Bokken, Lonneke MD, PhD; Rethans, Jan-Joost MD, PhD; Jöbsis, Quirijn MD, PhD; Duvivier, Robbert; Scherpbier, Albert MD, PhD; van der Vleuten, Cees PhD
Contact with patients has always been an integral part of undergraduate medical education. Early patient contacts increase students' motivation, ease the transition from preclinical to clinical training, and teach students things that cannot be learned from books, such as empathy, responsibility toward patients, and professional identity.1–3 Patient contacts also help students build integrated skills for clinical reasoning, communication, history taking, and physical examination.1,2,4,5
Patient contacts in medical education may involve real patients or simulated patients (SPs). Both contacts make unique contributions to medical education.6 Although many studies have compared the instructiveness of encounters with real patients or SPs with more traditional teaching methods, such as lectures or instruction by faculty members in the teaching of skills to medical students, only a few studies have compared the instructiveness of real patient contacts with SP contacts in the teaching of clinical skills.6–10 With regard to the instructiveness of the contact, all of these studies focused on the performance of students in the domain of clinical skills—for example, students' scores on an objective structured clinical examination. Most of the studies found no difference between the performance of students taught by real patient contacts and that of students taught by SP contacts. Nevertheless, our students often comment on the instructiveness of real patient contacts and SP contacts being dissimilar. In this context, the instructiveness of the contact refers to the educational value of the contact as perceived by students. The perceived instructiveness of real patient contacts or SP contacts thus seems to differ from the instructiveness of the contacts in terms of student performance.
We performed an experiment in which first-year medical students were randomized to having a real patient contact or an SP contact. The aims of this experiment were (1) to evaluate which contact (real patient or SP) is perceived as most instructive by students and teachers and (2) to evaluate which variables contribute to the perceived instructiveness. A mixed-method evaluation approach was used with quantitative (questionnaires) and qualitative (focus groups) evaluations of perceived instructiveness.
Once every three weeks, undergraduate students (Years 1–3) of the Maastricht Medical School have the opportunity to practice communication and physical examination skills in encounters with SPs at the Skillslab, the department where medical students are taught medical skills.11 These simulated doctor–patient encounters involve one SP and two students, with one student in the role of doctor and the second student observing. The students change roles for each SP encounter. After the encounter, SPs provide the “student-doctor” with feedback. SP encounters are recorded on DVD and discussed one week later in a tutorial group composed of 10 students and a teacher. The SP program is fully integrated with the undergraduate curriculum of the Maastricht Medical School.11
For this experiment, encounters with real patients were organized within the structure of a regular SP encounter on asthma in Year 1. The experiment was performed during this SP encounter because the recruitment of real asthma patients seemed relatively easy and because students were offered the opportunity to practice a full consultation (including physical examination) during the encounter. The routines of the SP encounters and real patient encounters were comparable. Both encounters occurred at the Skillslab, included one student in the role of doctor and a second student observing, were recorded on DVD, and were discussed in a tutorial group one week later. SPs and real patients both provided the student-doctor with feedback on the encounter.
All first-year students were informed about the experiment by an announcement on the electronic learning environment of the Maastricht University. The 163 students registered to perform the consultation in the role of student-doctor were randomized to having a real patient encounter or an SP encounter. An e-mail sent a few days before the consultation informed the students about the encounter, specifically about which type of patient (real or simulated) they were assigned to. Comparable preparation advice for both encounters was offered. The students, all involved in a module that addressed asthma, were assumed to be able to manage a consultation with a patient suffering from asthma. All students had had at least one regular SP encounter in the role of student-doctor before the experiment.
Recruitment and selection of real patients and SPs
For the experiment, we recruited adult patients with asthma via the practices of general practitioners (GPs) in Maastricht, because we thought it was important to have the GPs' approval for participation of their patients. Several criteria were laid down for the selection of real patients: (1) To ensure that the patient suffered from mild-to-moderate persistent asthma (according to the GINA guidelines),12 patients had to be using inhalation steroids; (2) to avoid medical histories too complex for first-year students, patients could not be currently treated by a pulmonologist; (3) patients had to be available during at least one of the scheduled sessions; and (4) patients had to feel comfortable with having a consultation (including physical examination) with a first-year medical student.
