An approach to teaching the psychiatric interview

Abstract

The psychiatric interview serves as the cornerstone of psychiatric practice. It is therefore essential that we find effective ways of teaching students how to conduct a psychiatric interview. The present paper arises from two faculty members at Memorial University of Newfoundland and Labrador considering how to improve the quality of teaching of the psychiatric interview to preclerkship undergraduate medical students, before they begin the clinical portion of their training. The interview is taught in discrete pieces initially (e.g., discussing confidentiality, screening for suicidal ideation, taking a history for depressive disorders, etc.) before being assembled into a whole interview.

The sessions are led by psychiatrists and residents who play the role of the patient. They use prewritten cases but can improvise to challenge or direct the students. Students receive real time feedback. The flexibility allows for students to repeat and vary their approach in response to feedback.

Anonymous course evaluations showed improvement in student satisfaction with the new psychiatry clinical skills teaching. Prior to implementing the new approach student satisfaction was at 3.9/5. With the new method scores improved to 4.7/5 and 4.5/5 in the following two years. Clinical skills OSCE scores remained stable with modest improvement following implementation. The class average was 8.5 in the year prior to implementation and were 9.1, 8.6 and 8.8 in the years following. As a side benefit, the approach lent itself well to being delivered remotely so it continued to function well during the disruption resulting from COVID-19.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12909-024-06529-1.

Introdution

The psychiatric interview serves as the cornerstone of psychiatric practice. It is therefore essential that we find effective ways of teaching students how to do a psychiatric interview. This is challenging for a number of reasons. A psychiatric interview focuses on different aspects than interviews in other areas of medicine and so often requires more time than traditional assessments in other areas of medicine. In addition, a psychiatric interview involves teaching students to screen for conditions which often involve lack of insight into the presence of illness as a symptom of the illness.

The present paper arises from two faculty members at Memorial University of Newfoundland and Labrador considering how to improve the quality of teaching of the psychiatric interview to preclerkship undergraduate medical students, before they begin the clinical portion of their training. The students are in the second year of their four-year medical degree. Outside of a lecture on the subject, this is the learners’ first exposure to interviewing psychiatric patients. The goal of the teaching at this stage of training is to ensure that the learners are aware of the structure and practiced in a complete psychiatric interview to prepare them for their clinical rotations when they will assess real patients.

We begin by describing the ‘traditional’ approach there. Considering that approach highlights some of the challenges involved in teaching the interview, as well as revealing the shortcomings of our own institution’s approach. We then describe a review of the available literature on approaches to teaching the psychiatric interview. This search did not lead us to a satisfactory alternative approach, although did provide some direction for the approach we developed. That approach is discussed in the third and fourth sections. The paper concludes with some reflections on the success of that approach thus far.

The ‘traditional’ approach at Memorial University

The long-established approach at Memorial consisted of four 2 h sessions in which a psychiatrist or psychiatry resident led a group of a dozen or so medical students. During these sessions, a real patient with a psychiatric history volunteered to be interviewed for approximately fifty minutes. The patient would then leave, and the interview would be discussed. The real patient volunteers were chosen on the basis of having a psychiatric history of some sort, but most often were relatively well at the time of the session, on the theory that this would allow them to adequately provide the students with a full history.

This format lent itself to an interview experience that was often quite artificial. After an initial session in which the students watched the psychiatrist or resident interview a patient, the students would share the role of interviewer, changing from one student to the next after a designated period of time or shift in topic so that the whole group collectively carried out a full psychiatric interview. The interview often focused on the past since typically the patient was not unwell at the present time. While this often takes place in outpatient practice, we felt it was not the most effective way to teach and solidify interviewing skills at this level of training. Furthermore, the fact that the interview subjects were volunteers meant there was little control over the type of psychopathology that could be covered. Volunteers had knowledge of their own illness experience, but no formal training, and as a result the history encountered by the students was unpredictablee and subject to luck of the draw. This approach lent itself to a discussion of the interview but not to a significant opportunity to repeat and vary the students’ approach to the interview. The volunteers left once the fifty-minute interview was completed and a different physician would typically lead the subsequent session so opportunities to put coaching from the physician into practice were very limited. Overall, it was an approach that provided some opportunity for the students to familiarize themselves with the general structure of a psychiatric interview but not to acquire firsthand experience at carrying out all the components of an interview and certainly not to repeatedly practice different aspects of the interview.

