Effect of soft skills and emotional intelligence of health-care professionals on burnout: a Lebanese cross-sectional study / Effekte von Soft Skills und emotionaler Intelligenz auf Burnout von Fachkräften im Gesundheitswesen: eine Querschnittsstudie aus dem Libanon

A cross-sectional study was conducted among working health-care professionals all over Lebanon for a period of 3 months starting from March, when our first participant was recruited, till June 2021, when our data collection ended.

The questionnaire included three main parts and required 7–10 min to complete. It was first written in English and then translated to Arabic in order to target a larger number of participants.

The first part included sociodemographic questions such as those on age, gender, residence, profession, and working region. The second part aimed at work conditions such as working schedule, work type, years of working experience, and job satisfaction. The third part of the questionnaire included the Maslach Burnout Inventory for Health Services Survey (MBI-HSS) and the Trait Meta-Mood Scale (TMMS-24) to assess emotional intelligence, as well as a soft skill assessment scale using a five-point Likert scale ranging from strongly disagree (=1) to strongly agree (=5).

All the scales used were already developed and validated. The TMMS-24 was used to assess emotional intelligence. It is a version of the Trait-Meta Mood Scale (TMMS), which was originally developed by Salovey et al. (1995) and reduced and validated by Fernández-Berrocal et al. (2004); the scale had a Cronbach α = 0.929. Burnout was assessed using the MBI-HSS, which comprises 22 items regrouped into three subscales: EE, DP, and personal accomplishment (PA) (Ibtissam et al., 2012); it had a Cronbach α = 0.842. The soft skills scale was adapted from Nikitina and Furuoka (2012); it had a Cronbach α = 0.972.

The study was carried out via an online survey targeting health-care professionals from all the Lebanese districts. The data collection lasted approximately 4 months during which a questionnaire, developed through a Google form, based on our literature search, was sent and completed online via social networks using the snowball technique.

The main inclusion criteria were: (1) being a health-care professional; (2) currently working; (3) having a minimum working experience of 2 years; and (4) providing their written consent to participate in the study. Of the 345 contacted participants, 21 were excluded, which led to a total of 324 participants randomly recruited from all over Lebanon, including physicians, nurses, pharmacists, and other health-care professionals. Among the excluded participants, five were not health-care professionals and nine had not been working for 2 years or more, of which one was not currently employed and others were duplicates.

We calculated the minimum required sample size using Epi Info version 7.2.4.0. Based on a systematic review conducted by Parola et al. (2017), we found that the prevalence of burnout among health-care professionals is estimated at 17.3%. With an acceptable margin of error of 5%, a design effect and a cluster of 1, and a 95% confidence interval (CI), our estimated minimum required sample size was 220 health-care professionals.

Taking the fact that this is an observational study, the Lebanese University ethics committee waived the need for approval, provided that consent is obtained from participants and their confidentiality and anonymity are maintained.

SPSS 25 for Windows was used for analyzing the data. Regarding the bivariate analysis, T test, Mann–Whitney test, analysis of variance (ANOVA), and Kruskal–Wallis tests were used to compare the means, whenever the association between a qualitative and a quantitative variable was tested.

Pearson and Spearman correlations were used to test the association between two quantitative variables. P-value <0.05 was considered as significant, and the CI was set at 95%. A Cronbach alpha was recorded for reliability analysis of all the scales.

A linear regression was used to assess the effect of soft skills and emotional intelligence on burnout. Variables with a P-value <0.2 were included in the multivariable analyses.

A total of 345 health-care workers participated in this study; of them, 324 were included, involving 45.1% nurses, 32.1% pharmacists, and 8.3% physicians. The mean age of the participants was 30.95 ± 8.69 years, and the majority of them were females (80.2%). As for their region of residence and work, the majority of them were from Mount Lebanon (see Table 1).

The soft skills scale was not normally distributed, with a mean ± SD of 138.59 ± 18.33.

Soft skills were significantly associated with profession (P = 0.008), age, working experience in years, emotional intelligence, and burnout (P < 0.001). A positive, yet weak correlation was found between soft skills and emotional intelligence (R = 0.382). As for burnout, a negative, yet weak correlation was found with soft skills (R = −0.278; see Table 2).

Emotional intelligence was normally distributed with a mean ± SD of 87.00 ± 14.99. Emotional intelligence was significantly associated with soft skills (R = 0.382) and burnout (R = −0.169; see Table 2).

The burnout scale was normally distributed with a mean ± SD of 59.04 ± 11.61.

Burnout was significantly associated with profession (P = 0.010), region of work (P = 0.034), job satisfaction (P < 0.001), soft skills (R = −0.278), and emotional intelligence (R = −0.169; see Table 2).

Based on the bivariate analysis, variables with a P-value <0.200 were included in the multivariate analysis in order to predict burnout. These variables were soft skills, emotional intelligence, job satisfaction, profession, work region, region of residence, and gender (see Table 3).

Higher burnout was associated with lower soft skills (β = −0.137) and lower job satisfaction (β = 8.064). Dentists, radiologists, midwives, nutritionists, psychotherapists, and speech therapists (grouped together in the variable “other professions”) had lower burnout levels than nurses (β = −4.595). As for burnout among pharmacists and physicians, no significant difference was found compared to nurses (P = 0.283 and 0.631, respectively). Moreover, it is worth noting that people working i n Baalbek, Akkar, Beqaa, North and South had lower burnout levels compared to those working in Beirut (β = −9.015). As for emotional intelligence, no statistically significant association was found with burnout (P = 0.116), as well as with gender and region of residence (see Table 3).

