Introduction

In recent years, burnout syndrome has been a major concern widely discussed in the area of occupational health1,2. It was described as a state of physical and emotional exhaustion resulting from extended exposure to a stressful and demanding situations work environment3,4. It may occur in a very wide range of work contexts, and in particular in demanding jobs such as healthcare providers5,6,7. Healthcare workers (HCWs) belong to the most devoted servants to humanity which gives them a lifetime of professional gratification8,9. However, healthcare was listed among the top high-stress professions that provoke a high level of burnout5. Despite the variation in the extent of burnout among HCWs, several studies found that all medical staff including physicians10,11, pharmacists12, nurses13,14, and lab technicians15 experienced burnout.

Of note, physicians are among the top potential candidates for burnout16. Notably, burnout among physicians begins to cultivate its seeds during their education period, goes along with the residency, and finally matures and crowned their practicing life17,18,19. This could be ensuing of the exposure of physicians to high levels of work distress, persistent tension, extensive working hours, a wide range of tasks, and interaction with patients and their relatives, and colleagues as well20,21. They also deal regularly with several complex situations, including responsibility for the health of patients, high patient and family expectations, patients’ and families’ aggressive behaviors complaints and high expectations, and coping with death and injury22,23,24. Physicians who encounter these issues are more likely to have psychological and physical exhaustion which leads them to be cynical about their work25,26,27.

Several studies conducted before the COVID-19 pandemic have indicated a high prevalence of burnout among physicians28,29,30. The prevalence of burnout among physicians varies between countries; ranging from 3.7 to 54.1%29,31,32. In Arab countries, such prevalence ranged from 12.6 to 41.94%33. Another study estimated that one in every three physicians would suffer from burnout at a given time34,35. Of note, burnout among physicians has devastating personal and professional consequences and could incite them towards turnover, early retirement, and poor job performance36,37. Besides, it impacted negatively the quality of care provided to patients and increased the risk for medical errors38,39,40.

In the era of the COVID-19 pandemic, the prevalence of burnout among physicians is snowballing. Physicians experienced ever-increasing pressure in their daily lives, particularly at their work41,42,43. This upsurge was reported in some studies conducted worldwide44,45. Similar to other countries, Lebanon experienced many challenges imposed by the COVID-19 pandemic on its healthcare system which was already in a fragile state even before the pandemic46, the economic collapse47, and the Beirut blasting48. It was overwhelmed by the humanitarian crisis revealed by the influx of more than one million Syrian refugees49. However, the COVID-19 pandemic overlapped with an economic crisis that has its roots in the aftermath of the civil war goaded by corruption and mishandling of the country’s resources50. This economic crisis was ranked by the World Bank among the world’s three worst crises since the mid-1800s affecting living standards where the Lebanese pound has lost more than 90% of its value since the fall of 201951. This was later followed by the devastating Beirut blast, which was coupled with a meteoric soar in COVID-19 infections and hospitalizations where ICU occupancy in the hospitals touched 95% in January 202152. In comparison with other HCWs, physicians bear the large toll of the pandemic53. In addition, the growing number of physicians diagnosed with COVID-19 unveiled gaps in policies and laws intended to warrant physician safety such as coverage for healthcare, disability, and death54. As a result of these consecutive and combined events, Lebanese physicians are leaving to find a better life elsewhere55. In such a typical context of multiple calamities that fueled mental health problems and burnout; it is of great interest to assess the level of physical and psychological burnout experienced by Lebanese physicians in its three domains: personal, work-related, and client-related burnout using a recognized free-of-charge validated tool and to understand as well its determinants in order to prevent such syndrome and reduce its negative impact. Of note, concerns about the pandemic56 and financial wellbeing57 were able both to instigate psychological illnesses and could interact and increase the burnout among physicians.

This study aimed to assess the prevalence of burnout among Lebanese physicians stranded amid the mixture of crises, along with how sociodemographic factors, work-related factors, economic factors, and pandemic-related factors affect the intensity of burnout. Besides, we targeted to assess the combined effects of the COVID-19 pandemic and economic crisis on burnout.

