Participant characteristics
The survey was sent to 134 EPs, of which 121 completed the survey, but 5 surveys were considered incomplete. As such, only 116 responses were considered for analysis, demonstrating a good response rate of 86.7%. The demographic and occupational characteristics of the participants are summarized in Table 1. The majority of participants (42.2%) were in the age group of 35-44 years old followed by 30.2% who were 25-34 years old. Eighty-one participants (69.8%) were male. Most participants were based in SMC (46.6%), followed by KHUH and BDF (25.9% and 23.3% respectively), where 5 participants did not state their affiliation. In terms of marital status, 81% were married, 14.7% were single, and 4.3% were widowed or divorced. Twenty-four (20.7%) participants were residents, 49 (42.2%) were senior residents, 30 (25.9%) were chief residents, and 13 (11.2%) were Consultants. Of the 81 participants who provided a response on their residency year, 23.3% had completed their residency within the past 6 months, 22.4% were between their first and third year of residency, and 24.1% were in their fourth year. There was an equal distribution among physicians among the categories for years of experience, where 18.1% of EPs had 1 to 4 years’ experience, 28.4% had 5 to 9 years, 21.6% had 10 to 14 years, 14.7% had 15-19 years, 11.2% had 20 to 24 years, and 6.0% had 25 to 39 years. Regarding the hours of work per week, 45 (38.8%) EPs worked 40 hours, whereas 61.2% were working for 50 hours or more. Sixty-one physicians (52.6%) had taken at least 1-day sick leave in past year. Sleep disturbance occurred among 102 (87.9%) EPs, and 74 (63.8%) had been exposed to violence in their workplace at least once in the last year, of which 43 (37.1%) had been exposed to violence 1-5 times in their workplace last year.
Table 1. Socio-demographic and occupational characteristics of Bahrain’s Emergency Physicians (n= 116)
Characteristics
|
Frequency
|
Percentage
|
Total (n)
