1. Background
Physician burnout is an important issue associated with physician attrition, mental and physical health, and self-reported medical errors (1). The "burnout" phenomenon caused by work-related stress is a challenge for academic psychiatry both conceptually and professionally. More than 140 definitions have been suggested from the first description of burnout in 1974 (2). Therefore, a consensual definition of occupational burnout is currently lacking (3). However, according to a recent systematic review, several studies focusing on burnout have used Maslach Burnout Inventory (MBI) to assess burnout (4). Considering MBI, burnout results from chronic exposure to stress: comprising emotional exhaustion (EE), depersonalization (DP), and a reduced sense of personal achievement (PA) (5). Burnout has been gaining much attention, mainly due to its adverse impacts such as physicians’ poor health and decreased functional outcomes, lower quality of care, the shrinking physician workforce, and compromised patient safety (6). Burnout for each individual is a result of the interplay between his/her characteristics and environmental factors. However, Stigma and fear of professional consequences appear to be a leading barrier for physicians to express their burnout and access services (7). Furthermore, this phenomenon has high costs for healthcare systems (8). Thus, any additional evidence on this issue from all over the world is of great importance.
The coronavirus disease 2019 (COVID-19) pandemic has brought several challenges to an underprepared healthcare system. Healthcare providers face unpredicted acute workplace stress as well as a higher rate of physician burnout (9). The rates of physician burnout, depression, and suicide increased over the past 50 years. However, during the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and COVID-19 pandemics, these numbers are predicted to show a steep increase due to the increased work demands, social isolation, decreased self-care, and increased exposure to emotionally traumatic events at work and home (10). Although the issue of burnout among physicians is not new, the COVID-19 pandemic is expanding the negative consequences of inadequate support by the authorities for solving this problem (11). The pandemic has already posed strain on the entire healthcare system (12). Consequently, it is vital to provide precise data on the prevalence of burnout during different periods since the beginning of the pandemic.
2. Objectives
The present study aimed to compare the prevalence of overall burnout and its related sub-dimensions (using MBI) among interns-residents and general physician (GPs)-specialists in three different periods since the beginning of the COVID-19 pandemic. In addition, our study takes a new look at the association between different variables, including working in COVID-19 wards, with the overall burnout of both mentioned groups.
3. Methods
This cross-sectional study was conducted among 220 interns-residents and 212 (GPs)-specialists working at hospitals affiliated with Shiraz University of Medical Sciences, southern Iran. In this study, we tried to investigate the prevalence of burnout among physicians at various time intervals after the onset of the COVID-19 pandemic. Therefore, the study was performed in three periods of time, including March-April 2020, June-July 2020, and December 2020-January 2021. These physicians had a regular working schedule during the COVID-19 outbreak.
Due to the lack of the cooperation of physicians, we were not able to apply a sampling framework in this study, so we used a convenience sampling method to recruit participants. We reached the sample size in each group based on the participation of physicians. To collect data, a trained questioner referred to university-affiliated teaching hospitals and their related outpatient clinics in Shiraz. Questionnaires were completed by participants who met inclusion criteria. Experience of at least six months presence at clinical wards was considered inclusion criteria for interns. Willingness to participate in the study was considered an inclusion criterion for all participants.
In this study, we gathered data using a researcher-made checklist and a valid questionnaire. The checklist comprises demographic features and job characteristics, including working experience and taking care of COVID-19 patients. We also applied the Persian version of the Maslach Burnout Inventory (MBI), which included nine items on emotional exhaustion (EE), eight items on depersonalization (DP), and five items on personal accomplishment (PA) (15). The scoring of these items ranges from zero (never) to six (every day). The MBI has frequency and intensity ratings. To avoid redundancy, the intensity rating was used in this study. The Persian version of the MBI was previously validated in the study of Sepehri Shamloo et al. with the Alpha Cronbach value of 0.76 (16).
In this study, a physician with a high score for either the depersonalization and/or emotional exhaustion dimensions was considered to have at least one manifestation of job burnout. The level of EE was classified high if the score was greater than 26, intermediate if it was from 17 to 26, and low if the score was less than 17. In the DP dimension, the overall score of more than 12 was considered high depersonalization, and the scores from 7 to 12 and less than seven were deemed to be intermediate and low levels, respectively. PA dimension scores include higher than 39 (top-level), from 32 to 38 (moderate-level), and those below 32 (low-level). According to Li et al., an EE score equal to or more than 27 or a DP score equal to or more than 13 was considered overall burnout (17).