Methods

Study design and population

A web-based quantitative cross-sectional study was conducted among Lebanese physicians from the eight Lebanese provinces using a snowball sampling technique. It was conducted in December 2020. Participants were identified via professional groups and health facilities and were electronically invited to participate.

Physicians were contacted via phone call and notified about the survey and its purpose. Upon their agreement to participate, an online questionnaire using a Google form was sent to them via email or WhatsApp as per their preference. They were invited if possible to disseminate the link of the survey among their colleagues. All practicing Lebanese physicians who had access to the internet were eligible to be part of the study. Physicians who are not practicing currently, those who were out of the country at the time of the survey, retired physicians, interns, and those who refused to give informed consent were excluded.

Sample size calculation

Using an estimated population of 10,918 physicians58, an expected response of 50%, a 95% confidence level, and an estimated absolute error of 5%, the requisite sample size was calculated using the RAOSOFT digital sample size calculator which yielded the least required sample size of 372 participants.

Ethical consideration

The study has no foreseeable risks and written informed consent was obtained in an electronic format. The study was conducted following the standards issued by the World Medical Association’s Declaration of Helsinki guidance. The study was exempted from ethical approval by the Lebanese Ministry of Public Health. Participants were reassured that their participation is voluntary. All information was gathered anonymously and handled confidentially. The study design assured adequate protection of study participants and do not imply any risk for them. No reward was received by participants in return for participation59.

Instrumentation

A questionnaire was developed in the Arabic and the English languages through Google forms. The utilized scales used were translated into Arabic following the guidelines concerning the forward and backward translation. A consensus was used to resolve inconsistencies between the original and translated versions. A pilot survey was also conducted on 15 physicians, and reformulations for some questions were made throughout its course. The answers to the pilot survey were excluded from the final data of this study. The finalized anonymous, self-administered questionnaire took 10 to 13 min to be completed and consisted of three sections: (a) basic sociodemographic characteristics, (b) work-related and exposure to COVID-19 variables, and (c) the measurements.

The first section collected sociodemographic data of the participants, including gender, age, marital status, specialty, urbanicity, health status, and living conditions. It also included questions about the history of illnesses and the health status of people living with the participant. Physicians were also asked about the type of health facility where they worked. The second section covered the topic of exposure to COVID-19 in addition to work-related variables. Physicians were queried to answer whether they have worked in the frontline, treated COVID-19 patients, and been tested or diagnosed as a COVID-19 case. They were also asked if they had a family member or a colleague infected by COVID-19 and had previous experience in a pandemic/infectious disease outbreak. Of note, the term “working in the frontline” referred to physicians who reported direct contact with suspected or documented COVID-19 infected patients while previous experience with pandemic/outbreaks referred to prior work of the physicians during infectious disease outbreaks such as SARS, MERS, H1N1, or Ebola60.

The third section consisted of four validated scales to objectively assess financial well-being, threat perception, fear of COVID-19 (FOC), and burnout. The scales were used after requesting permission from their copyright owners when required.

The perceived threat (TP) and altruistic acceptance of risk questionnaire

This tool was developed by Chong61 to assess threat perception among HCWs. It consisted of ten items where nine of these items described HCWs’ perception of COVID-19 threat and one item asked for altruistic acceptance of COVID-19 risk. Since this scale was previously used among Lebanese HCWs, thus it could be used to assess this aspect among Lebanese physicians62. Ratings were given based on a five-point Likert scale from one (strongly disagree) to five (strongly agree). Responses were dichotomized into positive responses ‘agree’ or ‘strongly agree’, while ‘strongly disagree’, ‘disagree’, and ‘not sure’ were considered negative. The Cronbach alpha of this scale in this study was equal to 0.703.

The FOC scale

This tool developed by Ahorsu consisted of seven items63 and scored on a five-point Likert scale from one (strongly disagree) to five (strongly agree). The score is calculated by summing the answers and ranges from 1 to 35. Higher scores indicated a large extent of fear of COVID-19. This scale was previously used to assess the fear of COVID-19 among the Lebanese population64. In this study, the Cronbach’s alpha for this scale was 0.769.

