5.5Can Medical Interview Skills Be Taught?

Kraan, H. F., Crijnen, A. A. M., de Vries, M. W., Zuidweg, J., Imbos, T., & van der Vleuten, C. P. (1990). To what extent are medical interviewing skills teachable? Medical Teacher, 12(3–4), 315–328.

Teaching Interviewing Skills 

The medical interview is the medium for the exchange of information between the patient and the physician and further operationalizes the physician-patient relationship (Lipkin et al., 1984). The interviewing skills of the physician in large measure determine to what extent communication between patient and physician takes place and therefore directly bear on what physical and mental problems are presented (Goldberg & Huxley, 1980).

Traditional medical skills need detailed data collecting by means of a medical history and depend on the quality of the physician’s interviewing skills. The physician’s competence in medical interviewing then relates to his capacity to solve medical problems (Elstein et al., 1978), and to enhance patient satisfaction and compliance with medical regimes (DiMatteo & DiNicola, 1982).

In response to the recognition of the importance of the quality of medical interviewing to patient care, there has been a rapid growth in programs that teach interviewing skills in medical curricula (Kahn et al., 1979). The effectiveness of these teaching programs is, however, unclear and research has presented conflicting evidence (Carroll & Monroe, 1980; Sanson-Fisher et al., 1981). For example, what is known about the ‘natural course’ of interviewing skill acquisition in medical schools which did not provide interview training?

Without interview training, medical students tend to drift to a more directive interviewing style

Studies carried out in these settings demonstrate a tendency for students to drift to a more directive interviewing style over the years. At the end of their training, these students gathered more factual and ‘organic’ information while psycho-social issues received less attention, as shown by their infrequent use of open-ended questions and their offering of less reassurance, support and empathy to the patient (Helfer, 1970; Scott et al., 1975). Other less controlled studies reported that students’ interviewing style persisted throughout the years of clinical training (Wright et al., 1980).

Training and interviewing style

What is our knowledge about the influence of training programs on the interviewing style? After a course in communication techniques Werner & Schneider (1974) reported that students showed a significant increase in the ability to empathize and to ask exploratory and affect-related questions. Stillman et al. (1977) also showed that interviewing skills improved after 1 year of training students using actors that simulated patient’s problems.

Several other investigators (Sanson-Fisher & Maguire, 1980; Engler et al., 1981), however, noted that after some initial improvement resulting from training programs, a subsequent decrease in interviewing skills occurred. Kauss et al. (1980) further found little evidence that the training of interviewing skills regardless of the duration and quality of teaching resulted in significant long-term gains.

What are we to make of these mixed results? Interviewing skills and teaching methods were poorly defined in these studies, rendering it difficult to discern which interviewing skills actually improved or worsened and under what teaching conditions. Methodologically, studies are needed that clearly describe the training conditions and present subsequent student interview behavior in quantifiable and replicable results.

In the present study we make such an attempt and report the effects of a continuous 4-year training program with simulated patients on the interviewing skills of 563 medical students throughout their 6-year undergraduate curriculum.

Measuring Students’ Progress in Interviewing Skills

The students’ progress is measured using well-defined interviewing parameters:

1. Collecting patient-centered information;
2. Obtaining present and past medical data;
3. Providing information about the illness condition and giving advice to the patient;
4. Opening, phasing and closing the interview;
5. Promoting effective information exchange and establishing rapport with the patient. 

These five theoretically distinct categories of interviewing skills were accentuated and evaluated during the training program. Evaluation results demonstrated differential effects of teaching on each of the five categories, thus yielding a more detailed answer to the question: To what extent are medical interviewing skills teachable?

Teaching program of medical interviewing skills

A substantial part of the 6-year medical curriculum at the University of Limburg Medical School in Maastricht, The Netherlands, is devoted to training medical interviewing skills. Students receive intensive feedback of videotaped encounters with simulated patients (Kraan et al., 1986). Training takes place in small groups of a maximum of eight students, guided by a physician and a behavioral scientist. Students grow accustomed to interviewing simulated patients and being observed or videotaped.

Students grow accustomed to interviewing simulated patients and being observed or videotaped

During the first 4 years of the medical curriculum the small groups meet 15 times a year for 2 hours. During the fifth and sixth year there is no formal training in interviewing skills. The purpose of using simulated patients is to allow students to practice medical interviewing skills from the first pre-clinical years on. 

