Psychiatric symptoms, burnout and associated factors in psychiatry residents

Abstract Introduction Mental health in training physicians is a growing issue. The aim of this study was to investigate emotional distress in psychiatry residents. Method This web-based survey evaluated 115 (62%) psychiatry residents in training in the Brazilian State of Rio Grande do Sul. The DSM-5 Self-Rated Level 1 Cross-Cutting Symptom Measure-Adult, the Patient Health Questionnaire-2, the Alcohol Use Disorders Identification Test-concise, and the Maslach Burnout Inventory were all administered. Linear regression models were estimated with burnout dimensions as dependent variables. Result Positive screening rates were 53% for anxiety, 35.7% for somatization, 16.5% for depression, and 7% for suicidal ideation. Half of the male residents were at risk of alcohol abuse and dependence. Regarding burnout, 60% met criteria for emotional exhaustion, 54.8% for depersonalization, and 33% for low personal accomplishment. The most consistent risk factors were the nature of the relationships with preceptors, relations to the institutions themselves, age, and the quality of relationships with family. Conclusion Besides disconcerting rates of psychiatric symptoms, the study revealed that characteristics of the workplace (i.e., the nature of relationships with preceptors and relations to the institution) can be regarded as potential targets for development of interventions aimed at improving mental health during training periods.


Introduction
Mental health in physicians is an emergent issue nowadays and the occurrence of burnout is a growing concern; often referred to as an epidemic phenomenon in the literature. 1,2 In 2019, 44-47% of US physicians described themselves as feeling burned out or at least reported symptoms of burnout. 3 America. They found that, on average, the literature suggests that burnout and mental health concerns affect 30-60% of all physicians and residents. There was some overlap among papers discussing burnout, depression, and suicidal ideation, suggesting that work-related stress may lead to the emergence of more serious mental health problems as well as addiction and substance abuse. Residency training was shown to produce the highest rates of burnout.
In addition, they found that papers discussing causes of deterioration of mental health in physicians (20%) and prevalence of mental illness (16%) were the least common.
Burnout is a syndrome that includes emotional exhaustion, depersonalization, and low sense of personal accomplishment. Emotional exhaustion (EE) is described as lack of enthusiasm and energy, leading to a feeling of resource depletion. Depersonalization (DP) is defined as emotional insensitivity, characterized by disillusionment with the service provided, culminating in dehumanization and impersonal treatment of patients and colleagues. Low sense of personal accomplishment (PA) at work refers to a sense of inadequacy and low self-esteem connected to a belief that professional goals have not been met. 6,7 In 2018, the World Health Organization (WHO) recognized burnout as an occupational phenomenon and included it in the 11th Revision of the International Classification of Diseases (ICD-11). 8 Medical training can be associated with uncertainties about the future, feelings of insecurity, high levels of responsibility, and high workload. Current studies show that the prevalence of burnout in residents is about 25-75%, varying by specialty, country, and methods of measurement. 9,10 Research investigating psychiatry residents suggests that there is a 23-36% prevalence of burnout [11][12][13] and reveals associations with various demographic, learner, and workplace factors. These include non-parental status, being married, increased workload, insufficient rest, lack of supervision at work, being in early years of training, lower priority of psychiatry as career choice, decreased empathic capacity, poor coping skills, increased medical errors, more stressors, and low self-efficacy. 10

Methods
In this cross-sectional study, all psychiatry residents in training in the Brazilian State of Rio Grande do Sul (n = 185) were invited to participate over a period of one month at the end of 2018. Data were collected with an online questionnaire sent by e-mail. We chose an electronic questionnaire because of its ease of response, and because it has the potential advantage of enhancing reliability by augmenting the perception of anonymity.
Subjects could only access the questionnaire if they had agreed to the online Informed Consent Form. After completion, the questionnaire provided telephone and electronic contact information for suicide prevention and support centers located in Brazil. The study was approved by the Hospital de Clínicas de Porto Alegre ethics committee (Porto Alegre, Brazil) (protocol CAAE 70231617.6.0000.5327).