|
1. Age
|
|
|
|
25-34
|
35
|
30.2%
|
116
|
35-44
|
49
|
42.2%
|
|
45-54
|
25
|
21.6%
|
|
55-64
|
5
|
4.3%
|
|
≥65
|
2
|
0.2%
|
|
2. Sex
|
|
|
|
Male
|
81
|
69.8%
|
115
|
Female
|
34
|
29.3%
|
|
3. Hospital Affiliation
|
|
|
|
Bahrain Defence Force
|
27
|
23.3%
|
111
|
King Hamad University Hospital
|
30
|
25.9%
|
|
Salmaniya Medical Centre
|
54
|
46.6%
|
|
4. Marital Status
|
|
|
|
Single
|
17
|
14.7%
|
116
|
Married
|
94
|
81.0%
|
|
Divorced
|
4
|
3.4%
|
|
Widowed
|
1
|
0.9%
|
|
5. Nationality
|
|
|
|
Bahraini
|
68
|
58.6%
|
116
|
Non-Bahraini
|
48
|
41.4%
|
|
6. Number of Children
|
|
|
|
0
|
28
|
24.1%
|
116
|
1
|
13
|
11.2%
|
|
2
|
39
|
33.6%
|
|
3
|
20
|
17.2%
|
|
4
|
14
|
12.1%
|
|
5
|
1
|
0.9%
|
|
>5
|
1
|
0.9%
|
|
7. Job Position
|
|
|
|
Resident
|
24
|
20.7%
|
116
|
Senior Resident
|
49
|
42.2%
|
|
Chief Resident
|
30
|
25.9%
|
|
Consultant
|
13
|
11.2%
|
|
8. Residency Year
|
|
|
|
Completed within last six months
|
27
|
23.3%
|
81
|
Year 1
|
9
|
7.8%
|
|
Year 2
|
7
|
6.0%
|
|
Year 3
|
10
|
8.6%
|
|
Year 4
|
28
|
24.1%
|
|
9. Years of experience
|
|
|
|
1-4
|
21
|
18.1%
|
116
|
5-9
|
33
|
28.4%
|
|
10-14
|
25
|
21.6%
|
|
15-19
|
17
|
14.7%
|
|
20-24
|
13
|
11.2%
|
|
25-39
|
7
|
6.0%
|
|
10. Hours of Work Per Week
~ 40 hours
~ 50 hours
~ 60 hours
~ 70 hours
~ 80 hours
|
45
49
13
6
3
|
38.8%
42.2%
11.2%
5.2%
2.6%
|
116
|
11. Have you taken at least 1 sick leave during the last year?
Yes
No
|
61
55
|
52.6%
47.4%
|
116
|
12. Have you personally dealt with sleep disturbances?
Yes
No
|
102
14
|
87.9%
12.1%
|
116
|
13. Do you tend to take short meals, staggered, or not eat all?
Yes
No
|
95
21
|
81.9%
18.1%
|
116
|
14. Have you been exposed to violence in the workplace at least once in the last year?
Yes
No
|
74
42
|
63.8%
36.2%
|
116
|
15. If you answered yes to Q14, number of times of exposure to violence?
1
1 ≤ 5
6 ≤ 10
|
9
43
22
|
7.7%
37.1%
18.9%
|
74
|
Prevalence and distribution of burnout and stress rates among EPs
Tables 2 and 3 provide descriptive statistics of scores obtained and the prevalence of each type of CBI. The prevalence of burnout among EPs was 81.0% for personal burnout (M: 63.0, SD: 22.4; CI: 67.0% - 83.3%), 69.8% (M: 60.3, SD: 21.6; CI: 60.6%- 78.0%) for work-related burnout, and 40.5% (M: 43.1, SD: 25.4; CI: 31.5% - 50.0%) for patient-related burnout. The data for the SOS-S subscales and its related items for the EPs are displayed in Tables 2 and 3, with personal vulnerability (M: 13.89, SD: 5.04), event load (M: 16.73, SD: 4.80), and full SOS-S (M: 30.57, SD: 9.37).
Table 3 combined the descriptive data from both the CBI and the SOS-S using a mean score of 50 as the cut-off for the presence of burnout for each scale based on previously validated studies.44,45 While the SOS-S does not have cut-off scores to determine whether participants are stressed, it categorizes participants based on categorical scoring within four quadrants of risk: High Risk (high PV, high EL), Challenged (low PV, high EL), Fragile (high PV, low EL), and Low Risk (low PV, low EL). In Figure 1, EPs categorical SOS-S scores were determined by splitting each SOS-S subscale at its mean (PV-S mean = 13.89; EL-S mean = 16.73) and crossing the scales to form a 2 x 2 matrix. Out of 113 responses, results yielded 41.6% (47) of participants as Low Risk (Low PV, Low EL), 0.9% (1) as Fragile (High PV, low EL), 33.6% (38) as Challenged (low PV, high EL), and 23.9% (27) as High Risk for illness (High PV, high EL), and all the figures are highlighted in Figure 1.
Table 3. Prevalence and distribution of burnout and stress overload among Emergency Physicians
Type of Burnout
|
Mean Score (SD)
|
Prevalence (%, Confidence Interval)
|
Personal burnout
|
63.0 (22.4)
|
94 (81.0%, CI: 67.0% - 83.3%)
|
Work-related burnout
|
60.3 (21.6)
|
81 (69.8%, CI: 60.6%- 78.0%)
|
Patient-related burnout
|
43.1 (25.4)
|
47 (40.5%, CI: 31.5% - 50.0%)
|
SOS-Short (SOS-S)
|
Mean Score (SD)
|
|
Total Personal Vulnerability (PV-S)
|
13.9 (5.0)
|
-
|
Total Event Load (EV-S)
|
16.7 (4.8)
|
-
|
Total SOS-S
|
30.6 (9.4)
|
-
|
Association between burnout, stress, and demographic and occupational variables
As per Table 4, Pearson correlations showed a strong and statistically significant correlation between the CBI and SOS-S across all their subscales with Pearson coefficients ranging from 0.595 to 0.830 between the SOS-S subscales and the CBI dimensions (p < 0.0001), illustrating that higher personal, work-related, and patient-related burnout were all positively correlated with personal vulnerability, event load, and stress overload.