This teaching approach continues unchanged over the 4-year training program, but each year a different set of interviewing skills is accentuated. 

• In the first year, basic interviewing skills are stressed, such as asking open- and closed-ended questions, discussing emotional topics, empathizing and summarizing. 
• In the second year, attention is paid to medical history-taking skills and to obtaining information about psycho-social issues. 
• The third year, the program focuses on the effective exchange of information between physician and patient. 
• In the fourth year, the students learn to manage interviews in ‘difficult’ situations, such as with aggressive patients, dying patients and patients with sexual problems. 

Effectiveness of teaching programs of this kind is generally claimed (Carroll & Monroe, 1980). The students’ interviewing skills are evaluated each year along with other medical knowledge and skills as part of the examination system.

Growth of Interviewing Skills – Discussion

The data of our study suggest that the five categories of medical interviewing skills are differently influenced by a continuous small group teaching format with expert and peer review of videotaped encounters with simulated patients. In skills like History-taking, Presenting Solutions and Structuring a positive teaching effect is clearly discerned. To a lesser extent, the same holds for the skills used for Exploring Reasons for Encounter, but their acquisition is suppressed by counteracting influences to be discussed later in this section. In the learning of Basic Interviewing Skills, adverse influences decrease the net effect of the described teaching format to nearly zero from the first to the fourth year. To analyze the effects of teaching and their counteracting influences we look more closely to the five categories of interviewing skills.

Exploring Reasons for Encounter

Figure 1 shows the scores of students in the section Exploring Reasons for Encounter over the course of Medical School. In the second year, there is a significant decline in the number of times students perform these interviewing skills compared to first year. During the next two years there is a gradual return to the first-year level. Sixth year students almost perform on the lowest level, comparable with the second-year students.

The curve of the skill to Explore the Reasons for Encounter (‘patient-centeredness’) agrees with often heard remarks (Helfer, 1970; Scott et al., 1975) that in the course of the curriculum students omit more and more to ask information from the patients’ frame of reference, unless they are trained and stimulated to perform these interviewing skills. Students perform these skills relatively well in their first year.

An explanation may be their lack of medical knowledge, restricting their medical interviewing to the patients’ frame of reference. Low scores at second- and third-year level seem to be the result of the rapid growth of knowledge and of the strong motivation of students to solve medical problems. The preoccupation with medical problem solving may hinder students to follow the patient in his own ‘medical story’ and related emotional cues. Comparison of the levels of first- and sixth-year students throws light on the natural history of these patient-centered interviewing skills. The highly significant decline from the first to the sixth year-group, both relatively untrained, suggests that these skills gradually impair during the medical curriculum.

Patient-centered interviewing skills tend to deteriorate over medical school, but can be regained with communication training

The finding can be explained as an effect of the clinical rotations where students are focused on obtaining factual information related to medical problems. This may lead on the one hand to more efficient data collection and on the other hand to a one-sidedness in retrieving data from the patient. Information from the patients’ frame of reference is not obtained. These results agree with the findings of Helfer (1970), who compared the interviewing skills of randomly selected freshmen and senior medical students. He found that freshmen asked fewer leading questions and raised more issues related to interpersonal aspects of the patients’ problem than the seniors. Nevertheless, the incline from second to fourth year level suggests that these skills can be learned under the influence of an intensive interviewing course. Werner & Schneider (1974) also demonstrated an – at least temporary – increase in the number of exploratory and affect-related questions during a medical interview after a course in communication techniques.

History-taking

Figure 2 shows an interesting pattern about the course of History-taking skills. In the second year a significant decline from the first year can be seen, followed between the second and fourth year with a rapid and significant growth in history-taking skills. Fourth year performance is significantly better compared to first year performance. The large standard deviation in the first year decreases in the course of the curriculum. The performance of the sixth-year students approximately equals that of the first year.

The growth pattern of medical History-taking is rather erratic. The first-year students’ relatively high score is reflecting the extensive and sometimes unselective reviewing of systems during history-taking. The remarkable sharp fall between the first and second year may be a result of the increasing body of knowledge and the yet insufficiently developed problem-solving skills. A hesitating style of interviewing replaces the learned schematic style of history-taking and reviewing of systems, characterizing the first-year students. After this transient effect students achieve a high level of performance from the second year onwards.

