Workplace factors and burnout in Brazilian anaesthesiologists
Description
There is a high prevalence of burnout among anesthesiologists, which may affect their mental health and safety in clinical anesthesia practice. Despite this, there is limited knowledge about workplace factors associated with burnout among anesthesiologists, and there is a lack of interventional studies focusing on the environment. Therefore, we designed a cross-sectional study of Brazilian anesthesiologists to assess workplace factors related to burnout syndrome and its dimensions. The survey included sample characteristics, the Maslach Burnout Inventory - Human Services Survey for Medical Personnel, and a 'job difficulty score.' Bivariate and multivariate Poisson regression models calculated the prevalence ratios (PR) and assessed the relationship between variables. The response rate was 11.11% (973/8755). Burnout prevalence was 7.2%. The prevalence rates of high emotional exhaustion, high depersonalization, and low personal accomplishment were 28%, 10.9%, and 17.8%, respectively. Burnout was associated with strained relationships (PR 3.05, 1.64-5.68, p= 0.00), feeling disrespected at work (PR 2.45, 1.37-4.40, p= 0.00) and scoring at the 'job difficulty score' (PR 1.26, 1.16-1.37, p= 0.00). Emotional exhaustion was linked to productivity pressure (PR 1.34, 1.01-1.78, p= 0.46), working sick because of staff shortages (PR 1.67, 1.22-2.30, p= 0.00), moral harassment (PR 1.54, 1.15-2.05, p= 0.00), strained relationships (PR 2.09, 1.55-2.84, p= 0.00), conflict of values (PR 1.42, 1.05-1.92, p= 0.02), managing patients' airways during the COVID-19 pandemic (PR 1.52, 1.06-2.18, p= 0.02) and scoring at the 'job difficulty score' (PR 1.15, 1.10-1.2, p= 0.00). Depersonalization was related to strained relationships (PR 2.30, 1.41-3.78, p= 0.00), feeling disrespected at work (PR 1.87, 1.16-3.02, p= 0.01), and scoring at the 'job difficulty score' (PR 1.21, 1.13-1.29, p= 0.00). The low personal accomplishment related to not having access to decisions that affect the workplace (PR 1.27, 1.02-1.57, p= 0.02), working part-time in public and part-time in private hospitals (PR 0.78, 0.62-0.98, p= 0.03) and scoring the "job difficulty score" (PR 1.26, 1.16-1.37, p= 0.00). Overall, the workplace has distinct effects on each burnout component, and mapping this information will allow for the design of customized interventional studies in anesthesia.
Files
Steps to reproduce
The Distance Education and Health Information Technologies Center (NEAD.TIS) from the São Paulo State University (UNESP) hosted the four-part survey on LimeSurvey. The link was sent to all Brazilian anesthesiologists affiliated with the Brazilian Society of Anesthesiology from September/ 2020 to March/2021. This document contains all the data collected. Any participants with missing data were excluded from the final analysis. The first part included an introductory message explaining the study’s aims, potential risks and benefits, and the researchers’ contact information. The second part consisted of four questions regarding participants’ baseline characteristics. The third part was a 27-item questionnaire on workplace conditions. The questions outlined the workplace characteristics of our sample and the challenges presented by such an environment. We took questions 7–24 from the third section of the survey to create a composite variable called the ‘job difficulty score.’ We transformed the answers to these questions from categorical to binary: when an answer expressed the worst possible workplace scenario, it received a score of 1, otherwise a 0. Because the ‘job difficulty score’ consisted of 18 questions, the worst score was 18, and the best score was 0. The greater the score, the higher the number of difficult situations faced in the workplace. The punctuation in this score was also related to burnout in our analysis. The fourth section was the Maslach Burnout Inventory—Human Services Survey for Medical Personnel MBI-HSS (MP), which was included in this final part of the online survey after Mind Garden Inc. granted permission to include it in the fourth part of the online survey. To analyze data, we adopted a bivariate Poisson regression model as a screening test to examine the relation between workplace and personal factors with each burnout dimension and with burnout syndrome diagnosis (high Emotional Exhaustion + high Depersonalization + low Personal Accomplishment). We then analyzed variables with a p-value < 0.2 using a multivariate Poisson regression model. Afterward, we analyzed only the variables that had passed the screening test. We then calculated the prevalence ratio, 95% CI, and p-value using a multivariate Poisson regression model. We considered associations statistically significant for the final analysis if p < 0.05. We used SPSS 22 (IBM Corp.) to perform the analysis.