Table 4: Pearson Correlations between the CBI dimensions and the SOS-S total scores and its subscales*
|
Personal Burnout
|
Work-related Burnout
|
Patient-related Burnout
|
Personal Vulnerability
(r)
|
Event Load (r)
|
SOS-S Total (r)
|
Personal Burnout
|
-
|
0.830
|
0.700
|
0.616
|
0.622
|
0.650
|
Work-related Burnout
|
0.830
|
-
|
0.732
|
0.659
|
0.675
|
0.699
|
Patient-related Burnout
|
0.700
|
0.732
|
-
|
0.613
|
0.595
|
0.638
|
Personal Vulnerability
|
0.616
|
0.659
|
0.613
|
-
|
0. 811
|
|
Event Load
|
0.622
|
0.675
|
0.595
|
0. 811
|
-
|
0.949
|
SOS-S Total
|
0.650
|
0.699
|
0.638
|
0.954
|
0.949
|
-
|
*All Pearson-correlation values show a p-value of <0.0001.
Table 5 analyses the differences in the CBI dimensions and SOS-S categorical scores (High Risk, Challenged, Fragile, and Low Risk groups) according to the demographic and occupational variables. A higher number of Bahraini nationals were experiencing personal and work-related burnout (p= 0.005). The 59% of physicians who took sick leave were also experiencing patient-related burnout (p=0.005). A significant number of physicians experiencing personal burnout, work-related and patient-related burnout had sleep disturbances. More physicians experiencing personal burnout have experienced violence at the workplace (82.4%, p= 0.028). There were also significant differences between the four SOS-S quadrants based on the number of working hours, sick leave status, and violence at workplace status of the physicians (p = 0.020, p =0.004, p= 0.048 and 0.009, respectively). Those who were more likely to work 50 hours or more, take sick leave, and/or suffered from violence at the workplace were more likely to be in the High-Risk group.
Table 5: Chi-square test for CBI and SOS-S domains based on demographic / occupational variables
|
CBI Domains
|
SOS-S Categories (Risk for Illness)
|
|
Personal Burnout
(score ≥ 50)
N (%),
p-value
|
Work-related burnout
(score ≥ 50)
N (%),
p-value
|
Patient-related burnout (score ≥ 50)
N (%),
p-value
|
High risk
N (%)
|
Challenged
N (%)
|
Fragile
N (%)
|
Low Risk
N (%)
|
SOS Categori-cal Scores
(p-value)
|
Total
|
94 (81.0 %)
|
81 (69.8%)
|
47 (40.5%)
|
27 (23.5 %)
|
38 (33.0%)
|
1 (0.9%)
|
46 (40.0%)
|
|
Gender
Male
Female
|
64 (79.0 %)
30 (88.2 %)
p = 0.034
|
53 (65.4 %)
28 (82.4 %)
p = 0.061
|
35 (43.2 %)
12 (35.3 %)
p = 0.520
|
17 (21.0 %)
10 (29.4 %)
|
27 (33.3%)
11 (32.4%)
|
0 (0.0%)
1( 2.9%)
|
36 (44.4 %)
10 (29.4 %)
|
0.168
|
Age
25-34
35-44
45-54
55-64
≥65
|
29 (82.9%)
33 (67.3%)
22 (88.0%)
3 (60.0%)
1 (50.0%)
p= 0.178
|
25 (71.4%)
35 (71.4%)
18 (72.0%)
2 (40.0%)
1 (50.0%)
p= 0.619
|
12 (34.3%)
20 (40.8%)
12 (48.0%)
2 (40.0%)
1 (50.0%)
p= 0.875
|
9 (25.7%)
11 (22.4%)
7 (28.0%)
0 (0%)
0 (0%)
|
16 (45.7%)
13 (26.5%)
8 (32.0%)
0 (0%)
1 (50.0%)
|
0 (0.0%)