The InCharge financial distress/financial well-being scale (IFDFW)

This tool was developed by Prawitz65 including eight items measuring the perceived financial distress/financial well-being on a linear scale from one to ten. Higher scores reflect lower financial distress and higher well-being. Of note, the IFDFW scale was used before in Lebanon in a study assessing the mental health outcomes of the COVID-19 on the Lebanese population64. For this study, the reliability of this scale was checked66 and the Cronbach’s alpha for IFDFW was 0.85.

The Arabic version of the Copenhagen Burnout scale A-CBI

The validated Arabic version of the CBI which consisted of 19 items was used67. The CBI evaluates personal-related (PB) (six items), work-related (WB) (seven items), and client-related (six items) (CB) burnout. Of note, the term “clients” referred to patients in this study. Participants were asked to rate how often they felt exhausted. Ratings were given based on a five-point Likert scale. Each item was scored from 0 to 100 (0 = never, 25 = Seldom, 0 = Sometimes, 75 = Often, 100 = Always). Of note, some questions were answered using another five-point Likert scale (to a very high degree, to a high degree, somewhat, to a low degree, to a very low degree). The mean items score was calculated per scale. A cut-off of 50 was used to assess the prevalence of burnout among physicians. A score of more than 50 is considered a high or moderate burnout level whereas a score less than 50 signifies a low burnout level or its absence. The score was valid and reliable according to many previous studies66. In this study, the Cronbach’s alpha of this scale was equal to 0.879.

Statistical analysis

The generated data through google forms were downloaded in an excel sheet, then transferred to SPSS® software (Statistical Package for Social Sciences) version 24.0 for analysis. No missing data were recorded since the response to all questions was mandatory. For descriptive analysis, frequency and percentage were used for categorical variables, and the mean and standard deviation for quantitative variables. The normality distribution of CBI items was confirmed by the calculation of skewness and kurtosis values. (< 1)68. For the bivariate analysis, to compare the means between the two groups, the Student’s T-test was used. Levene’s test was used to check the assumption of the homogeneity of variances before running a One-Way analysis of variance ANOVA to compare three groups or more.

To limit the possibility of getting a statistically significant test resulting from the run of many simultaneous independent and dependent statistical tests, post hoc analyses using Bonferroni correction were performed which sets the significance cut off at α/n. (α: error type 1 and n: number of tests)69. The correlation between continuous variables (burnout subscales and other scores used) was explored using the Spearman correlation coefficient and the gamma coefficient was calculated to explore the association between ordinal variables70. The Eta squared was used to compare means and the coefficient of variation (r-squared) to estimate the effect size of the correlations71.

The assumptions required before running the multivariable analysis were checked including the absence of multicollinearity, the residues normality, the homoscedasticity assumptions, and the linearity of the relationship. Then, four multiple linear regressions were performed using the stepwise method to identify the correlates of dependent variables (CBI subscales) in the whole sample and to reach the most parsimonious model. As for independent variables, all variables that showed a p-value < 0.2 in the bivariate analysis were introduced in the multivariable including sociodemographic, family, health, Fear of COVID-19, work-related and economics-related variables were also included. Based on the sample size, into account, the maximum number of variables allowed to be introduced in the analysis were taken into account. The R-squared and adjusted R-Squared were calculated for the full model, and the partial Eta squared for individual items. p-value < 0.05 was considered statistically significant.

To assess the interaction between the TP and the financial wellbeing (IFDFW) scales, a multivariate analysis using the General Linear Model was conducted on the same dependent variables using the enter method. The estimated marginal means were calculated for burnout among subjects according to their TP of COVID-19 and IFDFW (high/low categories). Of note, the dichotomization of the two variables (TP and IFDFW) into high and low categories was done according to the median of each scale.

