At both medical schools that I was affiliated with, students rated highly the preclinical teaching programs of patient interviewing , probably because of their enthusiasm about meeting patients for the first time. Their most common difficulties were reflected by the following critical comments that were expressed during the programs and the subsequent clerkship rotations.

What we learn is different from what we see

The main difficulty arose from the mixed messages that students received from their tutors. The disease-centered orientation of many clinical tutors was inconsistent with the patient-centered approach that was advocated in preclinical lectures. Preclinical students were taught to ask open-ended questions (“Tell me about your pain”) and to listen without interrupting the patient’s narrative. Yet, during clerkship, they encountered doctors who frequently interrupted patients by asking closed questions (“Did the pain radiate to the arm?”) Student difficulties in reconciling these inconsistencies bred a feeling that, “What we learn in the course on interviewing skills is not what real medicine is about.” Other authors have similarly observed a decline in students’ patient-centered attitudes as they progress through medical school [14].

I felt embarrassed to be lectured on how to talk to people

Students acquire the ability to communicate with other people before beginning their medical education , and they bring to the classroom the well-formed habits that are believed to define a civilized person. Consequently, some of them felt uncomfortable, and even offended, when asked to show what appeared to them to be elementary courtesy, such as greeting a patient. They also appeared reluctant to adopt interviewing guidelines that were inconsistent with their well-entrenched attitudes. Examples of such “counterintuitive” guidelines included the requirement to sustain the patient narratives rather than to interrupt them by asking closed questions; and to wait a second or two after a patient has stopped talking, rather than ask the next question immediately.

You should have changed the subject when the patient started weeping

This comment was made by a student after she had observed me while demonstrating to a group of 20 students an interview of a real patient. In response to my question “could you tell me more about your family?” the patient described his children and wept while he told us about his recently deceased wife. The student’s remark reflected the difficulty of many medical trainees in dealing with a patient’s expression of emotion. Students avoided emotional issues and changed the subject when patients shared personal problems. Other authors have similarly reported that medical students felt intrusive as they explored a patient’s psychosocial situation and felt insecure in coping with emotionally loaded topics [15].

There are so many rules to remember

Many students appeared overwhelmed by the large number of interviewing rules and system-review items that they received in lectures and handouts. In many cases, these rules appeared to them contradictory and incoherent, creating the impression that patient-centered interviewing was rule bound. Their difficulties may have been compounded by the lack of an appropriate vocabulary to understand the rules. For example, some tutors defined a “doctor’s control of the interview” as an effort to prevent patient digressions, whereas for others it was an attempt to sustain the patient’s narrative. The students’ difficulties with comprehending the guidelines may have added to their fears of appearing foolish in their supervised patient interviews.

Only now I understand what you meant by…

This comment was recurrent after students watched me interview a patient. It underscored the students difficulties when they were provided only with verbal guidance or verbal feedback after communicating with patients. Similar to any other skill, patient interviewing appeared to necessitate demonstrations: students needed to be shown, rather than just be told, what to do. The absence of live demonstrations in the teaching programs may have been due to the belief that verbal explanations to students, and the viewing of a single videotaped interview by a respected clinician, are sufficient. It may have also been due to a lack of tutors who felt confident in their ability to demonstrate interviewing techniques . Although doctors had the necessary experience, they were rarely familiar with patient-centered interviewing techniques . On the other hand, social scientists were familiar with the literature on patient-centered care, but lacked the experience or confidence to interview patients.

I feel that we have had enough interviewing practice

Some students complained about the repetitive nature of their practice interviewing sessions during which they were required to elicit the same information from different patients. Even though their tutors rejected this complaint (“No two patients are alike,” “practice makes perfect,”) it identified a major weakness of the program: the absence of a well-defined structure. Consequently, the program failed to produce in students a feeling that they were progressing in their ability to deal with problems of increasing complexity .

Table 4.2 summarizes some of the students’ comments. These comments, as well as published observations by other authors, identify problems that seem to be unique to teaching patient interviewing .

First, students encounter a marked inconsistency between the patient-centered interview style that is taught in preclinical programs, and the observed doctors’ behavior during clerkship [16, 17]. Indeed, emphasis placed on specific communication skills varies widely between and within teaching programs, and this variability has led to calls for uniformity in teaching patient interviewing [18].

Second, unlike other clinical skills , patient interviewing may require that students change their preconceived habits and overcome their reluctance to explore feelings. Such “reconditioning” is more intellectually demanding than learning of new subjects [20]. Students may be unresponsive to authoritarian teaching or written handouts summarizing the rules of clinical interviewing. They may be bored by the “should” rules of interviewing that they perceive as self-evident (e.g., showing respect of patients); they may reject the “should not” rules that are inconsistent with their natural tendencies (e.g., refrain from avoiding the emotional expressions of patients); they may resist attempts to impose upon them specific interviewing habits (e.g., avoiding “why” questions). In some cases, students may even have difficulties in understanding interviewing rules that are presented out of their clinical contexts.

Third, throughout medical school, the various disciplines are taught by subject-matter specialists. Even medical schools that encourage self-directed, problem-based learning provide students with opportunities to consult experts. However, the teaching of patient interviewing appears to be an exception to this practice. The realization that clinical faculty are not experts in communication skills has led to attempts to involve behavioral scientists in the clinical training of medical students. These attempts were only partially successful because, as I stated earlier, behavioral scientists lack clinical experience and credibility with students. Furthermore, Engel [21] has warned that such combinations may perpetuate the split between somatic disease and psychosocial problems by implying that the latter are the exclusive domain of behavioral specialists rather than part of the clinical practice of all physicians.

Fourth, several authors have questioned the appropriateness of the methods of assessing students’ interviewing skills. These methods consist of observing trainees either for an entire single interview [22] or during an objective structured clinical examination (OSCE) [23], and of assessing performance either with detailed checklists [24] or with global ratings of a small number of evaluation items [25]. The advantage of global ratings is their purported ability to measure domains not amenable to coding; however, evidence suggests that there are significant deficiencies in such global evaluations [25]. The advantage of checklists is their high inter-rater reliability [26]; however, it has been claimed that “students frequently interpret these [checklists] to mean that they should ask as many content [disease] related questions as possible in a limited timeframe in order to earn checks…. For some, the OSCE inadvertently reinforces a close-ended, disease-focused approach to the interview” [27].