0 (0.0%)
1 (4.0%)
0 (0.0%)
0 (0.0%)
|
10 (28.6%)
23 (46.9%)
8 (32.0%)
5 (100%)
1 (50.0%)
|
0.318
|
Marital Status
Married
Single
Divorced
Widowed
|
70 (74.5%)
14 (82.4%)
4 (100%)
0 (0 %)
p = 0.179
|
63 (67.0%)
14 (82.4%)
4 (100%)
0 (0 %)
p = 0.129
|
40 (42.6%)
4 (23.5%)
3 (75%)
0 (0 %)
p = 0.183
|
23 (24.5%)
3 (17.6%)
1 (25%)
0 (0 %)
|
28 (29.8%)
7 (41.2%)
3 (75.0%)
0 (0.0%)
|
1 (1.1%)
0 (0.0%)
0 (0.0%)
0 (0.0%)
|
40 (43.5%)
6 (37.5%)
0 (0%)
1 (100 %)
|
0.822
|
Nationality
Bahraini
Non- Bahraini
|
58 (85.3%)
30 (62.5%)
p = 0.005
|
56 (82.4%)
25 (52.1%)
p = 0.005
|
30 (44.1%)
17 (35.4%)
p = 0.347
|
21 (30.9%)
6 (12.5%)
|
23 (33.8%)
15 (31.3%)
|
1 (1.5%)
0 (0.0%)
|
21 (30.9%)
26 (54.2%)
|
0.069
|
Job Position
Resident
Senior Resident
Chief Resident
Consultant
|
18 (75.0%)
34 (69.4%)
26 (86.7%)
10 (76.9%)
p= 0.384
|
18 (75.0%)
30 (61.2%)
24 (80.0%)
9 (69.2%)
p= 0.321
|
8 (33.3%)
15 (30.6%)
17 (56.7%)
7 (53.8%)
p= 0.082
|
5 (20.8%)
9 (18.4%)
8 (26.7%)
5 (38.5%)
|
11(45.8%)
12 (24.5%)
11 (36.7%)
4 (30.8%)
|
0 (0.0%)
0 (0.0%)
0 (0.0%)
1 (7.7%)
|
7 (30.4%)
27 (56.3%)
10 (34.5%)
3 (23.1%)
|
0.124
|
# Work Hours
~ 40 hours
~ 50 hours
~ 60 hours
~ 70 hours
~ 80 hours
|
33 (73.3%)
39 (76.9%)
9 (69.2%)
5 (83.5%)
2 (66.7%)
p = 0.884
|
29 (64.4%)
36 (73.5%)
9 (69.2%)
5 (83.5%)
2 (66.7%)
p = 0.833
|
14 (31.1%)
17 (34.7%)
9 (69.2%)
5 (83.5%)
2 (66.7%)
p = 0.016
|
8 (17.8%)
7 (14.3%)
7 (53.8%)
4 (66.7%)
1 (33.3%)
|
13 (28.9%)
22 (44.9%)
2 (15.4%)
0 (0.0%)
1 (33.3%)
|
1 (2.2%)
0 (0.0%)
0 (0.0%)
0 (0.0%)
0 (0.0%)
|
22 (50%)
20 (40.8%)
3 (25.0%)
1 (20.0%)
1 (33.3%)
|
0.020
|
Sick Leave
Yes
No
|
50 (82.0%)
38 (69.1%)
p= 0.106
|
47 (77.0%)
34 (61.8%)
p= 0.074
|
36 (59.0%)
11 (20.0%)
p= 0.005
|
20 (32.8%)
7 (12.7%)
|
21 (34.4%)
17 (30.9%)
|
1 (1.6%)
0 (0%)
|
16 (26.2%)
31 (56.4%)
|
0.004
|
Sleep disturbances
Yes
No
|
82 (80.4%)
6 (42.9%)
p= 0.002
|
75 (73.5%)
6 (42.9%)
p= 0.019
|
45 (44.1%)
2 (14.3%)
p= 0.033
|
26 (25.5%)
1 (7.1%)
|
35 (34.3%)
3 (21.4%)
|
1 (1.0%)
0 (0%)
|
37 (36.3%)
10 (71.4%)
|
0.05
|
Short/Staggered/ no meals
Yes
No
|
73 (76.8%)
15 (71.4%)
p= 0.60
|
68 (71.6%)
13 (61.9%)
p= 0.382
|
38 (40.0%)
9 (42.9%)
p= 0.809
|
24 (25.3%)
3 (14.3%)
|
32 (33.7%)
6 (28.6%)
|
1 (1.1%)
0 (0%)
|
37 (39.4%)
10 (52.6%)
|
0.175
|
Violence at workplace
Yes
No
|
61 (82.4%)
27 (64.3%)
p= 0.028
|
56 (75.7%)
25 (59.5%)
p= 0.069
|
34 (45.9%)
13 (31.0%)
p= 0.114
|
23 (31.1%)
4 (9.5%)
|
17 (23.0%)
21 (50.0%)
|
1 (1.4%)
0 (0%)
|
30 (42.3%)
17 (40.5%)
|
0.009
|
In Table 6, based on the SOS-S categorical scores, all four groups (High Risk, Challenged, Fragile, Low Risk) reported to suffer from personal burnout and work-related burnout. In terms of patient-related burnout, all four groups reported to suffer from this except the Fragile group, which was represented by 1 participant only.