The recruitment and selection of patients turned out to be very difficult because of insufficient cooperation from GPs' practices. In only one practice were the GPs able to recruit patients themselves, and one practice agreed to allow the main researcher (L.B.) to recruit patients. In total, nine patients were recruited and selected, with an age range of 54 to 82 years. All patients were informed about the encounters with the students. The mean number of student encounters per real patient was six. Five SPs were recruited from our SP bank, with an age range of 39 to 66 years. The mean number of student encounters per SP was 20. Both SPs and real patients were paid €8 ($12) per hour for their participation.
A questionnaire regarding the perceived instructiveness of real patient encounters and SP encounters was developed on the basis of the literature and previous qualitative research.13 The draft questionnaire was revised on the basis of comments of several experts in the field of medical education, second-year medical students, and teachers. The final questionnaire consisted of two parts. The first part contained general questions regarding the patient encounter, such as position of the student during the encounter (student-doctor or observer) and type of patient encounter (real patient or SP). The second part, shown in Table 1, contained statements on seven aspects of the patient encounter: preparation, authenticity of the consultation, content of the consultation (learning communication, history taking, and physical examination skills), safety during the consultation, feedback provided by the patient, difficulty of the encounter, and impact of the encounter (on learning in general). The students were asked to indicate their agreement with the statements on a four-point Likert scale (1 = complete disagreement, 4 = complete agreement). In addition, students could assign a general mark (on a 10-point scale) for the perceived instructiveness of the patient encounter and could make free comments. The questionnaire was anonymous so students would feel free to report their views.
To explore the views of teachers of the groups in which the patient encounters were discussed, a second questionnaire was developed. This questionnaire, shown in Table 2, was intended for teachers of groups in which students had had encounters with real patients and SPs. Teachers could also make free comments.
Focus group sessions.
To gather qualitative information on the perceived instructiveness of the patient encounters and to elaborate on the results of the questionnaire, two focus group interviews, each lasting 1 to 1.5 hours, were held. All students who had performed the consultation in the role of student-doctor were sent an e-mail invitation to participate in a focus group session. To accommodate the 17 students interested in participating, we held two focus groups. Students were paid €15 ($22) for their participation. The focus group sessions followed the guidelines offered by Morgan and Krueger.14 The focus group interviews were audiotaped for later transcription and guided by an independent moderator. It was explained to the students that the analysis and the reporting of the results would be anonymous. The interview was semistructured on the basis of a preestablished interview guide (List 1), which was based on the results of the questionnaire.
We analyzed the data with SPSS Version 15.0 for Windows (SPSS Inc., Chicago, Ill.), calculating frequencies for all items. Differences between evaluations of students who had an SP encounter and those of students who had an encounter with a real patient were calculated using the t test for two independent samples. The t test was used because the evaluations of students were normally distributed by approximation because of the large numbers of students. ANOVA was used to test whether the role of the student during the consultation (student-doctor or observer) had an effect on the difference between the evaluations of students who had an SP encounter and those of students who had a real patient encounter. In cases in which the role of the student seemed to be of influence, a simple effects analysis was carried out. Because scores on item 9 and on items 15 to 19 of the questionnaire were nominal (Yes or No), differences with regard to these items were calculated using the chi-square test. Effect sizes were calculated following the guidelines described by Hojat and Xu.15 A P value smaller than .05 was considered statistically significant. Differences with effect sizes around 0.20 were considered to be of limited practical importance. Differences with effect sizes around 0.50 were considered to be of moderate practical importance, and differences with effect sizes around 0.80 were considered to be of large practical importance.
Focus group interviews.
The recordings of the two focus groups were transcribed by one of the authors (L.B.). The transcripts were summarized and sent to the students for approval. The transcripts were imported into the software program ATLAS-ti (ATLAS.ti Scientific Software Development GmbH, Berlin, Germany) and were reviewed and coded by two reviewers independently (L.B. and R.D.). When the two reviewers disagreed, they discussed codes until agreement was reached. The main codes corresponded with the questions discussed in the focus groups (List 1). Additional codes were used for new themes raised by the participants during the discussions.