The second year of medical school is hard on students: emerging problem-solving skills are very demanding on their problem-space at the cost of interviewing skills

Further integration of knowledge and problem-solving strategies with history-taking skills appears to be necessary for adequate performance. The difference between the fourth- and sixth-year levels may be explained by the lack of systematic training of the latter. The viewpoint that the growth of medical history-taking skills during the curriculum is due to the growth of medical knowledge does not hold. Comparison of growth patterns in medical knowledge and in history-taking skills shows a linear increase of the former during medical school (Sprooten & Van der Vleuten, 1983), whereas the curve of the history-taking skills has a different pattern. Our results disagree partly with those of Barbee et al. (1967) who found a rapid improvement in history-taking skills during medical school. However, the decline at second year level and the incline during the next years agree strongly with the findings of Wright et al. (1980), who described exactly the same pattern of growth.

Presenting Solutions

Figure 3  marks the significant growth in the ability to Present Solutions. The ability of the sixth-year students (clerkship) is significantly lower than that of the fourth year.

The improvement of the skill to Present Solutions (negotiation about problem definition and solutions, providing information and giving advice) may be due to a combined result of the increase of medical knowledge and the ongoing interviewing course during medical school. The results of the sixth-year students suggest that they acquire these skills somehow in a ‘natural way’ during their clinical rotations. However, in absolute terms the scores of the total group of students are not very high.

Presenting Solutions are poorly developed over Medical School and require a real effort by teachers for improvement

The results point to the necessity of special teaching of the skills of negotiation and information provision during both last curricular years. In the literature no research evidence about educational efforts to improve the skills of Presenting Solutions could be found.

Structuring

In Figure 4, a significant increase in students’ capacity for Structuring the interview can be seen. The ability of the sixth-year students is slightly lower than that of the fourth year.

Between their second and fourth year most students seem to acquire the skill to Structure the medical interview. The large standard deviation in the third year indicates that some of the students are already able to structure the interview, while others are still far from this. Towards the end of the medical curriculum the standard deviation becomes smaller, whereas a high scoring level on this skill is attained in the fourth year.

Though very demanding, most students manage to structure the interview with their patient

Wright et al. (1980) found the same pattern of growth in their study. However, the relatively high scores of the sixth-year students suggest that these skills will be partly learned by ‘experience’, in other words in the absence of formal training.

Basic Interviewing Skills

In Figure 5,  the fluctuation in performing Basic Interviewing Skills is shown. The decline from the first to the second year is significant. The growth from second to third year is also significant, as is the decline from the third to the fourth-year level. However, the level of performance of the first compared to the sixth year did not show a significant difference. The standard deviations decrease over the course of the curriculum.

Examination of Figure 5 leads to the conclusion that the net effect of the training on Basic Interviewing Skills is not high. The scores of first year students are already relatively high. In this study design (lacking a control group for the first year) it cannot be discerned if this result is due to an effect of training or to a pre-existing, ‘human’ ability to reflect emotions, ask open questions, empathize, avoid technical terms and summarize. The decline during the second year may be attributed to the earlier described disturbing effect of growing medical knowledge, and rising interest in medical problem solving, which may hamper these pre-existing Basic Interviewing Skills. The improvement in the scores of the third- and fourth-year students may be attributed to the effect of the training.

Students benefit from interview training and don’t undergo the decrease in Basic Interviewing Skills, the natural history of these skills over Medical School

According to studies of the ‘natural history’ of Basic Interviewing Skills (Helfer, 1970; Scott et al., 1975) we would expect a decrease in quality during medical school, when no interviewing course is provided during the curriculum. Several authors (Werner & Schneider, 1974; Stillmann et al., 1977; Sanson-Fisher & Maguire, 1980) reported an – at least temporary – improvement in Basic Interviewing Skills as result of training. Our findings suggest that students benefit from interview training in the sense that they don’t undergo the decrease in quality which is described as the natural history of Basic Interviewing Skills and maintain the relatively high level of the first year.