In Table 6, results of a General Linear Model Test using a univariate ANOVA across the four SOS-S quadrants (High Risk, Challenged, Fragile, Low Risk) can be viewed. Results showed that personal burnout had main effects for personal vulnerability [F (1,190) = 43.72, p < .0001], event load [F (1,288) = 52.06, p < .0001] and for their interaction [F (233) = 26.09, p < .0001]. There was also a significant difference in the means of personal burnout across all four quadrants. There were significant differences in the burnout levels of each CBI dimension between the quadrants. The high-risk group was more likely to suffer from significantly higher levels of personal burnout (M: 83.8 ± 14.4), work-related burnout (M: 83.1 ± 11.3), and patient-related burnout (M: 74.4 ± 15.5), followed by the Challenged, Fragile, and Low Risk groups.
A similar pattern was seen for work-related burnout, with effects for personal vulnerability [F (1,241) = 57.40, p < .0001], event load [F (1,284) = 64.54, p < .0001] and for their interaction [F (240) = 35.18. p < .0001]. Patient-related burnout displayed the same pattern with effects for personal vulnerability [F (440) = 39.47, p < .0001] and event load [F (405) = 48.42, p < .0001] and for their interaction [F (50) = 26.23. p < .0001]. Significant differences in all four quadrants for work-related and patient-related burnout is indicated in Table 6.
Table 6: Prevalence, mean, and standard deviations of the CBI dimensions across the SOS-S categories
|
Personal burnout
|
Work-related burnout
|
Patient-related burnout
|
|
n (%)
|
(Mean ± SD)
|
n (%)
|
(Mean ± SD)
|
n (%)
|
(Mean ± SD)
|
|
Yes
88 (75.8%)
|
No
28 (24.1%)
|
Kruskal-Walis test <0.001
|
Yes
81 (69.8%)
|
No
35 (30.1%)
|
Kruskal-Walis test <0.001
|
Yes
47 (40.5%)
|
No
69 (59.4%)
|
Kruskal-Walis test 0.02
|
High risk
|
26 (29.5%)
|
1 (3.6%)
|
83.8 ± 14.4
|
26 (32.1%)
|
1 (2.9%)
|
83.1 ± 11.3
|
24 (51.1%)
|
3 (4.3%)
|
74.4 ± 15.5
|
Challenged
|
34 (38.65)
|
4 (14.3%)
|
66.3 ± 16.11
|
30 (37.0%)
|
8 (22.9%)
|
68.3 ± 10.5
|
17 (36.2%)
|
21 (30.4%)
|
62.5 ± 13.3
|
Fragile
|
1 (1.1%)
|
0
|
1 (66.67)
|
1 (1.2%)
|
0
|
1 (66.67)
|
0
|
1 (1.4%)
|
1 (66.67)
|
Low risk
|
24 (27.3%)
|
23 (82.1%)
|
62.6 ± 10.9
|
21 (25.9%)
|
26 (74.3%)
|
60.5 ± 9.2
|
5 (10.6%)
|
42 (60.9%)
|
59.1 ± 13.9
|
Data in Table 7 demonstrates the distribution of the continuous scores for the CBI subscales and the SOS-S total scores and its subscales categorized according to the demographic and occupational variables. Bahraini nationals suffered from higher levels of personal burnout (p=0.002), work-related burnout (p= 0.0001), personal vulnerability (p=0.010), event load (p=0.002) and total SOS-S scores compared to non-Bahrainis. Chief Residents followed by Consultants had higher levels of personal burnout (p= 0.035) and work-related burnout (p=0.035) in comparison to other job positions. There were no significant differences based on patient-related burnout and job position. In terms of work hours, physicians who worked more hours had higher levels of personal and work-related burnout with a weak positive correlation coefficient of 0.18 (p=0.04). However, only those who worked 60 hours or more suffered from work-related burnout.
