key: cord-0258366-7haa3z2g authors: Moser, C. M.; Tietbohl-Santos, B.; Arenas, D. L.; Xavier, A. Z.; Ornell, F.; Borges, R. B.; Hauck, S. title: Validation of the Brazilian Portuguese version of The Copenhagen Burnout Inventory (CBI) in healthcare professionals date: 2020-09-23 journal: nan DOI: 10.1101/2020.09.21.20198408 sha: ed9bf3b25826ea6184b757cd1e90d94cb185bde1 doc_id: 258366 cord_uid: 7haa3z2g The objective of this study was the validation of the Brazilian Portuguese version of The Copenhagen Burnout Inventory (CBI). The study sample consisted of 1.256 Brazilian health professionals. Data were collected during one month (May-2020 to June-2020) through an online self-administered questionnaire, including the CBI and the Patient Health Questionnaire 9 (PHQ-9) to evaluate depressive symptoms. All consistency measures showed values > 0.90. The split-half correlation values with Spearman-Brown reliability were higher than 0.8. The parallel analysis suggested two factors: the items of CBI's personal burnout (PB) and work-related burnout (WB) sub-dimensions that were associated with factor 1 (personal and work-related fatigue and exhaustion) and the items of CBI's client-related burnout (CB) sub-dimension were associated with factor 2 (exhaustion that emerges while working with people). The correlation between the PHQ-9 score and factor 1 was 0.76 (0.73; 0.79) and with factor 2, 0.49 (0.43; 0.55). Our study provides a valid and reliable Brazilian Portuguese version of CBI, encouraging the expansion of the burnout research field in our country. Burnout Syndrome (BS) is an emerging condition worldwide that can emerge in response to constant and prolonged stressors in the work environment, generating physical, emotional and mental exhaustion, leading to reduced performance. The devastating effects of burnout and its clinical significance have been vastly reported in the literature and health professionals are among the most affected 1 . BS has been associated with a higher frequency of medical errors, suboptimal care of patients, reduced empathy in general, absenteeism, reduced quality of life and higher costs for health care systems 2 . However, there is not a consensus about BS diagnostic criteria and about standardized measurement tools to assess it. In fact, the substantial disagreement on what constitutes burnout precludes conclusions even about the phenomena's prevalence 1, 3 . The instrument most commonly used to evaluate burnout is the Maslach Burnout Inventory (MBI) 4 . It consists of a three-factor questionnaire that assesses emotional exhaustion, depersonalization, and low sense of personal accomplishment. Some authors have pointed out that the three dimensions assessed by MBI are not weighted equally, and that there is no accuracy between the three subscales. Moreover, depersonalization could be regarded as an exhaustion coping mechanism, whilst lower personal accomplishment could be viewed as a consequence of exhaustion, rather than a dimension of burnout itself 5, 6 . In this context, The Copenhagen Burnout Inventory (CBI) was developed by Kristensen and colleagues to be a more straightforward measurement that considers fatigue and exhaustion as the core constructs of burnout 6 . The CBI differentiates three life domains from which emotional exhaustion may arise: personal, work-related, and client-related. Personal burnout is the degree of physical and psychological exhaustion one can experience, unrelated to their occupation. Work-related burnout is the degree to which physical and psychological exhaustion is perceived in relation to work activities. Client-related burnout is the level of exhaustion that stems from the professional relationship with clients. The CBI assesses burnout in a less complex manner then the MBI. Moreover, the CBI is a public domain tool that assesses the same overall construct in different contexts, as opposed to mixing burnout and its consequences in the same construct as the MBI is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 23, 2020. . https://doi.org/10.1101/2020.09.21.20198408 doi: medRxiv preprint does. Also, a greater range of occupations can be evaluated using the CBI, in contrast with the MBI, which evaluates only the human services sector 6 . The CBI has been translated, validated, and used in many countries and contexts across the world with a growing evidence base of good psychometric properties for measuring occupational burnout [7] [8] [9] . Therefore, the aim of this study was to validate a CBI Brazilian version. This is an instrument validation study. The cross-cultural adaptation and translation process of CBI to Brazilian Portuguese involved a series of standardized steps according to the protocol developed by our research group 10 The total sample was composed of 1.256 Brazilian health professionals that work in Brazil: 32,32% were physicians (n=406), 20.22% were nursing technicians (n=254), 13 .61% were nurses (n=171),12.18% were psychologists (n=153) and 21,67% were other health workers (n=272). Subjects with incomplete or missing questionnaires were excluded. Level of burnout was assessed with The Copenhagen Burnout Inventory (CBI) (Supplemental Material 1), a 19-item questionnaire that measures three burnout sub- is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 23, 2020. . https://doi.org/10.1101/2020.09.21.20198408 doi: medRxiv preprint dimensions: personal burnout (6 items), work-related burnout (7 items), and client-related burnout (6 items). It is a self-reported questionnaire answered by a Likert scale of five points: "Always" or "To a very high degree" (100 points), "Often" or "To a high degree" (75 points), "Sometimes" or "Somewhat" (50 points), "Seldom" or "To a low degree" (25 points) and "Never/almost never" or "To a very low degree" (0 points). The CBI items in each subscale are summed and averaged to obtain the scores. The higher the scores, the higher the level of burnout 6 . Depressive symptoms were assessed with Brazilian version of Patient Health Questionnaire 