Health professionals are not exempt from physical and psychological disorders. Interest in identifying mental health problems or “psychological vulnerabilities (1)” in physicians that could interfere with their well-being and quality of their patient-related services has been increasing over the last four decades.

Varying rates of alcoholism, drug abuse, psychiatric disorders, and suicide had been reported in physicians, but they are always higher than in the general population (2–4). Certain characteristics and conditions of these jobs are unhealthy behaviors that lead to decreased general wellness and affect a professional's health (5, 6): difficult relationship with patients and their families, daily contact with death and pain, responsibility over others' health, workload, dealing with emergencies, the mismatch of personal expectations against institutional circumstances, and poor social networks.

Previous research has evaluated mental health and psychopathology among undergraduate medical students. Although results are sometimes contradictory, they show a significant prevalence of psychiatric illness, mainly anxiety disorders and depression (7), and the evidence that medical students are more susceptible to different kinds of psychopathology than their non-medical peers (8–12).

Symptoms of obsessive-compulsive behavior (13), learning problems, addictions, anxiety, and affective disorders (14–16) among residents in different specialties have been reported. Psychiatry resident impairment was recently identified as an emerging field of study for psychiatrists in academic settings (17). The prevalence of psychological symptoms among psychiatry residents has not been formally studied; it was estimated to be from 4% to 22% in a methodologically limited research study (18). Gender differences had not been reported previously.

The aim of this cross-sectional study is to describe the frequency and degree of possible psychopathology among psychiatry residents undergoing training at a major psychiatric hospital in Mexico (the biggest in the country and with the largest number of residents enrolled).

Subjects

The eligible sample population comprised a total of 112 residents (from 4 postgraduate years, enrolled in Hospital Psiquiátrico Fray Bernardino Álvarez during Spring 2008. Ninety-two (49 women and 43 men; mean age: 27.6 years) completed the SCL-90 in one session. Twelve refused to sign the informed consent, and eight were rotating overseas. The study was approved by the local Ethics Committee.

The Symptom Check List–90 (SCL-90) is a self-administered inventory with 90 items, originally developed by L. Derogatis (19) for screening psychiatric symptomatology; it includes nine symptomatic dimensions: Somatic Worries, Obsessive-Compulsive, Interpersonal Sensitivity, Depression, Anxiety, Hostility, Phobic Anxiety, Paranoid Ideation, and Psychoticism. Its validity and reliability were confirmed in a general population sample (20).

Analysis

We compiled descriptive measures, such as frequency, means, medians, and standard deviations (SD) for each symptom. Bivariate analysis using SAS statistical software was applied for exploring the association between gender and SCL-90 results. The cutoff for statistical significance was<0.01, mainly because of the sample size.

Table 1 presents descriptive values of SCL-90 symptom dimensions. Interpersonal Sensitivity, Depression, Obsessive-Compulsive, and Somatic Worries were considered more frequent or more severe than the rest. Also, we show differences between men and women. Men reported fewer significant symptoms than women. The strongest statistical significance was shown on the Somatic Worries dimension; however, significant differences were also found in Obsessive-Compulsive, Interpersonal Sensitivity, Depression, Anxiety, and Hostility.

Data were compared with a previous study, which applied SCL-90 to a general population (20), as seen in Table 2. This sample included 228 healthy Mexican subjects (115 women and 113 men; mean age: 34).

A significant proportion of psychiatry residents in our sample reported psychological symptoms, although they were described less frequently and less severely than in the general population. Demographic aspects about the comparison sample that could explain the observed differences are not known.

A noteworthy fact is Interpersonal Sensitivity as the most frequent and severe symptomatic category among psychiatry residents, which differs from the general population. This dimension focuses on personal feelings of inadequacy and inferiority, shyness, self-referential behavior, and low tolerance of criticism—features that could lead to problematic scenarios within clinical practice.

Our study has several limitations: The sample is small, but it represents 30% of psychiatry residents enrolled from every residency program in Mexico. We utilized a single symptom-checklist, whereas other studies simultaneously apply more instruments. The SCL-90 is not used for determining any specific diagnosis; as a screening instrument, it presents the frequency and degree of possible psychopathology. We did not explore the presence of other possible confounding or associated factors, such as burn-out syndrome or existing medical conditions.

It is possible that most anxious, depressed, hostile, or otherwise less healthy residents chose not to complete the questionnaire; probably, their inclusion would yield larger group differences. The possibility of both recall and rater bias cannot be ruled out because of the self-report nature of our instrument. Finally, our study was cross-sectional, and, thus, no inferences about the time of onset, sequence of symptomatology, or causation could be made.

There are substantial knowledge gaps about the mental health of physicians; only more detailed studies about psychiatry residents and other mental health professionals will help us to address any neglected treatment needs (21).