Informed consent

Informed consent for participating in the study was obtained digitally through Google Forms from all subjects, and all methods were carried out in accordance with the relevant guidelines and national regulations for the Non-clinical studies. Specifically, at the beginning of the questionnaire, participants were asked whether they agree to participate in the research in order to be included in the study. Participants were also informed that their participation was voluntary and that they had the right to leave at any time without providing any explanation. No incentives were provided to the study participants.

Results

Baseline information of the participants

A total of 398 physicians participated in the survey. The majority of them were male (52.8%); married (60.1%), aged between 40 and 49 years old (43.2%), and residents of Mount Lebanon province (34.7%). Around half of participants had currently a dependent child (47.7%) or were living with the elderly (53%) or a family member with comorbidities at home (53.8%). More than two-thirds (69.85%) of surveyed physicians had a professional experience larger than 10 years and a previous experience in working in pandemics (74.12%). The highest percentage of participants were working on the frontlines (62.1%) and 51.9% of them were caring for COVID-19 cases. Only 15.3% of them had a previous history of COVID-19. However, 44.2% of the participants had a family member diagnosed with COVID-19 and 90.2% of them had a colleague diagnosed with COVID-19 (Table 1). Of note, the majority of surveyed physicians (39%) were specialized in internal medicine (Fig. 1).

Table 1 Socio-demographics characteristics of surveyed physicians (N = 398).
Full size table
Figure1
figure 1

Distribution of surveyed physicians by specialty.

Full size image

Description of the scales

CBI had a mean of 65.34 (SD = 17.39) while the values for TP scale, FOC scale, and IFDFW were 35.53 (SD = 2.88), 17.88 (SD = 1.4), and 22.85 (SD = 7.64) respectively. The normality of the scales was assumed since skewness and kurtosis were lower than 1 and the sample size was larger than 300. The used scales showed good reliability; IFDFW (α = 0.85); FOC (α = 0.769); TP (α = 0.703) and CBI (α = 0.879). The lower scores of IFDFW reported in all items of the scale reflected higher financial distress and lower well-being. The highest burnout level was shown in WB (71.5 ± 16.33) followed by PB (64.8 ± 17.32) (Table 2).

Table 2 Descriptive statistics of the scales used in the study.
Full size table

Prevalence of burnout among Lebanese physicians

Moderate and high level of burnout was detected among 90.1% of surveyed physicians, while 19.1% had a high level of burnout. PB ranked first among other burnout aspects (80.5%) with 45.8% of physicians reporting high PB levels. As for WB, it was detected in moderate and high levels among more than three-quarters of physicians (75.6%), where 60.3% exhibited a high level of WB. Moderate and high CB was found among 69.6% of participants (Fig. 2).

Figure 2
figure 2

Prevalence of burnout and its three dimensions among Lebanese physicians.

Full size image

Socio-demographic characteristics and burnout

Female gender, younger age, being single or divorced, physicians who had a dependent child, and those who live with elderly and family members with comorbidities had a significantly higher level of burnout. Similarly, all these sociodemographic variables were significantly associated with a high level of PB, WB, and CB except the age which was not significantly associated with the work-burnout dimension. The largest effect size was observed in age, marital status, and presence of a dependent child at home (Table 3).

Table 3 Association between socio-demographic characteristics and CBI subscales (N = 398).
Full size table

Economic characteristics and burnout

Surveyed physicians who have private health coverage and those who subjectively classified themselves as having a low socioeconomic status currently had significantly higher burnout in all its aspects (PB, WB, and CB). Besides, physicians who earned less than two Million Lebanese pounds per month and those who considered that pandemic or economic crisis highly impacted their monthly income showed also high burnout. However, financial well-being (FWB) was negatively correlated with high burnout. Regarding burnout, the largest effect size was seen in low economic status after the pandemic and economic crisis, major impact of the economic crisis on the income and FWB (Table 4).