This study aimed at testing the effect of soft skills and emotional intelligence on burnout among health-care professionals in Lebanon. The main findings showed that better soft skills lead to lower burnout. It is worth noting that the bivariate analysis showed a significant association between burnout and emotional intelligence. These results are in line with previous findings which showed that health-care professionals with higher emotional intelligence experienced less EE and DP (Ünal et al., 2014). Another study conducted in Tehran showed that emotional intelligence was negatively associated with EE (Mansoor et al., 2011).

The health and political situation that Lebanon is facing might demotivate health professionals, leading to high levels of job burnout (Harari et al., 2017). In fact, based on previous literature, physicians and nurses are more likely to experience burnout. A meta-analysis conducted by Gómez-Urquizaetal et al. (2017) showed that 30% of emergency nurses experienced at least one of the three Maslach Burnout Inventory subscales. This can be explained by the fact that nurses are in direct contact with the patient, which automatically puts them under pressure, in addition to their work schedule and work load.

As for physicians and medical students, a study conducted in the USA showed that 50%–60% had at least one of the burnout symptoms (Rothenberger et al., 2017).

In the Arab countries such as Saudi Arabia, Jordan, and Lebanon, a systematic review showed that high levels of burnout were reported among nurses (Al-Turki et al., 2010. Elbarazi et al., 2017; Ibtissam et al., 2012).

Furthermore, a systematic review showed that 20%–81% of health-care workers experience high levels of EE, 9.2%–80% experience high DP levels, while 13.3%–85.8% have low PA levels (Lahoud et al., 2019). In Lebanon, similar results were found in a study conducted among nurses where job satisfaction was significantly associated with burnout and nurses experienced moderate burnout levels (Ibtissam et al., 2012).

In addition, a positive association was found between soft skills and emotional intelligence. This replicates other published data regarding the effect of soft skills such as leadership, communication, and job satisfaction on emotional intelligence (Brackett et al., 2011; Dacre et al., 2012). Health-care workers with higher soft skills have better self-confidence and improved interpersonal, leadership, and communication skills, which improves their capacity to manage their social and personal emotions, which, in turn, eventually improves their emotional intelligence (Harari et al., 2017). According to previous literature, nurses improve their mental health, communication capability, and job productivity after undergoing a communication skills training. Also, 60% of oncology physicians who attended communication skills workshops had the sense of PA (Rothenberger et al., 2017; Elbarazi et al., 2017; Al-Turki et al., 2010).

Finally, this study also showed that health-care professionals who were satisfied with their job had a significantly lower job burnout. Also, people working in Beirut had a higher burnout level comparing to those who work in Baalbek, Akkar, Beqaa, North and South. This may be related to the stress experienced by the health-care professionals due to the overpopulation in Beirut. Further studies are suggested to explain these contextual findings.

Since this study confirmed the impact of soft skills and emotional intelligence on burnout among health-care professionals in Lebanon, many interventions can be suggested in order to reduce burnout levels.

This will consequently reduce work fatigue and improve their mental health and physical well-being. Thus, training health-care professionals during their education will improve their soft skills and emotional intelligence, and consequently improve their behavior in stressful situations and reduce their job burnout. Subsequently, during their work journey, many interventions can be addressed, including reducing their working hours, increasing their financial income, and providing them with the needed psychological support (Rahme et al., 2020).

The limitations of this study are as follows:

First, regarding participants, the sample mainly included nurses and pharmacists with no physicians and other health-care professionals. Moreover, the majority of the participants were from Beirut and Mount Lebanon. This might be related to the background of the researchers, in addition to the fact that physicians were not cooperative. In addition, due to the economic situation in Lebanon, many health workers were unemployed, which affected our sample size. As for the gender distribution, the majority were females, which reflects the demographic characteristics of health workers in Lebanon. Second, since we used the snowball technique, we could not know the number of health workers that received the questionnaire, and thus, we could not calculate the response rate. These factors collectively affected the representativeness of our sample.

In addition, many factors were not addressed during data collection, such as financial situation and income, psychological support, and antipsychotics consumption (Fond et al., 2019).

A volunteer bias might thus be present since it is an observational study and participants had the option to refuse participation. In addition, a non-response bias might as well occur, which affects the precision of the study by reducing the sample size.

Third, the scales used in this study are not validated in Lebanon, which might cause some non-differential information bias. Furthermore, the collected data was based on self-reported measures, which may lead to an information bias.

Despite the previously mentioned limitations, our study had many strengths. In fact, the questionnaire was developed in two languages, targeting a larger number of participants. Moreover, it mainly included short questions and the scales were frequently used in the reviewed literature and were validated outside Lebanon. Regarding the minimum required sample size, it was around 220 participants, while our study included 324 health-care professionals, which increased the precision of the outcomes and might have ensured a good representativeness of the Lebanese health-care professionals, especially nurses and pharmacists, thus improving the generalizability of our results. Finally, additional research should be conducted using other methods in order to support our findings. The authors assumed that the obtained results are original and similar to other recent studies.