Of the 163 students registered to perform the consultation, 61 students (37%) had encounters with real patients, and 102 students (63%) had regular SP encounters. The total response on the student questionnaire was high (87%). Of the students randomized to real patient encounters, 61 student-doctors (100%) and 57 observers (93%) responded to the questionnaire. Of the students randomized to SP encounters, 86 student-doctors (84%) and 80 observers (78%) responded to the questionnaire.
Table 1 shows the evaluations of student-doctors and observers with regard to real patient encounters and SP encounters. There were several significant differences between the evaluations of students who had an SP contact and those of students who had a real patient contact. The effect sizes of most differences were considerable, indicating moderate to large practical importance. For example, compared with SP encounters, students considered real patient encounters less helpful in practicing communication skills (item 7). Students felt real patients were more themselves during the encounter (item 6), and students were more curious to find physical abnormalities in real patients (item 12). Furthermore, the feedback provided by SPs was considered more relevant than the feedback of real patients (item 26) and was more helpful in the making of new learning goals (item 31). The position of the student during the patient contact (student-doctor or observer) seemed not to influence the differences that were found, except for the difference on item 7. Here, the evaluations of student-doctors were significantly different with regard to real patient encounters and SP encounters, but the evaluations of observers were not. The mean mark for the general instructiveness of real patient encounters and SP encounters was not significantly different (item 38).
The response on the teacher questionnaire was 67%. In all groups, students had encounters with real patients and SPs, with a mean number of three SP encounters and two real patient encounters per group. The evaluations of teachers regarding real patient encounters and SP encounters are shown in Table 2. The majority of the teachers indicated students showed better skills in SP encounters or indicated that there was no difference. SP encounters were considered more instructive than real patient encounters. Several reasons were mentioned. For example:
[S]tudents can learn the basic skills in a safe environment with trained SPs.
Students have difficulty structuring [the consultation] and you can better practice this with an SP.
Most teachers preferred the feedback provided by SPs to the feedback provided by real patients, and real patient encounters were considered more difficult than SP encounters. Many teachers felt the real patients were insufficiently informed about the purpose of the consultation, which contributed to the difficulty of the consultation.
Focus group interviews
Compared with SP encounters, most students had better prepared for the real patient encounters. A thorough preparation helped students to feel more self-assured.
I was afraid I would say something wrong because a real patient knows a lot more about medical aspects than an SP.
Most students considered the real patient encounters to be more authentic, as they felt some SPs were simply presenting preestablished symptoms or were deliberately withholding information.
An SP just knows what I need to know to make a diagnosis and real patients don't.
Real patients talk more of their own accord.
Because of their authenticity, real patient encounters made a more profound impression on the students.
Real patient encounters were considered less useful for practicing communication skills and learning goals related to communication skills, such as “learn to structure the consultation by making summaries.” Students felt that some real patients were ill informed about the purpose of the consultation because they did not make a clear request for help.
She said, “I don't have complaints. Yes, I do cough, but that's not a problem; I just drink tea with honey and then it's done.”
Some students also said the emphasis in the real patient encounters was less on communication skills and more on medical aspects.
The fact that real patients have more medical knowledge and can judge you on this makes it very hard for me because I have to practice how to make a summary, ask open-ended questions, and respond to the patient, instead of on the medical knowledge.
Many students felt that, had the real patients been better informed about the purpose of the consultation and what help to request, real patient encounters would be more useful than SP encounters to practice communication skills.
[T]he real patient actually suffers from those symptoms and he knows what he is talking about.
Real patient encounters were considered more useful than SP encounters in practicing physical examination skills, mainly because real patients have actual abnormal physical findings. Consequently, students were more motivated and more directed at the physical examination. Students performed fewer physical examinations in real patient encounters because they felt the physical examination was less relevant to the real patient contact, and some of them did not want to burden the patient.