3.1. Ethics Approval and Consent to Participate
The research protocol of this study was evaluated and approved by the Ethics Committee of Shiraz University of Medical Sciences (code: IR.SUMS.REC.1398.877). All the participants were asked to fulfill their written consent forms before completing the questionnaires. All participants completed the questionnaires willingly and were ensured of the confidentiality of the collected data.
3.2. Statistical Analysis
We analyzed data using IBM SPSS 18. Quantitative and qualitative variables were described by mean ± standard deviation (SD) and frequency (percent). The chi-square test was used to compare qualitative variables between the groups. The total score of burnout and related sub-dimensions were compared between the periods using analysis of covariance adjusted for age and sex. To determine factors related to the overall burnout, we employed a multiple logistic regression. Because of the limited number of the variables included in the study, all of them were entered into the regression model. A P value of less than 0.05 was considered statistically significant.
4. Results
The participants’ characteristics are shown in Table 1. The Mean age of the participants was 25.33 ± 1.89 (ranged 21 to 31) years for interns-residents and 37.55 ± 9.01 (ranged 26-62) years for GPs-specialists. Most GPs-specialists were married and female. Working experience in COVID-19 centers was significantly lower in December-January than March-April.
Table 1.Participants’ Characteristics Measured in Different Periods
| Characteristics | Interns-Residents | GPs-Specialists | |||||
|---|---|---|---|---|---|---|---|
| Mar-Apr (n = 158) | Dec-Jan (n = 62) | P Value | Mar-Apr (n = 76) | Jun-Jul (n = 43) | Dec-Jan (n = 93) | P Value | |
| Age, y | 25.11 ± 1.83 | 25.88 ± 1.95 | 0.01 | 30.90 ± 4.43 | 41.62 ± 9.99 | 41.09 ± 8.24 | 0.01 |
| Work experience, y | 2.62 ± 1.01 | 2.25 ± 1.36 | 0.03 | 5.27 ± 2.62 | 11.51 ± 8.58 | 10.81 ± 8.35 | 0.01 |
| Gender | |||||||
| Male | 86 (54.43) | 25 (40.32) | 0.06 | 35 (46.05) | 20 (46.51) | 44 (47.31) | 0.98 |
| Female | 72 (45.56) | 37 (59.67) | 41 (53.94) | 23 (53.48) | 49 (52.68) | ||
| Marital status | |||||||
| Single | 140 (88.60) | 51 (82.25) | 0.21 | 45 (59.21) | 18 (41.86) | 16 (17.20) | 0.01 |
| Married | 18 (11.39) | 11 (17.74) | 31 (40.78) | 25 (58.13) | 77 (82.79) | ||
| Working in COVID-19 ward | |||||||
| No | 14 (8.86) | 40 (64.51) | 0.01 | 12 (15.78) | 0 (0.00) | 40 (43.01) | 0.01 |
| Yes | 144 (91.13) | 22 (35.48) | 64 (84.21) | 43 (100) | 53 (56.98) | ||
| Experience of resuscitation maneuver for COVID-19 patients | |||||||
| 0 | 30 (18.98) | 39 (62.90) | 0.01 | 13 (17.10) | 27 (62.79) | 42 (45.16) | 0.01 |
| 1 - 4 | 101 (63.92) | 11 (17.74) | 43 (56.57) | 12 (27.90) | 30 (32.25) | ||
| ≥ 5 | 27 (17.08) | 12 (19.35) | 20 (26.31) | 4 (9.30) | 21 (22.58) | ||
| Experience of dealing with patients died from COVID-19 | |||||||
| 0 | 19 (12.02) | 18 (29.03) | 0.01 | 11 (14.47) | 23 (53.48) | 25 (26.88) | 0.01 |
| 1 - 4 | 81 (51.26) | 22 (35.48) | 23 (30.26) | 15 (34.88) | 26 (27.95) | ||
| 5 - 9 | 38 (24.05) | 9 (14.51) | 21 (27.63) | 2 (4.65) | 12 (12.90) | ||
| ≥ 10 | 20 (12.65) | 13 (20.96) | 21 (27.63) | 3 (6.97) | 30 (32.25) | ||
The comparison of the total score of burnout and related sub-dimensions is presented in Table 2. For Interns and Residents, the mean score of burnout, depersonalization, and reduced achievement was significantly higher in December-January than March-April. In June-July, GPs and specialists reported the highest level of burnout, EE, and DP. Moreover, EE for interns and residents and reduced achievement for GPs and specialists were not significantly different between the periods.