Table 7. Distribution of continuous scores for CBI, SOS-S and its subscales by demographic and occupational variables
|
CBI
|
SOS-S
|
|
Personal burnout
(Mean ± SD, p-value)
|
Work-related burnout (Mean ± SD, p-value)
|
Patient-related burnout (Mean ± SD, p-value)
|
Personal vulnerability (Mean ± SD, p-value)
|
Event load (Mean ± SD, p-value)
|
SOS-S Total Scores
(Mean ± SD, p-value)
|
Gender
Male
Female
|
61.36 ± 23.52
66. 83 ± 19.17
p = 0.15
|
59.12 ± 21.95
64.07 ± 20.10
p = 0.24
|
44.13 ± 26.74
41.17 ± 22.49
p = 0.62
|
13.55 ± 5.09
15.00 ± 4.62
p = 0.057
|
16.37 ± 4.91
17.87 ± 4.24
p = 0.118
|
29.19 ± 9.62
32.71 ± 8.12
p = 0.058
|
Age
25-34
35-44
45-54
>55
|
62.14 ± 19.70
63.29 ± 23.28
68.33 ± 23.66
46. 42 ± 19.07
p = 0.08
|
62.34 ± 18.23
61.29 ± 23.05
61.00 ± 21.55
40.81 ± 21.01
p = 0.08
|
40.71 ± 26.62
43.96 ± 25.07
47.16 ± 26.20
34.52 ± 20.51
p = 0.66
|
13.77 ± 4.35
14.12 ± 5.57
14.54 ± 5.24
10.71 ± 2.69
p = 0.30
|
17.91 ± 3.74
16.00 ± 5.59
17.48 ± 4.29
13.00 ± 3.00
p = 0.049
|
31.68 ± 7.50
30.00 ± 10.82
32.04 ± 9.11
23.71 ± 5.28
p = 0.119
|
Marital Status
Married
Single
Divorced/
Widowed
|
63.27 ± 22.74
62.99 ± 19.48
58.33 ± 28.25
p = 0.99
|
59.68 ± 21.75
61.76 ± 15.60
67.14 ± 36.36
p = 0.99
|
44.14 ± 26.22
35.29 ± 20.94
22.82 ± 10.20
p = 0.34
|
13.93 ± 5.17
13.50 ± 4.38
14.40 ± 5.17
p = 0.85
|
16.54 ± 4.90
17.62 ± 3.89
17.20 ± 6.01
p = 0.612
|
30.12 ± 9.65
31.12 ± 7.47
31.60 ± 11.03
p = 0.770
|
Nationality
Bahraini
Non- Bahraini
|
68.46 ± 21.17
55.29 ± 21.92
p = 0.002
|
66.22 ± 20.06
51.93 ± 21.01
p < 0.0001
|
46.69 ± 26.87
38.02 ± 22.57
p = 0.097
|
14.85 ± 4.89
12.53 ± 4.96
p = 0.006
|
17.86 ± 4.34
15.16 ± 4.98
p = 0.003
|
32.63 ± 8.66
27.65 ± 9.62
p = 0.002
|
Job Position
Resident
Senior Resident
Chief Resident
Consultant
|
60.76 ± 21.24
57.83 ± 21.61
71.30 ±20.92
67.62 ± 26.19
p = 0.035
|
60.26 ± 18.40
56.19 ± 22.12
66.66 ± 19.36
61.26 ± 27.77
p = 0.035
|
42.01 ± 26.49
38.01 ± 24.15
49.30 ± 24.54
50.00 ± 28.51
p = 0.14
|
13.30 ± 4.03
12.89 ± 5.40
14.76 ± 4.10
16.61 ± 6.25
p = 0.068
|
17.91 ± 3.55
15.44 ± 5.19
17.48 ± 4.46
17.76 ± 5.19
p = 0.092
|
31.21 ± 6.91
28.31 ± 10.27
32.06 ± 8.13
34.38 ± 10.98
p = 0.074
|
# Working Hours
~ 40 hours
~ 50 hours
~ 60 hours
~ 70 hours
~ 80 hours
|
57.77 ± 19.83
62.52 ± 19.35
68.07 ± 31.86
84.02 ± 24.91
52.77 ± 33.67
p = 0.042
|
56.26 ± 21.36
60.64 ± 17.56
67.03 ± 27.51
77.38 ± 24.81
52.38 ± 39.82
p = 0.042
|
37.77 ± 24.35
41.07 ± 23.32
55.76 ± 28.18
68.75 ± 29.66
50.00 ± 18.16