Table 4 Association between economic factors and CBI subscales (N = 398).
Full size table

Occupational factors and burnout

Physicians working in public hospitals, those with limited professional experience (less than 10 years), and those who lacked a previous experience during pandemics had significantly higher levels of burnout compared to their counterparts. Furthermore, insufficient sleeping hours, extensive working hours, and physicians’ higher perception of COVID-19 impact on their work increased the overall burnout among participants. These factors had a large effect size related to the overall burnout. Similar occupational factors were associated with a high level of BP except for extensive working hours. In addition to the identified professional factors increasing burnout among physicians, working in hospitals located in urban areas had higher WB. In terms of CB, a higher level was associated with health facility type, previous pandemic experience, and extensive working hours (Table 5).

Table 5 Work characteristics and CBI subscales (N = 398).
Full size table

Exposure, perception of COVID-19 threat, fear of COVID-19, altruism, health characteristics, and burnout

Having a good health status, a history of COVID-19 infection and altruism were significantly associated with a lower level of burnout in all aspects. FOC and higher TP were correlated with higher burnout among physicians. Similarly, participants who perceived a major impact of the pandemic on their daily life and their familial relationship reported higher levels of burnout. The largest effect size was found for the TP of COVID-19, altruistic and COVID-19 impact on familial relationships. Altruism was significantly associated with a decreased burnout in all its aspects (Table 6).

Table 6 Association between COVID-19 exposure, health characteristics, COVID-19 impact, and CBI subscales (N = 398).
Full size table

Correlates of burnout and its subscales: a multivariable analysis

Higher overall burnout was associated with female gender, younger age, physician specialty, working in public hospitals, higher TP, insufficient sleeping hours, low income, extensive working hours, having a dependent child or family member with comorbidities, and limited professional experience. However, being married, financial well-being, good health, history of COVID-19, altruism, and previous pandemic experience were significantly associated with lower burnout. The full model could explain 76.1% of the overall burnout. PB was associated with younger age, female gender, having a single or divorced marital status, presence of an elderly, child at home, or family member with comorbidities. Higher TP, FOC, sleeping disturbance, extensive working hours, and low income were associated with higher PB. However, financial well-being, altruism, and good health were associated with lower PB levels. The full model could explain 67.2% of the PB. As for WB, similar factors were found positively associated with higher burnout along with the hospital’s type. The full model could explain 58.4% of the WB. In terms of CB, it was found that younger age, higher perception of threat, FOC, and low income were associated with higher CB. Similar to other aspects, altruistic and large professional experience and financial wellbeing were associated with a decreased level of CB (Table 7).

Table 7 Multivariable analyses: Correlates of CBI and its subscales.
Full size table

Interaction between TP of COVID-19 and financial wellbeing score

The multivariate analysis showed a significant interaction between the TP of COVID-19 and the financial wellbeing (IFDFW) scores on estimated marginal means of burnout. Estimated marginal means showed a significant linear increase of contrasts between the four categories of physicians: those with no TP and IFDFW (Category 0), those with TP and IFDFW (Category 1), those with TP and no IFDFW (Category 2), and those with no TP and no IFDFW (Category 3). Compared to physician with no TP and financial well-being (CBI = 58.9; PB = 60.1, WB = 63, CB = 53.5), TP of COVID-19 added some burnout (CBI = 63.9, PB = 62.8, WB = 67.5, CB = 55.9) followed by a higher increase related to financial distress (CBI = 64.1, PB = 65.7, WB = 70.5, CB = 57.1) while the highest increase in burnout was found in subjects presenting high COVID-19 threat perception of and financial distress (CBI = 65.7, PB = 67.8, WB = 73.1, CB = 59.9) (Fig. 3).

Figure 3
figure 3

Estimated marginal means of burnout and its dimensions through categories of threat perception scale (low and high) and financial wellbeing (IFDFW).

Full size image

Discussion

The COVID-19 pandemic has aggravated the levels of burnout among physicians who had to shoulder the burden of COVID-19. The present study aims to assess the level of burnout among Lebanese physicians along with how sociodemographic, occupational, economic, and pandemic-related factors affect the intensity of burnout. Besides, it aimed to explore the combined effects of the pandemic and the economic crisis on burnout. It is believed that this paper is the pioneer study in Lebanon focusing on burnout during the context of double hit and investigating factors associated with burnout and the combined effect of these crises among physicians.