Literature review

Dissatisfaction with our institution’s approach led us to investigate the literature on how to teach this crucial clinical skill. We found the literature on the subject to be quite sparse. There are only a very few articles describing an actual approach to teaching the psychiatric interview. Nuzzarello described the implementation of a course very similar to the one then in place at our university and so with the same attendant difficulties [1]. Lehman described the implementation of a course for psychiatric residents to prepare them to teach medical students the psychiatric interview [2]. This article illustrates the value of preparing psychiatric residents to teach the interview, but does not present an actual approach to doing so.

Several articles focused on particular aspects of the psychiatric interview, rather than an overall approach of the sort we were seeking. Ward and Stein described an approach focused not on the content of the interview but on creating a productive interpersonal environment between patient and physician [3]. Bremner presents a critical review of the literature on the use of standardized patients [4]. The authors concluded that standardized patients were helpful for teaching discrete operational skills and allowed for a broad range of psychopathology exposure but also concluded that standardized patients were less effective for teaching complex interpersonal skills. The article therefore provided reason for us to at least consider using standardized patients instead of the volunteers used in our current approach, but also identified some difficulties with this. Pohl et al. compared three approaches to teaching the Mental Status Examination (i.e., one aspect of the psychiatric interview): a didactic lecture, a video interview and a live simulation [5]. They concluded that all three were useful in teaching the mental status examination. This is unfortunately of limited value for our purposes since the mental status exam is based upon information and observations gathered in an interview, whereas we were seeking to teach the information gathering techniques themselves. Fiedorowicz et al. described the implementation of a single session group activity in which a simulated overdose patient was assessed [6]. They note that learners who were exposed to a simulation showed more confidence in communicating in this situation. Again, this provides a reason to think that simulation might have a useful role to play in teaching psychiatric assessment, but this article focuses on a particular type of psychiatric assessment rather than a general approach.

In summary, the literature review did not reveal a method for teaching the psychiatric interview aside from one similar to the unsatisfactory one already operating at our institution. It did, however, provide at least some reason to think that standardized patients would be a useful alternative to our own practice of using volunteers. There is also some reason based on the above to conclude that simulated patients are good for technical aspects of an assessment, while real patients are good for complex interpersonal skills. With these observations made and seeking a better approach than our current one, we turned to developing our own approach. We attempted to benefit from the best of each by using content experts as simulated patients. Other points to consider include that learners favor instruction with observed practice and feedback, residents value teaching on how to teach, and barriers to feedback need further exploration.

Rationale for the new approach

The literature review and reflection on our current approach led us to pursue two key principles in our new approach: repetition and flexibility. We sought a model that was flexible enough to allow students to encounter a broad variety of psychiatric conditions and allowed students the ability to practice their approach repeatedly in response to feedback.

Our approach was developed with the Four Stages of Competence learning model in mind [7]. This model breaks down the steps involved in learning a skill into four stages, as depicted in Fig. 1. Initially, learners are unaware of how little they know. As they recognize their lack of skill, they consciously work to develop it. Once developed, they apply the skill while intentionally thinking through the steps. Eventually, the skill can be utilized without having to intentionally think about the steps. At this point, the individual is said to have acquired unconscious competence [7].

Fig. 1.

Competence hierarchy [8]

With respect to the Psychiatric interview, a learner who has received didactic teaching in psychiatry may initially have some knowledge of the information they need to obtain, but be unaware that they lack the skill to obtain it. This is the Unconsciously Incompetent stage. As they attempt the skill, they become aware of their lack of ability, and enter the Consciously Incompetent phase. With focused practice, they progress to the Consciously Competent stage, in which they can conduct an assessment by focusing on the structure of the interview and what they need to do. With further practice, the learner will progress to the Unconsciously Competent phase. At this stage, they are able to obtain the relevant information from the patient without having to focus on the structure of the interview itself.