9 (PHQ-9), a widely used 9-item questionnaire for screening of major depressive episodes. The frequency of each symptom in the past two weeks is assessed on a Likert scale from 0 ("not at all") to 3 ("nearly every day") 13 . The reliability was assessed using Cronbach's alpha and split-half correlation with Spearman-Brown reliability. The suitability of data for factorisation was performed by the Kaiser-Meyer-Olkin measure and Bartlett's test of sphericity. Construct validity was established by exploratory factor analysis with oblique rotation. The number of factors was determined using the method of parallel analysis. All of these analyzes were performed using the polychoric correlation between the items. The concurrent criterion validity was analyzed by assessing Spearman's correlation coefficient between PHQ-9 score. The analyzes were performed in the R program using the psych version 1.9.12 and multicon version 1.6 packages. Where necessary, a significance level of 5% was adopted [14] [15] [16] . The Kaiser-Meyer-Olkin measure was 0.95 and Bartlett's test of sphericity rejected the null hypothesis (p <0.001). The parallel analysis suggested two factors. Table 1 shows the factorial loads: the items of CBI's personal burnout (PB) and work-related burnout (WB) that were associated with factor 1 (personal and work-related fatigue and exhaustion) and the items of client-related burnout (CB) were associated with factor 2 (exhaustion that emerges while working with people). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 23, 2020. . https://doi.org/10.1101/2020.09.21.20198408 doi: medRxiv preprint All consistency measures showed values > 0.90. The split-half correlation values with Spearman-Brown reliability were higher than 0.8 ( Table 2 ). The correlation between the PHQ score and factor 1 was 0.76 (0.73; 0.79) and with factor 2 was 0.49 (0.43; 0.55). The CBI's Brazilian Portuguese version developed by our group had an excellent internal consistency reliability (Cronbach alpha coefficient >0.9) and an adequate concurrent validity with the PHQ-9 (moderate to strong positive correlation with depressive symptoms) in healthcare professionals. However, instead of the 3 factor loads of the original instrument, our analysis suggested 2 factor loads. This is in line with other CBI validation studies that pointed out a good discriminant validity between client-related burnout and both work-related and personal related burnout, but did not establish a satisfactory discriminant validity between personal and work-related burnout [7] [8] [9] . These findings can be explained by an essential overlap between the two constructs that assess the fatigue and exhaustion dimensions of Burnout Syndrome. Our results may contribute to the debate regarding the BS construct. In fact, through our findings, one could hypothesize that the MBI's exhaustion and depersonalization dimensions could correspond respectively to the PB + WB and the CB CBI factors, at least partially. Indeed, the MBI depersonalization dimension refers to the effect of the BS on the client-related relationship. Further studies should explore this hypothesis, aiming at clarifying the BS criteria which would be of great value to this field of research. Our study has some limitations. The sample was comprised only of health professionals; hence the instrument must be validated in other populations. Also, other methods of analysis may be applied in the future, including test-retest reliability. However, a public domain tool that evaluates burnout is extremely needed in our milieu, even more considering the consequences of the COVID-19 pandemic. In this sense, our study provides a valid, reliable, and accessible Brazilian Portuguese version of the CBI that was produced through a very rigorous transcultural adaptation and translation process. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 23, 2020. . https://doi.org/10.1101/2020.09.21.20198408 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 23, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 23, 2020. . https://doi.org/10.1101/2020.09.21.20198408 doi: medRxiv preprint Physician Stress and Burnout. The American journal of medicine Understanding the burnout experience: recent research and its implications for psychiatry. World psychiatry : official journal of the World Psychiatric Association (WPA) Reflections on the study of burnout The Copenhagen Burnout Inventory: A new tool for the assessment of burnout Investigating validity evidence of the Malay translation of the Copenhagen Burnout Inventory Psychometric properties of the Copenhagen Burnout Inventory-Chinese version Translation and validation of the Copenhagen Burnout Inventory amongst Greek doctors Transcultural adaptation and translation process of the instruments Sick Control One Stone Fat Food questionnaire (SCOFF), Eating Disorder Examination Questionnaire Clinical Impairment Assessment questionnaire (CIA) to Brazilian Portuguese Principles of Good Practice for the Translation and Cultural Adaptation Process for Patient-Reported Outcomes (PRO) Measures: report of the ISPOR Task Force for Translation and Cultural Adaptation Literature review of methods to translate health-related quality of life questionnaires for use in multinational clinical trials Sensitivity and specificity of the Patient Health Questionnaire-9 (PHQ-9) among adults from the general population R: A language and environment for statistical computing Procedures for Personality and Psychological Research (Software multicon: Multivariate Constructs. R package version 1.6. (Software) is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprintThe copyright holder for this this version posted September 23, 2020. . https://doi.org/10.1101/2020.09.21.20198408 doi: medRxiv preprint It is made available under a perpetuity.is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprintThe copyright holder for this this version posted September 23, 2020. . https://doi.org/10.1101/2020.09.21.20198408 doi: medRxiv preprint