Main findings

A significant burnout level was detected among physicians during these unprecedented times. A strong association was found between sociodemographic variables, occupational, economic, and exposure factors with higher levels of burnout. However, financial well-being, altruism, good health, and history of COVID-19 were significantly associated with lower levels of burnout. The analysis of the combined effect of the COVID-19 pandemic and financial wellbeing demonstrated that the presence of both TP and financial hardship significantly increased the level of burnout.

The findings of this study revealed that burnout hits more than 90% of the Lebanese physicians and around 20% suffered from a high level of burnout. Combining moderate and high levels of burnout, more than the third quarter of them expressed PB (mean = 64.8) and WB (mean = 71.5). As for CB (mean = 58.7), it was detected among 69.6% of participants. Several studies found in the literature documented burnout and its effects among physicians72,73,74,75 as well as its increasing trend of burnout during the pandemic. For example, a study reported that 45.8% of US physicians had experienced burnout29. Another study conducted among Austrian physicians showed a substantial increase of 30% in burnout rates during the pandemic compared to other studies conducted before the COVID-19 outbreak7,8,45.

In a systematic review covering 176 studies, an overall burnout rate of 48.7% was found76. Burnout syndrome was also found prevalent among 57.7% of Jordanian physicians77. Of note, the use of different tools for assessing burnout impedes the comparison of the results of this study directly with the findings of other previous studies such as the one conducted among Lebanese physicians in 201330. Therefore, it was difficult to ascertain the increasing trend of burnout among the study population. In comparison with other studies using the CBI scale whether before or after the pandemic, the study’s findings were much higher than those reported in these studies78,79. For example, a study conducted among emergency physicians in Bahrein using the CBI scale found a prevalence rate of 81.0% for PB, 69.8% for WB, and 40.5% for CB78. Another study conducted among German general practitioners showed that one-third of physicians suffered from PB symptoms, one quarter showed WB while 12% of them reported a high prevalence of CBI79. Altogether, the crippling effect on mental health revealed by the alarming prevalence of burnout among Lebanese physicians was foreseeable. It could be understood in the light of the particular Lebanese context that cumulates the traumatic effect of the COVID-1980 and the unprecedented economic crisis. Hence, urgent measures that tackle this looming epidemic of burnout are required to save an already ailing health sector.

In terms of sociodemographic factors, our findings showed that higher burnout was associated with the female gender. However, studies in the literature reported dissimilar results in terms of gender. While a number of studies reported no gender differences in terms of burnout, other studies found that females experienced more burnout compared to males2 such as McMurray et al.73 who reported that women physicians had increased odds of burnout when compared to men. Consistently with the study findings, Kannampallil et al. also found a higher prevalence of burnout amongst women during the pandemic81. This could be explained by the high exposure to risk for female physicians given their predominance in patient-facing roles, gender expectations in care, with high workloads at their homes82.

Furthermore, this study highlighted the association between younger age and a high level of burnout. Our findings were consistent with the results of a study among Hungarian general practitioners and residents which considered younger age as the strongest predictor of burnout83. Conversely, another study conducted among Portuguese physicians reported that younger age and female gender were independent determinants of burnout84. Such a result could be explained by the fact that older physicians, learned during their journey, through their day-to-day practice and their previous encounters with stressful events how to anticipate, cope, and prepare for potentially tough situations. Therefore, it could be easier for them, than younger physicians to engage in their work, adopt positive adaptation, and apply emotion management skills85. To address this issue, specific programs to prevent burnout should be designed and implemented for physicians just starting their careers, such as coping and self-care strategies.

Another important aspect of burnout, noticed in this study was that being married decreased the level of burnout. The findings of Shanafelt et al.29 supported our results concerning the presence of a partner (being married) and the decreased risk of burnout6. This could be explained that physicians who are supported or feel supported by their partners or loved ones experienced less burnout when compared to those who do not. Interestingly another study showed that spouse support decreased burnout by 40%3. Further studies were suggested to explore the association between marital status and burnout.