To sum up 

• To sum up we found that interviewing skills like History-taking, Presenting Solutions and Structuring improve as a result of a continuous small group teaching format with expert and peer review of videotaped encounters with simulated patients. Structuring the interview may also improve without formal training.
• To a lesser extent, effects of training are seen in the skill Exploring Reasons for Encounter, but its acquisition is easily suppressed by the growth of knowledge, the increasing ability to solve medical problems and the influence of clinical rotations.
• The net effect of the described teaching format on the Basic Interviewing Skills starting from a relatively high level during the first year, is nearly zero as a result of the aforementioned counteracting influences.

Methodological considerations 

In the light of these findings some remarks about methodological restraints should be made.

Quasi-experimental Design

In this quasi-experimental (Cook & Campbell, 1979) cross-sectional design, each year group of students is compared with its preceding year group, which serves as a control group. So, the compared groups differ—as to medical interviewing skills—in 1-year training. In using cross-sectional data for longitudinal purposes, as in this study, decisive inferences are only allowed when three assumptions are fulfilled.

• First, the ‘treatment’ (the training in interviewing skills) should be kept constant over the years. As described earlier the teaching format was held constant, but in each year a certain category of interviewing skills received special attention. The sixth year-group had some training in interviewing skills, but not with the intensive and systematic format of the other year-groups. Their skills levels are mainly the result of ‘natural growth’. Moreover the growing medical knowledge and increasing experience in medical problem solving of students may be influences that confound the assessment of the real effects of training in interviewing skills, although other studies (Stillmann et al., 1977; Wolraich et al., 1981; Sprooten & Van der Vleuten, 1983) suggest that an increase in medical knowledge does not entail an improvement of interviewing skills.
• Second, selection effects may threaten the quasi-experimental design. Selection is, however, not very probable because students were not assigned to the year-groups on the basis of their interviewing skills.
• Third, disturbing instrumentation effects (e.g. invariance of the instrument or inconsistency of measurement in different cases) are to some extent probable.

Case-effects

Aforementioned studies with the MAAS (Crijnen & Kraan, 1987), as well as other studies of medical interviewing (Swanson et al., 1981) and of medical problem-solving (Elstein et al., 1978) show so-called case-effects. These case-effects are systematic differences in measurement of competencies, when different cases are used. These effects generally have three sources: differences in the nature of the medical problem, in the patterns of complaints and in the way of self-presentation by the patient (Crijnen & Kraan, 1987). In this study the latter two sources are controlled because of the use of a pool of simulated patients. Nevertheless, some case effects may be present.

With regard to these three assumptions for the use of cross-sectional data for longitudinal purposes, it can be argued that they are reasonably fulfilled. Causal inferences, however, about the effects of training on medical interviewing skills should be drawn with care. This study could be replicated in a stronger longitudinal design with a control group of students without training. Such a replication is, however, far less feasible, because of the impossibility of recruiting a control group without training from the same medical school.

Concluding Remarks

Nevertheless, the results with our study design suggest that intensive, continuous group teaching format with interviewing of simulated patients and with expert and peer reviews throughout the whole curriculum (including clinical rotations) is necessary to obtain a growth in the students’ History-taking, Presenting Solutions and Structuring skills.

However, with this teaching format only a maintenance of acceptable levels in the skills to Explore the Reasons for Encounter and the Basic Interviewing Skills is possible. More effective ways of teaching these categories of skills should be looked for.

Summary  Growth patterns of medical interviewing skills during a 6-year undergraduate curriculum are assessed by studying 563 medical students taken from five year-groups interviewing simulated patients. In a cross-sectional, quasi-experimental design, their skills are rated by means of the Maastricht History-taking and Advice Checklist (MAAS-General), an observation instrument which measures five categories of interviewing skills pertaining to initial medical consultations.

The findings suggest that the skills for History-taking, Presenting Solutions and Structuring of the Interview are effectively learned. These learning effects result from a continuous small group teaching program with expert and peer review of videotaped encounters with simulated patients.

In a medical curriculum, History-taking, Presenting Solutions and Structuring can effectively be learned

The teaching effects of this program seem less for the skills pertinent to the phase of Exploring the Reasons for Encounter and to the Basic Interviewing Skills, because the students’ growing medical knowledge and the increasing ability to solve medical problems exert a counteracting influence on the acquisition of these easily deteriorating skills. The results might be helpful to curriculum planners in order to make their programs for medical interviewing skills more effective.

Exploring the Reasons for Encounter and Basic Interviewing Skills, however, may deteriorate when medical problem-solving skills kick in.