One-hundred and fifty-seven (71.4%) interns-residents and 141 (66.5%) GPs-specialists had overall burnout. The prevalence of overall burnout was not statistically different between the two groups (P = 0.27).
Table 2.Comparison of Total Score of Burnout and Related Sub-dimensions
| Burnout Dimensions | Interns-Residents | GPs-Specialists | |||||
|---|---|---|---|---|---|---|---|
| Mar-Apr | Dec-Jan | P Value a | Mar-Apr | Jun-Jul | Dec-Jan | P Value a | |
| Burnout | 55.24 ± 9.97 | 60.66 ± 12.24 | 0.01 | 53.65 ± 13.28a | 55.83 ± 9.27b | 47.02 ± 20.04ab | 0.01 |
| Emotional exhaustion | 27.99 ± 7.32 | 28.40 ± 7.19 | 0.94 | 29.31 ± 7.64a | 32.83 ± 6.93a | 23.45 ± 12.09a | 0.01 |
| Depersonalization | 8.86 ± 4.60 | 10.98 ± 4.86 | 0.01 | 7.40 ± 4.66 | 8.65 ± 4.38a | 6.12 ± 5.49a | 0.02 |
| Reduced achievement | 18.38 ± 6.00 | 21.27 ± 5.22 | 0.01 | 16.93 ± 6.45 | 14.34 ± 4.35 | 17.44 ± 7.77 | 0.05 |
a Comparisons between the periods of time are adjusted for age and sex.
In the rows, same lower letters indicate significant differences between the periods of time.
In the univariate analysis, for interns-residents, all determining factors except age (OR = 1.18, P = 0.04, 95% CI: 1.01 - 1.38) were not significantly associated with overall burnout (P > 0.05). Working in COVID-19 wards during the studied periods was not significantly related to overall burnout (OR = 1.04, P = 0.94, 95% CI: 0.33 - 3.24). Multiple logistic regression revealed that age (OR = 0.24, P = 0.01, CI: 0.04 - 0.47) and being married (OR = 0.31, P = 0.02, CI: 0.12 - 0.81) were positively associated with decreased odds of overall burnout among interns-residents. Detailed results are shown in Table 3.
Table 3.Determinant Factors of Overall Burnout Among Interns and Residents
| Determinant Factors | Overall Burnout | Multiple Logistic Regression | |||
|---|---|---|---|---|---|
| No | Yes | OR | 95% CI | P Value | |
| Periods * working in COVID-19 ward | |||||
| Dec-Jan * work (Yes) | 5 (22.7) | 17 (77.3) | 1.04 | 0.33-3.24 | 0.94 |
| Dec-Jan * work (No) | 10 (25) | 30 (75) | - | - | - |
| Mar-Apr * work (Yes) | 43 (29.9) | 101 (70.1) | - | - | - |
| Mar-Apr * work (No) | 5 (35.7) | 9 (64.3) | - | - | - |
| Periods working in COVID-19 ward | |||||
| Dec-Jan | 15 (24.2) | 47 (75.8) | - a | - a | - a |
| Mar-Apr | 48 (30.4) | 110 (69.6) | - a | - a | - a |
| No | 15 (27.8) | 39 (72.2) | - a | - a | - a |
| Yes | 48 (28.9) | 118 (71.1) | - a | - a | - a |
| Age, y | 25.50 ± 1.84 | 24.92 ± 1.97 | 0.24 | 0.04-0.47 | 0.01 |
| Gender | |||||
| Male | 33 (29.7) | 78 (70.3) | - | - | - |
| Female | 30 (27.5) | 79 (72.5) | 1.19 | 0.64-2.32 | 0.56 |
| Marital status | |||||
| Single | 52 (27.2) | 139 (72.8) | - | - | - |
| Married | 18 (62.1) | 11 (37.9) | 0.31 | 0.12-0.81 | 0.02 |
| Experience of resuscitation maneuver for COVID-19 patients | |||||
| No | 22 (31.9) | 47 (68.1) | - | - | - |
| 1 - 4 | 34 (30.4) | 78 (69.6) | 0.83 | 0.37-1.83 | 0.64 |
| ≥ 5 | 7 (17.9) | 32 (82.1) | 2.21 | 0.67-7.22 | 0.18 |
| Experience of dealing with patients died from COVID-19 | |||||
| 0 | 15 (40.5) | 22 (59.5) | - | - | - |
| 1 - 4 | 30 (29.1) | 73 (70.9) | 1.67 | 0.66-4.23 | 0.27 |
| 5 - 9 | 11 (23.4) | 36 (76.6) | 2.03 | 0.66-6.24 | 0.21 |
| ≥ 10 | 7 (21.2) | 26 (78.8) | 1.99 | 0.57-6.94 | 0.27 |
a To prevent redundancy, main effects of periods of time and working in COVID-19 ward were not included in the multiple logistic model.