Remarkably, having a dependent child or having a family member with comorbidities were both associated with higher burnout levels among physicians. Our results were comparable to those reported by Koh et al. and Maunder et al. both suggest that having children is a predisposing factor to burnout4,5. However, McMurray et al.3 found that women physicians who had young children to look after reported a decrease in burnout by 40%. The higher burnout level detected among these physicians could be explained by their concerns and anxiety about transmitting the disease to their vulnerable family members86.

In terms of pandemic-related factors, a higher TP was also associated with a higher level of burnout. This could be due to the uncertainty surrounding the pandemic in terms of healthcare policy reform and compensation changes that have the potential to instigate burnout. Overall, it is well recognized that intense fear and TP when people experience physical and psychosomatic disorders lead to anxiety, burnout, and emotional exhaustion87,88,89.

In terms of economic factors, a current low socioeconomic status and income, and negative financial well-being were found associated with a higher burnout level. Of note, a previous higher socioeconomic status and a current fear of poverty were found associated with higher stress and burnout, whereas current financial wellbeing was correlated with lower burnout90,91,92. Such penetrating association in low- and middle-income countries is leading to several mental disorders93. Of note, the Lebanese physicians with savings in the country’s banks and who were unable to reclaim their money represented a typical example. Moreover, the massive depreciation in the country’s currency led to a loss of more than 80% in physicians’ income8. It was revealed that the current situation had detrimental consequences among physicians, including soaring burnout, and psychiatric illnesses94,95 in addition to an exodus of physicians who left the country searching for stability, financial wellbeing, and safety. On other hand, the association between escalating poverty and economic insecurity and stress is well known57 which in turn, can lead to burnout and demission. Since the economic crisis is expected to escalate, health facilities were in danger of laying off employees, postponing some services, or completely closing their doors.

In terms of occupational factors, our findings showed that physicians who specialized in internal medicine and infectious diseases were more prone to suffer from higher levels of burnout compared to their colleagues. The role of specialties as a contributor to burnout found in this study may be partly due to differences in exposure to COVID-19 cases as ID specialists, and internal medicine physicians such as pulmonologists and cardiologists were more involved than other physicians in the treatment of COVID-19 cases. This dissimilarity of burnout among specialties was also highlighted by a meta-analysis conducted by Lee et al.96. Besides, the findings of this study highlighted that burnout rates were highest amongst physicians involved in frontline care. However, this was anticipated since their job presented a higher risk of infection due to their direct contact with COVID-19 cases. A study conducted by Kannampallil et al.97 showed similar results concerning the higher prevalence of burnout (46.3%) reported among physicians who were exposed to COVID-19 patients compared to those who were not exposed (33.7%)14. However, there are disparities regarding the correlation between burnout and working on the frontline76. For example, Wu et al.28 found that medical staff working on the frontline had a lower level of burnout compared to those working on usual wards explaining this unexpected trend, by suggesting that frontline workers may have felt a greater sense of control over the situation.

One peculiar finding in this study was that working in public hospitals was associated with higher burnout levels. This could be understood since public hospitals were firstly designated by health authorities to treat and isolate COVID-19 patients, hence physicians working in these hospitals were more exposed to COVID-19. In the light of the deep economic collapse which lead to a shortage of funds, the government was unable alone to support hospitals with much-needed resources and supplies. This called for the support of foreign and local non-governmental aid to import essential supplies and equipment, including personal protective equipment.

Similar to other studies, our findings showed that insufficient sleeping hours and extensive working hours were associated with a higher level of burnout3. In this regard, several studies highlighted that sleep deficiency is a key risk factor for burnout among physicians18,98. With the rise of COVID-19 cases, physicians are facing intense workload, and extensive working hours, which eventually impacted physicians sleeping hours. Of note, the role of sleep disorder was found, even in normal conditions, to be associated with four-fold bigger odds of burnout17.