p = 0.030
|
12.86 ± 5.22
13.48 ± 4.33
17.53 ± 4.99
18.40 ± 4.15
12.33 ± 6.42
p = 0.015
|
15.65 ± 4.75
16.77 ± 4.68
19.16 ± 4.36
20.16 ± 4.91
15.00 ± 4.35
p = 0.074
|
28.52 ± 9.40
30.26 ± 8.50
36.50 ± 9.35
39.20 ± 9.14
27.33 ± 10.69
p = 0.035
|
Sick Leave
Yes
No
|
69.63 ± 22.80
55.68 ± 19.56
p< 0.0001
|
66.51 ± 20.25
53.44 ± 21.05
p = 0.001
|
50.88 ± 26.68
34.46 ± 21.05
p < 0.0001
|
15.71 ± 4.65
11.94 ± 4.72
p < 0.0001
|
18.10 ± 4.44
15.25 ± 4.75
p = 0.001
|
33.75 ± 8.70
27.20 ± 8.91
p < 0.0001
|
Sleep disturbances
Yes
No
|
65.66 ± 21.48
43.75 ± 19.59
p = 0.001
|
62.85 ± 20.92
41.83 ± 17.12
p = 0.001
|
45.46 ± 25.41
25.89 ± 18.57
p = 0.008
|
14.41 ± 5.00
10.21 ± 3.57
p = 0.003
|
17.23 ± 4.58
13.14 ± 4.89
p = 0.006
|
31.58 ± 9.10
23.35 ± 8.19
p = 0.003
|
Short/Staggered/ no meals
Yes
No
|
63.30 ± 22.51
61.70 ±22.15
p= 0.486
|
61.12 ± 21.84
56.63 ± 20.26
p= 0.334
|
43.81 ± 25.16
39.88 ± 27.08
p= 0.56
|
14.13 ± 4.97
12.68 ± 5.29
p= 0.27
|
17.01 ± 4.70
15.40 ± 5.12
p= 0.15
|
31.08 ± 9.18
28.00 ± 10.06
p= 0.17
|
Violence at work place
Yes
No
|
67.19 ± 22.57
55.65 ± 20.20
p= 0.004
|
64.14 ± 21.42
53.57 ± 20.35
p= 0.009
|
47.24 ± 26.39
35.81 ± 22.14
p= 0.038
|
14.80 ± 5.22
12.33 ± 4.30
p= 0.007
|
17.08 ± 4.90
16.11 ± 4.59
p= 0.290
|
31.81 ± 9.74
28.45 ± 8.37
p= 0.051
|
Note: significant p-values are listed in bold.
Hours of work was also positively correlated to patient-related burnout (r=0.273, p=0.03), personal vulnerability (r=0.24, p=0.015) and event load (r=0.20, p=0.035). In other words, those who worked for longer hours were more likely to report higher external workload, higher likelihood of perceiving their workload as stressful, and also higher patient-related burnout. The physicians who had taken at least one sick leave had a higher mean of personal burnout score (p<0.0001), work-related burnout (p=0.001), patient-related burnout (p< 0.0001), personal vulnerability (p< 0.0001), event load (p= 0.001) and total SOS-S scores (p< 0.0001). As expected, physicians with sleep disturbances had higher levels of personal burnout (p= 0.001), work-related burnout (p=0.001), patient-related burnout (p= 0.008), personal vulnerability (p= 0.003), event load (p= 0.006) and total SOS-S scores (p=0.003). The physicians who suffered from workplace violence were more likely to suffer from higher levels of personal burnout (p= 0.004), work related burnout (p=0.009), patient-related burnout (p= 0.038), and personal vulnerability (p= 0.007).