According to the univariate analysis, the higher level of burnout among the GPs-specialists was reported in June-July (OR = 4.44, P = 0.02, 95% CI: 1.23 - 16.03) and males (OR = 2.03, P = 0.01, 95% CI: 1.12 - 3.66). Also, higher level of overall burnout is reported in December-January (OR = 0.29, P = 0.01, 95% CI: 0.15 - 0.57) and older age GPs-specialists (OR = 0.95, P = 0.01, 95% CI: 0.93 - 0.99). Multiple logistic regression depicted that working in COVID-19 wards during June-July (OR = 13.93, P = 0.01, 95% CI: 3.62 - 53.51) was associated with increased odds of overall burnout. However, older age was associated with decreased odds of overall burnout (OR = 0.94, P = 0.01, 95% CI: 0.90 - 0.98). Detailed results of overall burnout and its determinant factors are presented in Table 4.
Table 4.Determinant Factors of Overall Burnout Among General Physician and Specialist
| Determinant Factors | Overall Burnout | Multiple Logistic Regression | |||
|---|---|---|---|---|---|
| No | Yes | OR | 95% CI | P Value | |
| Periods * working in COVID-19 ward | |||||
| Dec-Jan * work (Yes) | 27 (50.9) | 26 (49.1) | 0.73 | 0.33-1.57 | 0.42 |
| Dec-Jan * work (No) | 22 (55) | 18 (45) | - | - | - |
| Jun-Jul * work (Yes) | 3 (7) | 40 (93) | 13.93 | 3.62-53.51 | 0.01 |
| Jun-Jul * work (No) | 0 (0) | 0 (0) | - | - | - |
| Mar-Apr * work (Yes) | 18 (28.1) | 46 (71.9) | - | - | - |
| Mar-Apr * work (No) | 1 (8.3) | 11 (91.7) | - | - | - |
| Periods | |||||
| Dec-Jan | 49 (52.7) | 44 (47.3) | - a | - a | - a |
| Jun-Jul | 3 (7) | 40 (93) | - a | - a | - a |
| Mar-Apr | 19 (25) | 57 (75) | - a | - a | - a |
| Working in COVID-19 ward | |||||
| No | 23 (44.2) | 29 (55.8) | - a | - a | - a |
| Yes | 48 (30) | 112 (70) | - a | - a | - a |
| Age, y | 39.83 ± 8.77 | 36.40 ± 8.95 | 0.94 | 0.90-0.98 | 0.01 |
| Gender | |||||
| Male | 25 (25.3) | 74 (74.7) | 1.89 | 0.98-3.64 | 0.06 |
| Female | 46 (40.7) | 67 (59.3) | - | - | - |
| Marital status | |||||
| Single | 20 (25.3) | 59 (74.7) | - | - | - |
| Married | 51 (38.3) | 82 (61.7) | 0.99 | 0.45-2.15 | 0.97 |
| Experience of resuscitation maneuver for COVID-19 patients | |||||
| No | 29 (35.4) | 53 (64.6) | - | - | - |
| 1 - 4 | 29 (34.1) | 56 (65.9) | 0.9 | 0.39-2.05 | 0.8 |
| ≥ 5 | 13 (28.9) | 32 (71.1) | 1.87 | 0.65-5.35 | 0.23 |
| Experience of dealing with patients died from COVID-19 | |||||
| 0 | 20 (33.9) | 39 (66.1) | - | - | - |
| 1 - 4 | 19 (29.7) | 45 (70.3) | 1.35 | 0.52-3.50 | 0.53 |
| 5 - 9 | 9 (25.7) | 26 (74.3) | 1.63 | 0.51-5.20 | 0.4 |
| ≥ 10 | 23 (42.6) | 31 (57.4) | 0.83 | 0.28-2.41 | 0.73 |
aTo prevent redundancy, main effects of periods of time and working in COVID-19 ward were not included in the multiple logistic model.