In addition to the above, limited work experience was associated with a higher burnout level. Consistently, a Portuguese study showed that HCWs with larger experience were less affected by burnout99. Another study conducted among physicians in Lithuania found a significant reverse relationship between work and patient burnout and length of employment100. However, previous experience during a previous pandemic or emergency was associated with a decreased level of burnout among physicians. This could be explained that previous experience provides physicians with a sense of confidence and control over the situation and lessens their worries when dealing with patients. Physicians with good health status and previous history of COVID-19 experienced a lower level of burnout. Their good health status and a history of COVID-19 could lessen their concerns about their susceptibility as a previous infection could instigate their sense of being immune naturally.

The role of altruism in decreasing the level of burnout was supported by the study findings since physicians who accepted the risk of caring for COVID-19 cases had lower burnout levels in comparison with those who are not accepting this risk. Similar results were reported by a Turkish study that found a lower level of burnout among physicians who were actively involved in the fight against COVID-19 in comparison with their counterparts who are not actively involved101.

Lastly, the combined effect of the threat of the COVID-19 pandemic and financial hardship significantly increased burnout levels among physicians. Despite the scarcity of previous studies tackling such a topic, a review supported the effect of economic uncertainty on mental health in the era of COVID-1964. The increased risk of burnout among Lebanese physicians necessitates a combined approach to addressing the stressors resulting from the pandemic and economic crisis.

Limitations

Several limitations should be acknowledged in our study. First, the study had a cross-sectional design which does not allow us to deduce causality. Selection bias is possible due to the snowball technique which limits the generalizability of the findings. The collected data was based on self-reported information which makes it prone to social desirability. Although taking into consideration of some potential confounders in the multivariable models, residual confounding is still possible. Face-to-face studies would be suggested in the future to confirm our results. Further longitudinal studies as well as following up on the burnout of Lebanese physicians would be recommended in the future to confirm our results, especially since the economic crisis escalates sharply in December 2020.

Implications for clinical practice and research

The alarming level of burnout detected among Lebanese physicians represented only the tip of the iceberg of the crisis in Lebanon. Its negative impacts that begin to effervesce with the exodus of some physicians would not be restricted to those healthcare providers but would also affect the patient’s quality of care and the healthcare organizations20. However, to date, there were no realistic evidence-based interventions and tangible measures that focused on physician burnout in Lebanon. The benefits of preventing physician burnout are not restricted to the affected individual and could also benefit the patient care as well as the overall health care system by potentially preventing physicians from leaving clinical practice. Hence, it is important to address factors identified by this study that potentially contribute to burnout among physicians in order to mitigate the long-term negative consequences through oriented strategies. However, these approaches counter Lebanese physician burnout and need to be further explored. It should empower the active involvement of the physician, at the facility level, in developing guidelines and designing contingency plans. These plans should create a supportive network and ensure physicians’ access to feedback channels as well as giving them to communicate with experts. Training on emotion management strategies should be performed to improve their preparedness for stressful situations. The importance of self-care (rest, healthy lifestyle, breaks, and sufficient sleep) should be recognized by the organization. The latter should screen regularly physicians at increased risk of personal and work-related burnout. In addition, government and health facilities should address this comorbidity among physicians through enacting proactive policies and providing critical leadership and funding for burnout prevention programs through a collaborative effort between national and institutional leadership. More studies exploring possible interventions based on physicians’ preferences and the feasibility of such interventions were recommended. The association between burnout and intention to leave clinical practice or to go abroad for clinical work would be recommended to be explored.

Conclusion

After dealing with more than a year with the COVID-19 pandemic stressors combined with an unprecedented economic collapse, Lebanese physicians reached a crisis point and the problem is expected only to get worse in absence of urgent measures. This study found a huge and serious prevalence of burnout among Lebanese physicians which called for collaborative efforts from all stakeholders in healthcare to adopt urgent measures and to implement effective strategies to enhance the physicians’ wellbeing.