
Age
20-34 years
35-64 years
|
69 (78.4)
|
2.05 (0.88–4.80)
|
0.097
|
23 (63.9)
|
1
|
Sex
Male
Female
|
50 (70.4)
|
1
|
0.269
|
42 (79.2)
|
1.60 (0.69–3.70)
|
Marital status
Single
Married
|
36 (80.0)
|
1.64 (0.68–3.95)
|
0.267
|
56 (70.9)
|
1
|
Occupation
Doctor
Nurse
|
5 (71.4)
|
0.86 (0.16–4.68)
|
0.863
|
87 (74.4)
|
1
|
Number of years working in HMC
< 1 year
1-3 years
4-6 years
> 6 years
|
15 (78.9)
|
1.17(0.26–5.21)
|
0.835
|
48 (72.7)
|
0.83 (0.27–2.61)
|
0.754
|
13 (72.2)
|
0.81 (0.19–3.43)
|
0.777
|
16 (76.2)
|
1
|
|
Previous ICU experience
Yes
No
|
72 (72.7)
|
0.67 (0.23–1.95)
|
0.460
|
20 (80.0)
|
1
|
Posted to the ICU due to COVID-19
Yes
No
|
78 (75.7)
|
1.56 (0.57–4.30)
|
0.390
|
14 (66.7)
|
1
|
Family history of mental illness
Yes
No
|
7 (77.8)
|
1.24 (0.24–6.28)
|
0.799
|
85 (73.9)
|
1
|
Secondary stressors such as financial problems
Yes
No
|
50 (86.2)
|
3.57 (1.45–8.78)
|
0.006
|
42 (63.6)
|
1
|
Abbreviations: COVID-19 = 2019 coronavirus; HMC = Hamad Medical Corporation; ICU = intensive care unit; PSS = Perceived Stress Scale.
Although not statistically significant, variables such as previous ICU experience and assignment to the ICU due to COVID-19 were associated with moderate to high perceived stress. Respondents who were newly posted to the ICU were more likely to experience moderate to high stress compared with those who were already working in the ICU (adjusted OR: 1.41; 95% CI: 0.41–4.83; p = 0.589). Moreover, the staff with previous ICU experience were less likely to experience moderate to high stress than those without previous ICU experience (adjusted OR: 0.46; 95% CI: 0.13–1.70; p = 0.243).
3.2.1. PTSD symptoms among ICU staff
According to the PDS-5, a score of 28 was used as a cut off for the possible diagnosis of PTSD, with scores between 0–27 indicating no diagnosis, and 28–80 indicating probable diagnosis. One hundred and two respondents (82.3%) scored less than 28, indicating no diagnosis; however, 17.7% (n = 22) scored more than 28, indicating a probable diagnosis of PTSD.
3.3. Logistic regression of sociodemographic variables and probable diagnosis of PTSD
A total of 17.0% (n = 15) and 19.4% (n = 7) of respondents in the 20‒34-year and 35‒64-year age groups, respectively, had a result indicative of probable PTSD diagnosis (p = 0.751).
Overall, 16.9% (n = 12) of men and 18.9% (n = 10) of women had probable diagnosis of PTSD (p = 0.777). Approximately 13.3% of the respondents (n = 6) who were single and 20.3% (n = 16) who were married had a probable diagnosis of PTSD (p = 0.332). Finally, 28.6% (n = 2) of doctors and 17.1% (n = 20) of nurses had a probable diagnosis of PTSD (p = 0.440).
Furthermore, 24.0% (n = 6) of staff without previous ICU experience had a probable diagnosis of PTSD, and the proportion was 16.2% (n = 16) among staff with ICU experience (p = 0.359). Nineteen percent (n = 4) of respondents who were recently relocated to the ICU due to the COVID-19 pandemic and 17.5% (n = 18) of those who were already working in the ICU had a probable diagnosis of PTSD (p = 0.864).
Lastly, 20.0% of staff with previous mental health difficulties (n = 1) and 17.6% (n = 21) of those who did not have a history of mental health issues had a probable diagnosis of PTSD (p = 0.893). Moreover, 25.9% (n = 15) of respondents who had secondary stressors such as financial problems experienced moderate to high stress, and this proportion was only 10.6% (n = 7) among those who did not have any secondary stressors (p = 0.026). Univariate logistic regression of sociodemographic variables and patients with probable diagnosis of PTSD is presented in Table 3.
Table 3. Effect of sociodemographic variables on the probable diagnosis of post-traumatic stress disorder
Sociodemographic variables
|
PDS-5 ≥ 28; n = 22
n (%)
|
Unadjusted odds ratio
|
P value
|
Age
20–34 years
35–64 years
|
15 (17.0)
|
0.85 (0.32–2.30)
|
0.751
|
7 (19.4)
|
1
|
Sex
Male
Female
|
12 (16.9)
|
1
|
0.777
|
10 (18.9)
|
1.14 (0.45–2.89)
|
Marital status
Single
Married
|
6 (13.3)
|
0.61 (0.22–1.68)
|
0.335
|
16 (20.3)
|
1
|
Occupation
Doctor
Nurse
|
2 (28.6)
|
1.94 (0.35–10.72)
|
0.447
|
20 (17.1)
|
1
|
Number of years working in HMC
< 1 year
1–3 years
4–6 years
> 6 years
|
1 (5.3)
|
0.24 (0.03–2.33)
|
0.217
|
14 (21.2)
|
1.14 (0.33–3.95)
|
0.831
|
3 (16.7)
|
0.85 (0.16–4.43)
|
0.847
|
4 (19.0)
|
1
|
|
Previous ICU experience
Yes
No
|
16 (16.2)
|
0.61 (0.21–1.77)
|
0.363
|
6 (24.0)
|
1
|
Posted to the ICU due to COVID-19
Yes
No
|
18 (17.5)
|
0.90 (0.27–2.99)
|
0.864
|
4 (19.0)
|
1
|
Previous mental health issues
Yes
No
|
1 (20.0)
|
1.17 (0.12–10.97)
|
0.893
|
21 (17.6)
|
1
|
Family history of mental illness
Yes
No
|
1 (11.1)
|
0.56 (0.07–4.71)
|
0.593
|
21 (18.3)
|
1
|
Secondary stressors such as financial problems
Yes
No
|
15 (25.9)
|
2.94 (1.10–7.83)
|
0.031
|
7 (10.6)
|
1
|
Abbreviations: COVID-19 = 2019 coronavirus; HMC = Hamad Medical Corporation; ICU = intensive care unit; PDS-5 = PTSD Diagnostic Scale for DSM-5.
3.4. Logistic regression of perceived stress and probable diagnosis of PTSD
Logistic regression indicated that respondents who scored moderate to high on the PSS were more likely to have a probable diagnosis of PTSD compared with those who scored low (OR: 12.83; 95% CI: 1.69‒97.36; p = 0.014). This association was more notable in participants who perceived working in the ICU during the COVID-19 pandemic as a high-stress activity than among those who perceived it as a source of low to moderate stress (OR: 26.28; 95% CI: 5.22‒132.30; p < 0.0001).