Based on this learning model, we sought to devise a model that would allow for repeated exposure to aspects of the psychiatric interview both in terms of witnessing their colleagues interviewing someone and in terms of repeated opportunities for the individual learner to practice different aspects of the interview. For this to be accomplished within the limited curricular time available, we concluded we would need to shorten the length of a complete interview from the traditional 50 minutes. This necessitated abandoning the use of volunteer patients as their complex histories could not be gathered so quickly. We considered using standardized patients as we had seen described in the literature but realized this would not allow for sufficient flexibility on the part of the interview subjects. Given the unpredictability of the approaches novice learners would take, we wanted an interview subject who could tailor their responses to the needs of the individual learner. For example, if a student was struggling with screening patient for suicidal ideation, we wanted an interview subject who could vary their history subtly so that the learner could repeat the process multiple times without already knowing the ‘correct’ answer. This led us to the idea of using ourselves and our colleagues as the interview subjects. What using residents and psychiatrists might cost in acting skills would be made up by the content expertise that would allow them to vary the history as required.

The new approach

As previously, the new clinical skills sessions take place over four 2-hour long sessions. Prior to these sessions, the students attend a didactic lecture on the structure of the psychiatric interview and mental status exam (MSE). This includes a video demonstration of a simulated interview between this paper’s authors, with one acting as interviewer and the other as patient. This demonstration lasts about 20 min, covering all the standard elements of a psychiatric interview, but with a case designed to elicit pertinent negatives quickly so that it can be moved through efficiently. The goal of the didactic session was to provide a general explanation of the key pieces of the psychiatric interview and to show the students how those pieces could be assembled into a whole interview. The four sessions that followed mimicked this structure. The first 3 sessions focused on individual aspects of the interview, while the last 1 session focused on assembling those pieces into a whole.

The four clinical skills sessions involved groups of 5–6 medical students, with the same group working together throughout the 4 sessions and, when possible, with the same supervisor. As noted, the sessions were conducted without a real or standardized patient. Each session was co-facilitated by a psychiatrist and a psychiatry resident. All faculty and residents were invited and encouraged to participate. They were provided with clear written instructions in advance of each session. No additional time was required to prepare the facilitators. In the beginning, a brief overview was provided during a discipline faculty meeting. Monthly faculty meetings served as a platform to discuss progress. The facilitators were provided with several fictional cases that could be used to conduct simulated interviews. During the sessions, several simulated portions of interviews took place. One of the facilitators played the role of the patient, and the other observed as one of the learners conducted a focused interview. The rest of the group observed the interview. The group would then discuss the interview, and feedback was provided. The intent was that each learner would be able to conduct at least one interview at each session, with the opportunity to repeat and vary their approach, particularly if some aspect of the interview did not go well.

Each of the four sessions had a focus. The first covered beginning an interview (gathering patient identifiers, discussing confidentiality, and asking about the chief complaint), screening for depression and assessing for suicidality. The second covered screening for substance use, mania and psychosis. The third covered screening for anxiety disorders (Panic Disorder, Generalized Anxiety Disorder, Social Anxiety Disorder) as well as Obsessive Compulsive Disorder and Post Traumatic Stress Disorder. The goal of the fourth and final session was to give each of the learners the experience of taking a full history. For this session, the cases were crafted to allow for covering a full assessment in about twenty minutes. This approach allowed us to cover a wide breadth of psychopathology.

Once this approach was developed, it was trialed on a small scale. In one academic year (2017–2018), the authors of this paper acted as the co-facilitators for one group of learners. All other learners participated in the previously established psychiatry interview clinical skills teaching sessions. The authors elicited feedback from the group, and based on the unanimously positive response, proposed that the new method be piloted unchanged for the entire medical school class the following year. The proposal was accepted by Memorial University’s Discipline of Psychiatry, and the new method was implemented for the entire class during the 2018–2019 academic year. Following the completion of the first full class trial of the new approach, feedback was elicited from the participating faculty members. The feedback was unanimously positive, and after further discussion the discipline elected to continue with the newly developed method. It continues to use it currently. At the time of writing, the new method had been employed for four full cohorts, during the 2018/19, 2019/20, 2020/21 and 2021/22 academic years. The outbreak of the global COVID-19 pandemic posed many challenges for the 2020/21 academic year. Thankfully, the new format lent itself well to being conducted virtually. As a result, the Psychiatry sessions were among the very few that proceeded largely unaltered for the clinical skills course.