Survey instruments
The online questionnaire included questions regarding sociodemographic data, personal information, career status, workload, and mental health variables, including current psychiatric treatment, harassment, discrimination, and abuse at the workplace, alcohol and drug use, sleep patterns, and quality of relationships with family and friends. There is an important debate in the literature regarding MBI cut-off points. Therefore, we defined the prevalence of each dimension using the most common cutoffs  21 The WEEI's total and dimension scores and the environment category (i.e., healthy, risky, or toxic) were all related to burnout symptoms. 22 The DSM-5 Self-Rated Level 1 Cross-Cutting Symptom Measure-Adult was used to assess presence of psychiatric symptoms. This is a general screening measure for the main DSM 5 diagnostic categories that has been validated in Brazilian Portuguese. 23 The Patient Health Questionnaire-2 (PHQ-2) was administered to assess depressive symptoms. The The AUDIT-C was used to evaluate alcohol use. 25 This is a 3-item screening instrument that can help to identify persons who are high-risk drinkers or have active alcohol use disorder (including alcohol abuse or dependence). It is a modified version of the 10 questions AUDIT instrument. 26 The AUDIT-C has sensitivity of 79% and specificity of 56% in men (score ≥ 4) and sensitivity of 80% and specificity of 87% in women (score ≥ 3) for identifying patients with active alcohol abuse or dependence. 24 For men, scores of 0 to 3 were considered low risk; from 4 to 5 points, moderate risk; from 6 to 7 points, high risk, and from 8 to 12 points, severe risk. For women, scores of 0 to 2 were considered low risk; from 3 to 5 points, moderate risk; from 6 to 7 points, high risk, and from 8 to 12 points, severe risk. The AUDIT-C has been validated in Brazilian Portuguese. 27

Outcome
Levels of EE, DP, and PA were used as dependent variables in an analysis to identify risk factors for each burnout dimension.

Statistical analysis
The Statistical Package for Social Sciences (SPSS) version 18 was used to analyze the data. of 5% (p ≤ 0.05) was adopted for all statistical tests.
All tests were two-tailed.

Results
One hundred and eight-five psychiatry residents were invited to answer the questionnaire, 132 of them answered it, and 115 (62%) were included in our sample, after excluding 17 individuals because they did not answer all the questions. Table 1 Table 2. Spearman's rho correlations are shown in Table 3.
In the linear regression models (Table 4)

Discussion
Our results portray a sample of psychiatry residents with high levels of emotional exhaustion and depersonalization and a worrying prevalence of screening positive for anxiety, somatization, depression, and suicidal ideation. Their risk of alcohol abuse and dependence according to the AUDIT-C is also alarming, especially among the men.
Previous studies that evaluated psychiatry residents also show worrying rates of emotional distress among participants, namely symptoms of anxiety, depression, suicidal ideation, and use of psychotropic medications. 12,19 In counterpoint to this, one fact that caught our attention was that, despite the high rates of emotional distress in our sample, the level of satisfaction with the profession and the feeling of personal accomplishment were both high.
To our knowledge, this is the first study that evaluates the role of interpersonal and institutional aspects as risk factors for burnout in psychiatry residents, and several of the aspects it revealed are worth mentioning. Eighty-six percent of the subjects reported they had had at least one preceptor/ supervisor from whom they felt support. Nevertheless, 57.4% claimed to have suffered abuse/harassment from at least one preceptor/supervisor and 42.6% of these declared it had had a negative impact on their academic life. According to the WEEI, the environment was evaluated as healthy in 40.9% of cases, in 30.4% as risky, and in 28.7% as toxic. Moreover, according to the Kruskal-Wallis one-way ANOVA (Table 5), a healthy environment was related to lower EE and DP scores and higher PA scores than risky and toxic environments (p < 0.05). In the linear regression models, relations to the institutions themselves and the relationships with preceptors/supervisors were related to EE levels, and the relationship with superiors was also related to DP. with connecting to patients, which is a major concern in psychiatric treatment.

Disclosure
No conflicts of interest declared concerning the publication of this article.