Significant Predictors of Burnout
Factors associated with personal burnout
Factors that were significantly associated with each burnout subscale (based on bivariate analysis) were included as independent variables in the binary logistic regression models. Table 8 lists the significant predictors of each CBI subscale. Exposure to violence and Event Load were strong predictors of personal burnout as per this model. Physicians who reported having experienced violence at the workplace were more likely to suffer from personal burnout (OR = 3.45, 95 % CI = 0.99 – 11.98, p = 0.04). Event load (having high external stressors) was also a strong predictor of personal burnout, where those who reported higher event load were more likely to suffer from personal burnout (OR = 1.60, 95 % CI = 1.30 – 1.98, p <0.0001). This model for personal burnout showed a R2 value of 47.1%.
Factors associated with work-related burnout
Nationality and Event Load were strong predictors of work-related burnout as per this model, with a R2 value of 37.1%. Physicians of Bahraini nationality suffered from higher work-related burnout than non-Bahrainis (OR = 3.30, 95 % CI = 1.26– 8.63, p = 0.015). Event load was also a strong predictor of work-related burnout (OR = 1.27, 95 % CI = 1.13 – 1.42, p <0.0001).
Factors associated with patient-related burnout
Sick leave status and personal vulnerability were significant predictors of patient-related burnout as per this model, which had a R2 value of 58.2%. EPs who had taken sick leave were 5.16 times more likely to have patient-related burnout than those did not take sick leave (OR = 5.16, 95 % CI = 1.70– 15.60, p <0.005). Personal vulnerability was also a strong predictor of patient-related burnout (OR = 1.49, 95 % CI = 1.25 – 1.77, p <0.0001). Those who perceived demands as more stressful (high PV) were more likely to suffer from patient-related burnout.
In summary, Event Load was a predictor of both personal and work-related burnout, whereas Personal Vulnerability was a predictor of patient-related burnout. However, the total stress overload score was not a significant predictor of any of the three CBI dimensions of burnout.
Table 8. Multiple regression analysis of factors associated with CBI subscales
PERSONAL BURNOUT (Model with R2 = 0.471)
|
Factor
|
Unadjusted
|
|
Adjusted
|
|
OR
|
95% CI
|
P value
|
|
OR
|
95% CI
|
P value
|
Bahraini Nationality
|
3.45
|
1.42 – 8.47
|
0.006
|
|
|
|
|
Lack of sleep
|
5.46
|
1.70- 17.54
|
0.04
|
|
|
|
|
Violence at work place
|
2.60
|
1.09 – 6.22
|
0.03
|
|
3.45
|
0.99 – 11.98
|
0.044
|
PV
|
1.32
|
1.15- 1.51
|
0.00
|
|
|
|
|
EL
|
1.43
|
1.24 – 1.66
|
0.00
|
|
1.60
|
1.30 – 1.98
|
0.000
|
WORK – RELATED BURNOUT (Model with R2 = 0.371)
|
Factor
|
Unadjusted
|
|
Adjusted
|
|
OR
|
95% CI
|
P value
|
|
OR
|
95% CI
|
P value
|
Bahraini Nationality
|
4.29
|
1.84 – 9.96
|
0.001
|
|
3.30
|
1.26 – 8.63
|
0.015
|
Lack of sleep
|
3.70
|
1.17 – 11.65
|
0.02
|
|
|
|
|
PV
|
1.30
|
1.15 – 1.47
|
0.00
|
|
|
|
|
EL
|
1.30
|
1.16- 1.45
|
0.00
|
|
1.270
|
1.13 – 1.42
|
0.000
|
PATIENT–RELATED BURNOUT (Model with R2 = 0.582)
|
Factor
|
Unadjusted
|
|
Adjusted
|
|
OR
|
95% CI
|
P value
|
|
OR
|
95% CI
|
P value
|
>60 working hours
|
5.41
|
1.93 – 15.21
|
0.01
|
|
|
|
|
Sick Leave
|
5.76
|
2.50 – 13.27
|
0.00
|
|
3.71
|
1.32 – 10.39
|
0.12
|
Lack of sleep
|
4.73
|
1.0 – 22.25
|
0.05
|
|
|
|
|
PV
|
1.41
|
1.21 – 1.62
|
0.00
|
|
1.41
|
1.22 – 1.63
|
0.00
|
EL
|
1.35
|
1.20 – 1.53
|
0.00
|
|
|
|
|