Assessing the new approach

Ethical approval was sought from and approved by the Health Research Ethics Authority (HREA).

The psychiatry clinical skills sessions make up one component of twelve sections in a general medical clinical skills course at Memorial University Medical School (MED7720). Each year the students are given the opportunity to complete a survey of their experience with the course. They are surveyed on their opinion of the course overall and of the specific sessions. In an attempt to quantify the medical student perception of the new method, we obtained the survey results with regard to the psychiatry sessions. To add context to the interpretation of the results, the survey results for the sessions overall were also obtained. In addition, the course surveys allowed for freeform narrative feedback. This was reviewed and any comments directed at the psychiatry sessions were flagged. Feedback was obtained from the last year that the traditional method was used for the whole class (2016/17) onward. Unfortunately, feedback from the 2020/21 academic year was not yet available. This project falls under the category of secondary use of data. The survey data was collected by the university as part of their internal quality process. The survey was developed by the university, and not previously published. An example of the survey has been included as a supplementary file.

The only question focused on the specific sessions was “Please indicate the extent to which you agree (or disagree) that the following components of Clinical Skills III effectively met objectives.” Options were: Not Applicable (0), Strongly Disagree (1), Disagree (2), Neither Agree nor Disagree (3), Somewhat Agree (4), Strongly Agree (5). Table 1 shows the average score for the Psychiatry sessions and the sessions overall for each academic year.

Table 1.

Medical student feedback survey quantitative data

	2016-17 Last year the traditional approach was used	2017-18 Single group pilot of new approach	2018-19 1st year when only new approach used	2019-20 2nd year of new approach only
Did psychiatry sessions effectively meet objectives?	3.9 (n = 20)	3.9 (n = 18)	4.7 (n = 16)	4.5 (n = 10)
Did sessions overall effectively meet objectives?	4.0 (n = 218)	4.1 (n = 195)	4.6 (n = 189)	3.9 (n = 65)

Figure 2 depicts the numerical survey results as per Table 1.

Fig. 2.

Session feedback likert scores

The complete freeform quantitative comments for each academic year are included in appendix A. Regarding comments directed to the traditional approach, one student felt that the changing groups prevented them from getting comfortable enough to participate. One student enjoyed hearing real patients’ stories, but noted that often the patients wanted to chat instead of focus on the “psychiatric illness.” One student commented that the sessions did not allow for much practice. Another felt the sessions were artificial and would have preferred a standardized patient with an acute presentation.

With regard to the new approach, one student wrote that they “found the psychiatry sessions very helpful and well instructed. The skills were uniformly taught across groups as well and we knew exactly what was expected of us for the sessions.”. Another wrote, “I really enjoyed all of the sessions, especially the psychiatry sessions…”.

Figure 2 depicts the average survey scores for the psychiatry sessions, as well as for the course overall for each academic year. In the two years preceding the implementation of the new method for the entire class, the psychiatry sessions ranked just below the overall course averages. In the two years following the implementation of the new method for the entire class, the psychiatry sessions were ranked just above the overall course averages. It must be recognized, however, that the medical student survey asked only a single question regarding the psychiatry session, allowing limited inferences to be made from the results. Acknowledging this limitation, it would appear that the numerical trend is positive with regard to the newly implemented approach. A review of the freeform qualitative feedback also highlighted a positive trend following the transition from the traditional to the new method. This positive trend is in keeping with the informal feedback obtained from the faculty and residents that participated in the sessions.

The results are therefore encouraging although do not clearly document superiority of the new approach. Another possible source of evidence for effectiveness of the approach is the OSCE (Objective Structured Clinical Exam) taken by medical students at Memorial in their second year of studies. This exam contains stations dealing with the topics taught in clinic skills during the second year and includes a psychiatry station. There were some difficulties in obtaining this information, so it is incomplete. The available information is summarized in Table 2.

Table 2.

