The purpose of this study was to examine the psychometric properties of the Russian version of CBI-S. Following adaptation and psychometric tests, this study found that the survey was reliable and valid for assessing burnout among Russian-speaking medical students in Kazakhstan. The ICC analysis showed that the R-CBI-S had high stability within 2 weeks of the test-retest approached (mean ICC 0.81).
The internal consistencies of the four subscales were satisfactory, with all the Cronbach’s alpha values ranging from 0.874 to 0.926, and Cronbach’s alpha for R-CBI-S being 0.939. These results are slightly lower than those reported by Campos et al., with Cronbach’s alpha ranging from 0.875 to 0.931, and 0.957 for CBI-S. The results of the current study present good internal consistency values. The corrected item-total correlation values obtained for the items are relatively high, which demonstrated that the items of R-CBI-S are relatively homogeneous and are measuring the same overall construct.
R-CBI-S demonstrated satisfactory construct validity, as tested by CFA. The results indicated that most fit indices were in acceptable ranges. Sufficiency of the model was demonstrated by Bartlett’s test of sphericity and KMO measure.
The criterion validity of the survey was tested on all participants by adding the question “How burned out do you feel from your course?” as the gold standard for burnout. This test showed that the diagnostic efficacy of the survey was high, with an AUC in excess of 0.934. A higher value of SRB AUC than total scale AUC (0.947 and 0.934 respectively) can be explained by the similarity of an additional question with item 13 in the SRB subscale (“Do you feel burned out because of your studies?”). These results demonstrate a high capacity for detecting the students affected by burnout syndrome.
According to the result of psychometric properties testing, the present study shows that the R-CBI-S is a reliable and valid measure for assessing burnout syndrome among Russian-speaking medical students in Kazakhstan. This is a novel finding in that there are no other existing studies that describe burnout and related factors among medical students, or even non-medical students, in Kazakhstan, as a consequence of which authors compared the results by referring to foreign sources.
This study aimed to determine the prevalence of burnout and to identify factors associated with burnout among medical students at AMU. In one systematic review of 24 studies of 17,431 pre-residency medical students, overall burnout prevalence across the entire student population was estimated to be 44.2% (33.4%–55.0%) [11]. In the current study prevalence of burnout syndrome was 28% (broken down by severity: 25.4% moderate, 2.5% high, and 0.1% severe). Similar medical student burnout rates were found in studies from South and Central America (26%) and Europe (27.5%) [11]. This comparison cannot be reliably interpreted since these countries have different medical education system.
Our study also revealed that while prevalence of burnout was not dependent on gender, women showed significantly higher scores of personal burnout (p < 0.05). In a systematic review, Frajerman et al. found no significant association between gender and burnout [11]. In contrast, two systematic reviews of Chinese and Brazilian medical schools found higher levels of burnout in males [12,13]. The burnout prevalence among medical students was also different according to the year of study (p < 0.001); burnout rate were higher among students in the second year relative to students in the other years.
Many students during the training period take on additional work for various reasons: to make extra money or, try to improve their practical skills by working as junior medical personnel in clinics. Students doing additional work outside of the required curriculum represented 24% of our respondents. These students were found to have significantly higher prevalence of personal burnout (p < 0.05) compared to students who do not have any job.
Some prior research has found that burnout was not associated with different types of extracurricular activities [14,15]. However, it is known that extracurricular activities like those involving music and physical exercise may reduce burnout and other mental health problems [16,17]. In the current study, authors wanted to study the impact of extracurricular activities such as participating in scientific clubs and student societies and volunteering on burnout. We found that students with extracurricular activity had a higher level of colleague-related burnout comparison to students without it (p < 0.05).
Among US medical students Dyrbye et al. found that burnout can contribute to suicidal ideation, while recovery from burnout decreases the prevalence of suicidal ideation [18]. Data from the systematic review of Ishak et al. pointed to an association between burnout in medical students and suicidal ideation [19]. Our study found that students with suicidal ideation have more expressed burnout in total and across all studied dimensions (p < 0.001). Also, students who had suicidal ideation were almost eight (7.8) times more likely to develop burnout than those who did not have suicidal ideation.
Burnout is known to be associated with serious thoughts of dropping out [20]. In the current study we found the same association: students who report thoughts of dropping out from medical school show significantly higher burnout severity and prevalence (p < 0.001). Moreover, we found that students who are not satisfied with their academic performance have more pronounced burnout syndrome (p < 0.001). Suicidal ideation and thoughts of dropping out can be described as the manifestation of distress [21].
Salam et al. reported strong relations between stress and interpersonal relationship problems among medical students [22]. Participants of current study indicated the presence of relationship problems with family and friends more prone to burnout.
Substance use seems to play a meaningful role in burnout. 64 out of 480 students (13%) in the present study indicated that they smoke cigarettes. The study found that these students had higher burnout prevalence. Results obtained by Cecil, et al. indicated that being an ex-smoker was significantly predictive of higher emotional exhaustion scores [23]. Also, a study from Japan concluded that the mental health status of dental students among regular smokers was better than that of non-current smokers [24]. One longitudinal survey among German and Chinese students showed healthy lifestyle choices like choosing not to smoke are related to improvements in mental health over a one-year period [25]. On the other hand, a study from South-West Ethiopia indicates that smoking cigarettes was significantly associated with common mental disorders [26]. Alcohol use is also of note in the burnout research. Jackson et al. found that alcohol abuse or dependence was more common among medical students with burnout, high emotional exhaustion, and high levels of depersonalization [27]. Research among British medical students did not find significant correlations between any of the personality variables and alcohol consumption [28]. Cecil et al. reported that increased alcohol binge scores were significantly associated with higher personal accomplishment scores [23]. In this study, we found that higher burnout prevalence is found in students who drink alcohol (p < 0.001).
The degree to which students feel to have been empowered to make their own decisions regarding studies and career also seems to influence susceptibility to burnout. We note that the culture of Kazakhstan is distinct from that of more individualistic nations’ cultures that heavily promote independence and autonomy of young adults in that it is common for younger generations to strictly obey and yield to the opinions of older relatives. This influence often extends to decisions about which specialty children should pursue at university. In this work, it was found that when admission to a medical school is the decision of the student him/herself, or that he/she believes that it was, then during the study period, the student is likely to reported significantly lower levels of burnout (p < 0.05) than student counterparts who chose medical education for other reasons, including the urging of parents and close relatives. Thus we determined that an independent decision of the student to enter a medical school was a strong predictor for burnout development.
Finally, dwelling type plays a meaningful role. We found that compared to medical students who were not burned out, those who screened positive for burnout were more likely to be living at home or rented a house than in student housing. This might suggest there compounding network effects of burnout – that students living among other students might inherit be more likely to burn out themselves. It might also suggest that those living in non-student housing are more likely to have stronger support systems – family, relationship partners, or friends close enough to live with – that function as a burnout preventative.
Limitations
There are several limitations inherent in this study. First, not all means of validation analysis were employed during the study. Only one tool was used to measure burnout. Future studies could extend this work to comparisons using other well-established measures of burnout. Also, the study was conducted in Astana Medical University, and as such the results might be generalizable to other medical universities in Kazakhstan. Future work could extend this work to other groups of students in different fields across Kazakhstan, or non-student professional populations. Finally, this study stops short of validating the burnout scores with respect to their relationship to other determinants of quality of life, and with professional and life outcomes post-medical school. This limitation provides another fruitful potential avenue for future extension of this work.