Most students indicated that the real patient encounters seemed more difficult beforehand,
because you just don't know what to expect.
However, some students said the real patient encounters seemed to be easier during the consultation because the real patients talked more spontaneously. Some students, though, regarded the real patient encounters as more difficult because the emphasis was on medical aspects.
Students preferred the feedback from SPs to the feedback from real patients because the SPs' feedback was more specific.
SPs know what they have to pay attention to.
However, even though students preferred the SPs' feedback, they felt that feedback was often far-fetched.
The majority of students felt that the real patient encounter was scheduled too early in the curriculum. Students preferred to have three to four regular SP encounters before the introduction of real patient encounters, although some students preferred to have real patient encounters from the start.
On the questionnaire, first-year students gave comparably high mean marks to the general instructiveness of both real patient and SP encounters. However, there were several significant and practically relevant differences between the evaluations of real patient encounters and SP encounters. The focus group sessions yielded explanations for many of the differences found. Student-doctors, as well as teachers, felt that because some of the real patients were ill informed about the purpose of the consultation and therefore lacked a request for help, the encounters focused more on medical aspects than on communication skills. Consequently, students had trouble practicing the learning goals related to communication skills. In contrast, student-observers found no significant difference between real patient encounters and SP encounters with regard to practicing communication skills. We think this may be due to the inherently different perspectives of the student roles, with student-doctors actually practicing their communication skills, in contrast to observers.
If real patients were better informed about the purpose of the consultation and had requests for help, many students would prefer real patient encounters to SP encounters for practicing communication skills, mainly because of their authenticity. Because the real patient encounters were regarded as more authentic, they also made a more profound impression on the students. Authenticity is thus considered an important advantage of real patient encounters. However, an important disadvantage of real patients we encountered in this experiment is their difficult recruitment and selection.
Although many students would prefer real patient encounters, most students also preferred having several SP encounters before the real patient encounters. Teachers generally preferred SP encounters for teaching communication skills. The phase of the study and the aim of the consultation seem decisive in the use of real patients or SPs in medical education. Further research is needed on the phase of the curriculum in which real patient encounters can best be introduced—for example, a similar experiment in a later phase in the undergraduate curriculum or in a setup in which several SP encounters prepare students for real patient encounters.
Both students and teachers considered feedback from SPs more instructive than the feedback from real patients. This is not surprising because the SPs in this study were extensively trained to provide feedback, in contrast to the real patients. We did not train the real patients to provide feedback because we wanted the encounters to be as authentic as possible.
This study has several limitations. Because the number of patients recruited for the experiment was smaller than we had expected, fewer students than we had initially planned were allocated to have an encounter with a real patient. Because the students in our study were previously informed about the type of encounter (real patient encounter or SP encounter) they were having, they might have been biased by their expectations regarding the encounter. However, in the practice of medical education, students know which patient (real or simulated) they are to encounter. For this reason, we chose to inform the students about the type of patient encounter they were having. Further research is needed to explore the role of expectations of students regarding a patient encounter.
Although the real patients were informed about the encounters with the students, they were not familiar with the setting, in contrast to the SPs. This might have influenced the performance of the real patients. Furthermore, as mentioned earlier, some of the real patients seemed not to understand the purpose of the consultation with the student. This might have affected the perceived instructiveness of the encounter as students felt they could not perform a proper consultation with the patient. Perhaps we should have made the purpose of the encounter more clear to the real patients. Although we did not provide the patients with predetermined symptoms to keep them as authentic as possible, we might have given more guidance. Further research is needed to address these issues.
In conclusion, although both SP encounters and real patient encounters are perceived as being highly instructive, each encounter has specific advantages and disadvantages. Real patients are considered more authentic, whereas their recruitment and selection are very difficult. Advantages of SPs are their usefulness in practicing communication skills and their feedback.
The authors would like to thank Lonneke van Heurn of the Skillslab, Faculty of Health, Medicine and Life Sciences, Maastricht University, for her help in moderating one of the focus groups.
Ethical approval was obtained from the educational management board of the Faculty of Health, Medicine and Life Sciences of Maastricht University.