5. Discussion
In this study, we have provided further information on overall professional burnout and its related sub-dimensions among Iranian interns-residents and GPs-specialists during three separated periods during the COVID-19 pandemic. In addition, we have searched for a significant association between sex, age, and working in COVID-19 wards with the overall burnout in both groups.
While there was no significant difference between the prevalence of overall burnout among interns-residents and GPs-specialists, the highest mean score of burnout and its sub-dimensions including depersonalization, and reduced achievement for interns-residents, was attributed to December-January. This score for GPs-specialists was at the highest level during June-July, with an exception in the sub-dimension of reduced achievements. The higher values of overall burnout score in December and January correlate reasonably well with the study conducted by Amanullah and Ramesh Shankar (18) and support the concept of increasing burnout resulting from the workload due to COVID-19. These results substantiate previous findings in the literature, which has shown a statistically significant decrease in the physicians’ wellness during the pandemic of COVID-19 as a result of feeling less happiness at work, having more stress on days not being at work, having more trouble falling asleep, having a sense of dread when thinking of work needing to be done, and being more concern about ones’ health and the health of the beloved ones (19).
Moving through the rest of our results, the most striking point was the correlation between working in COVID-19 wards and burnout in June-July for GPs-specialists versus lack of any similar correlation for interns-residents, according to both univariate and multivariate analysis. In line with the results of this study, a study carried out by Sunjaya et al. points to the higher risk of experiencing depressive symptoms and burnout for the healthcare personnel who are responsible for treating COVID-19 patients over time (20). On the contrary, a recent study has shown that although the effect of COVID-19 on the work environment and personal perceptions, as well as fears about its impacts on lifestyle, have affected physicians’ well-being, burnout levels among emergency physicians remained stable during the initial weeks of the pandemic (21).
It seems that some factors such as workloads, personal and demographic characteristics, and organizational factors could be considered contributors to burnout among physicians (22) during the COVID-19 pandemic.
In the case of interns-residents as trainee physicians, although there was no special correlation between burnout and working in COVID-19 wards, there are many workplace-related factors leading to burnout which exist in all wards of the hospital, including COVID-19 wards (23, 24). Therefore it could be hypothesized that the lower responsibility of interns-residents to COVID-19 patients than GPs-specialists may justify the differences between these two groups.
On the other side, our study showed that age correlated with burnout for both groups of GPs-specialists and interns-residents so that being older was positively associated with the reduced risk of burnout. In this regard, it is reported that physicians with more professional experience demonstrated lower rates of burnout, while early-career physicians demonstrated higher rates of burnout (25). It is suggested that the peer support networks among physicians with more professional experience may lead to reduced burnout during the pandemics (26).
Based on the results of this study and related reports, it could be concluded that exposure to COVID-19 does not necessarily correlate with increased burnout among physicians. Therefore the issue is still could be considered an area for further investigation. Focusing on organizational resilience and institutional support (27), improving the working environment, rewarding physicians fairly and equitably (28), and fostering coping skills (29-31) could be suggested as the most essential strategies to reduce physicians’ burnout during pandemics.
5.1. Limitations of the study
One of the main limitations of the study was the lack of a sampling framework due to the limited cooperation of physicians to participate in the study. Moreover, the willingness of some individuals to participate in the study based on their degree of burnout and well-being may have an effect on the results of the study. Another limitation of this study was the reluctance of participants to complete the questionnaires due to the high volume of activities. We tried to address this problem by describing the importance of research to the participants. Moreover, given that our findings are based on a limited number of participants, the result from such analysis should be considered with the utmost caution.
5.2. Conclusions
This study revealed that the risk of burnout among GPs and specialists increased over time during the COVID-19 pandemic. Moreover, physicians who were older had a lower risk of burnout. Despite the contradictions regarding the relationship between professional burnout and the duration of exposure to patients during epidemics, the results of this study have the potential to remind policymakers of the importance of burnout issue among physicians during pandemics to taking appropriate action in order to prevent this phenomenon. Finally, it should be noted that the issue is still an area for more researches.
