To the Editor: The contact with simulated patients (SPs) bears the potential to improve medical teaching, including teaching in psychiatry and psychotherapy (1–3). To-date, only a few studies have addressed the question of whether SPs can be distinguished from real patients in the field of psychiatry. In the study by Sanson-Fisher and Pool, students were unable to tell SPs from patients (4). Krahn et al., on the other hand, observed that students could accurately identify SPs most of the time (5). Authenticity, that is, the impossibility of distinguishing SPs from patients, might be of particular relevance since it strongly modulates the experience of the students. In the study by Krahn and coworkers, 91% of the students who identified SPs reported feeling less empathy for the person. Our own experience is similar: Students frequently report having difficulty feeling and acting as doctors if they perceive the simulation as unrealistic. The aim of this study was to clarify whether SPs portraying psychiatric disorders can be distinguished from real patients.

Twelve videos showing a 15-minute exploration of six SPs (two with major depression; one with obsessive-compulsive disorder, one with borderline personality disorder, one with a manic episode, and one with schizophrenia) and six actual patients with the corresponding mental disorders were presented to 20 experienced psychiatrists blind to the status of the interviewed person (7 men and 13 women; minimum of 5 years postgraduate training in psychiatry). The raters were informed that some patients would be SPs. SPs had been trained for about 4 hours to develop authentic acting skills, that is, a high congruency between verbal and nonverbal communication, as well as emotional and cognitive characteristics of the role. Patients were recruited at the Department of Psychiatry, University Medical Center, Freiburg. The study was approved by the local ethic committee.

Authenticity was operationalized by having the raters retrospectively allocate the interviewed person to a group (possible answers: SP, Patient, unsure). Furthermore, the raters were asked to do a global rating of authenticity (6-point Likert Scale, 1: very good, to 6: very bad) and to assess the given diagnosis. Additional questions were asked for their usefulness in teaching medical students (e.g., whether symptoms were obvious).

A total of 240 questionnaires (100%) were filled in by the raters. In most of the cases, SPs were not identified: in 70 of 114 SP cases, raters believed they were seeing a real patient. In 6 cases, raters were unsure. Real patients were correctly identified in 107 out of 119 cases. In 6 cases, real patients were misjudged as an SP, and in 6 cases raters were unsure. The statistical analysis shows a significant difference between SPs and real patients (ratings of “unsure” were omitted) in favor of the real-patients group (χ²=20,238; p <0.0001). Detection of SPs was 38.5% for the “borderline” video, 26.9% for depression-2, 23.1% for mania, 7.7% for depression-1, and 3.8% for schizophrenia. The SP character in the SP video “obsessive-compulsive disorder” was not exposed as such at all. The number of correct diagnoses was the same in both groups (104 each; 87.4%). The ANCOVA calculations based on intrarater normalized ratings show that SPs and real patients differed significantly on 4 items: Real patients were rated as more authentic (1.63 versus 2.05; p=0.001). SPs were rated significantly better for the items “Case is typical” (1.75 versus 1.99; p=0.033), “Video is catchy for students” (1.63 versus 1.86; p=0.046), and “Symptoms are obvious to students” (1.48 versus 1.88; p=0.001).

The results show that SPs are not as authentic as real patients. This might not come as a surprise; however, it is worth noticing that, in the majority of the cases, SPs were not detected. Differences in the detection rate in comparison to previous studies might arise from different ways of training, level of experience of the rater, or from the portrayal of different disorders: in the study of Sanson-Fisher and Poole, for instance, SPs were free to choose a personal problem and did not simulate a specific mental health disorder (4, 5). This makes the simulation much easier.

Some critical issues noted by the raters of this study were the following: “gestures too quick for depression,” “did not present enough formal thought disorders,” or “answers were too textbook-like.” Similar statements can be found in the study by Hodges et al. about an OSCE (objective structured clinical examination), where 13% of the students did not find the simulations realistic and commented that the cases were too straightforward (6). This suggests that intricacies in the simulation that potentially reduce authenticity can be identified and thus trained.

Our results indicate that SPs can be used to teach medical students psychopathology. This agrees with other studies reporting that the use of SPs enriches medical instruction and offers students a wide-ranging spectrum of diseases (2, 3). Furthermore, SPs are even “better” and “catchier” in the presentation of typical symptoms. This likely owes to the fact that many patients in the clinic are not truly “typical” as described in the diagnostic criteria. Therefore, SPs might be better suited to familiarize beginners with the range of psychopathologic disorders, as suggested by Brenner (7).

In conclusion, our study suggests that SPs are not as authentic as real patients but, with proper training, can reach a high level of authenticity in presenting major psychiatric disorders when rated by experienced psychiatrists. Further studies should investigate in more detail which factors influence authenticity and what exact degree of authenticity is needed to make teaching with SPs effective.