Psychiatry OSCE average mark by year

Academic Year	Notes	Average Mark
2016-17	Last year - traditional approach	Not available
2017-18	Mixed traditional and new	8.5
2018-19	Full new approach	9.1
2019-20	No OSCE	None
2020-21	Virtual but using new approach	8.6
2021-22	In-person with new approach	8.8

This information is unfortunately not particularly illuminating. We were not able to obtain information concerning the marks prior to the new approach being introduced nor could we access information about the overall station averages for these years. There are therefore no clear inferences to be drawn from this information, although it can be said that the information does not suggest students are doing worse since the new approach was implemented and there is weak evidence that they may be doing better in that the lowest average grade occurred when the new approach was only being used in one group.

A final consideration when evaluating the new approach concerns the performance of this approach during the COVID 19 global pandemic. The new approach had been used twice for all of our students in the fall of 2018 and 2019. Then, like all medical schools, our curriculum was profoundly disrupted by the pandemic in 2020, with significant difficulties continuing to be encountered in delivering the curriculum by the time psychiatry clinical skills was scheduled in the fall of 2020. Serendipitously, however, our newly developed approach lent itself to virtual teaching very well. Aside from the shift to virtual teaching, the sessions proceeded with no interruption. The structure of the sessions was essentially identical, with the only difference being that they were conducted with video/audio as opposed to in-person. No real or standardized patients were required. This avoided the obstacles that these extra variables would have entailed, such as confidentiality, virtual platforms and technological issues. The small group sizes worked well in an online format, allowing for meaningful interaction between students and preceptors. As all the cases used were fictional, there were no concerns around encryption or confidentiality. The sessions were so successful in fact that the discipline of psychiatry made a decision to run them virtually in 2022 even though in-person instruction had resumed. It was felt that this allowed the sessions to serve the dual purpose of familiarizing the students with the newly essential skill of providing virtual care while still teaching the psychiatric interview effectively.

Conclusion

The project described in this paper began as an attempt to improve the quality of a teaching approach we felt was wanting. We reflected on the previously established approach and reviewed the literature to inform our plan. After a successful pilot, our new approach was implemented fully. Informal feedback from participating faculty and residents has been positive, as have medical student survey responses. OSCE scores suggest no concerns with the new approach. As a bonus, the timing of the transition to the new method avoided disruption of the teaching sessions by the COVID19 pandemic. Overall, there is reason to believe this new approach represents a significant improvement in the quality of teaching of the psychiatric interview at our institution. We encourage other programs to adopt a similar approach for their institution.

Limitations

One criticism of the traditional approach was the artificial nature of the interaction. It must be acknowledged that the new approach is also artificial. Neither approach is equivalent to interviewing real patients in a clinical setting.

It would have strengthened the assessment of the new method versus the traditional if there had been documentation of performance as the learners progressed through the course. It would also have been preferable to compare interviewing performance in clinical settings for the learners that participated in the traditional method against those that were part of the new approach. Unfortunately, no appropriate performance assessment was available. Using OSCE results and post course questionnaires were less robust methods.

Developing a formally documented assessment that occurs throughout the course will aid in assessing the effectiveness of the teaching. Developing a documented evaluation of interview performance in the clinical setting would further strengthen the assessment of the teaching.

Electronic Supplementary Material

Below is the link to the electronic supplementary material.

Author contributions

T. H. conducted the literature review and developed the initial plan for the novel teaching approach. T. H. and A. L. worked together to finalize the novel teaching approach, implement it, assess it’s impact and author this article.

Funding

No funding was sought or provided for this study.

Data availability

All data is included within the manuscript.

Declarations

Ethics approval and consent to participate

Ethical approval was sought and granted by the Newfoundland and Labrador Health Research Ethics Board (HREB), Reference # 2024.041. The HREB operates according to the Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans (TCPS2), ICH Guidance E6: Good Clinical Practice Guidelines (GCP), the Health Ethics Authority Act (HREA Act) and applicable laws and regulations. The ethical approval granted by the HREB included a request for waiver of consent.

Consent for publication

Not Applicable.

Competing interests

The authors declare no competing interests.

Authors’ information

T.H. and A.L. and practicing psychiatrists and full-time faculty members of the Memorial University of Newfoundland Faculty of Medicine.