gms | German Medical Science

Bedside teaching (BST) is a teaching method in medical education which offers students the opportunity to practice history taking and clinical examination with real patients under the supervision of physicians [1]. Comparing studies from last decades rather a decrease of BST can be found in the USA [1]. At the same time a decline in clinical abilities of prospective physicians with regard to physical examination is criticized [2], [3], [4]. The explicitly “practice and patient oriented” German Medical Licensure Act (ÄAppO) from 27 June, 2002 [5] attaches great significance to BST. Besides a mandatory minimum amount of 467 teaching hours, it is furthermore regulated in the ÄAppO that BST should be executed in equal terms of practical instructions with the patient present (with three students) and of patient demonstrations (with six students). Despite comprehensive administrative regulations regarding BST which are binding, leeway remains for its practical realization, especially regarding content design. Whether the BST which takes place complies with the sense of the German Medical Licensure Act and which significance it actually has for the education of medical students in Germany, has hardly been investigated so far.

While numerous recommendations and instructions for BST can be found in the literature [6], [7], only a few studies have examined the learning effects of different educational concepts for BST empirically. Studies indicate that BST is only well received by students when planned carefully and when adequate educational concepts are applied, and in this case it can lead to superior learning effects compared with other teaching methods [8], [9], [10]. Concerning patient selection for BST by teachers, a qualitative study showed that teachers mainly strive to select patients according to defined learning objectives in order to ensure an optimal learning context [11]. However, this selection process can be influenced by several “biopsychosocial” and “structural” factors [11]. For example, patients with diseases which require isolation or with marginal German language skills proved to be selected less frequently for participation in BST. Thereby, students are deprived of the didactic involvement with certain clinical patterns or difficult history-taking situations, respectively, for which BST could represent an important mode of learning if teachers are trained accordingly. In another qualitative study teachers stated busy wards and the difficulty to make students articulate their learning requirements clearly to be the greatest difficulties in BST [12].

Apart from these, no other studies were found in the literature which investigated commonly used practice regarding the organisation and the patient selection during daily hospital routine. Presumably specific approaches are needed in order to meet different basic hospital requirements. Therefore, the aim of this study was to investigate the mentioned characteristics in order to extract aspects which provide possible starting points for a structural and organisational optimisation of BST and which could be incorporated in teacher trainings.

For data acquisition concerning teacher characteristics, the organisation of and criteria concerning the selection of patients as well as difficulties regarding the execution of BST a questionnaire for teachers with closed and open questions was developed. In the first part of the questionnaire, the sociodemographic variables age, gender, occupational position, previous teaching experience in years as well as the number of lessons taught per trimester are inquired (6 items). The second part contains questions whether guidelines for patient selection exist and when the allocation of teachers to lessons as well as patient selection are carried out (3 items). In the third part, aspects of the individual prioritization of certain selection criteria (6 items) as well as the subjective importance of certain exclusion criteria concerning patient selection are inquired (17 items). In order to determine priorities regarding patient selection, six criteria which were identified to be important in [11] were provided to be ranked by allocating numbers from 1 (“most important”) to 6 (“least important”). In order to appraise the subjective importance of exclusion criteria for patient selection, teachers were asked to indicate the probability to exclude a patient actually fitting well for BST learning objectives due to 17 characteristics or situations on a scale from 1 (“very unlikely”) to 5 (“very likely”). These items were also worded on the basis of previously identified factors [11]. The aspect “fitting of the patients with the learning objectives” which had been identified as main factor in that study was selected and supplemented by some of the additionally identified minor criteria from “biopsychosocial” and “structural” domains as well as by further aspects, which had been denominated by students and teachers within the context of the evaluation. In the fourth part, teachers were asked in an open question to report about the subjectively most important difficulties and obstacles related to BST. The questionnaire was reviewed by two regular BST teachers and was valued as relevant and comprehensible.

The final questionnaire was sent to a total of 134 teachers from the departments of surgery (n=74), internal medicine (n=45) and psychiatry (n=15). These three departments comprise a major amount of BST in the undergraduate curriculum at Hamburg medical school and feature a range of different communication and practical skills to be practiced. In this study, the subject surgery was represented by teachers from the departments of abdominal and trauma surgery and the subject internal medicine by teachers from the departments of gastroenterology, cardiology, nephrology, oncology and pneumology. In the Hamburg undergraduate medical curriculum, courses are organized in thematic blocks of 12-week trimesters starting in semester five. In order to promote peer teaching, trimesters are completed in a self-chosen sequence, bringing students from different semesters together in the same courses.

To describe the sample, cases were analyzed unweighted and also weighted according to the number of BST hours taught per trimester, in order to gain a representative insight into teacher characteristics dependent on the actual frequency of their hours. In order to obtain evidence of possible biases due to drawing of the sample, teachers from the sample were compared with all contacted teachers from each discipline, respectively, regarding gender and occupational position. To analyze group differences related to central tendencies and frequency distributions, requirements for the use of parametric methods were tested primarily. As distortions of test statistics could not be excluded, all group comparisons were conducted using Kruskal-Wallis- or Mann-Whitney-U tests, respectively, as well as Fisher’s exact tests, to provide consistency and comparability also in the few cases in which parametric methods would have been justifiable. For all inferential statistical analyses a level of significance of p≤0.05 was chosen [13]. Additionally, effect sizes are provided for all tests. For group differences of interval scaled data the difference between the means divided by the pooled standard deviation d was calculated, for group differences of ordinal scaled data the effect size r according to Rosenthal [14] and for comparisons of frequency the phi coefficient Φ. The free text answers concerning difficulties and obstacles related to BST were analyzed using inductive creation of categories. The selection criterion was provided by the question, with respect to the level of abstraction a combination of abstract main categories and inherent, as concrete as possible subcategories was chosen. The reliability of the categorical system was verified formatively and summatively by both authors.

Fifty-one completed questionnaires were returned, which equates to a total response rate of 38.06% (internal medicine 55.56% (n=26), psychiatry 40% (n=6), surgery 25.68% (n=19). With regard to the representativeness of the sample, no biases were found concerning gender or occupational position. On average, the teachers in the total sample were 36.02 (SD: 6.34) years old and nearly three quarters of them were male (see table 1 [Tab. 1]). Approximately half of the sample consisted of residents, about a third were attending physicians and the rest were consultants. On average, teachers had been teaching BST for about six years with 8.58 lessons per trimester. After weighting the total sample by the number of BST lessons, an amplification of already existent age differences showed up to the effect that surgical and psychiatric teachers, respectively, are significantly or almost significantly younger than internal medicine teachers with large effect sizes (internal medicine vs. surgery: p=.000, d=1.47; internal medicine vs. psychiatry: p=.100, d=.91). Furthermore, in internal medicine attending physicians taught most frequently, while in contrast residents taught more frequently in the other two departments (internal medicine vs. surgery: p=.005, Φ=.47; internal medicine vs. psychiatry: p=.035, Φ=.86). In addition, teachers in internal medicine possessed longer teaching experience, accordingly (internal medicine vs. surgery: p=.003, d=.85; internal medicine vs. psychiatry: p=.035, d=.86).

While all teachers from internal medicine and psychiatry stated that they already receive their teaching schedule several weeks before the BST starts, surgery teachers receive their schedule only several days before their BST lesson (42.11%), on the same day of the lesson (36.84%) or shortly before the lesson (10.53%). Accordingly, teachers from internal medicine and psychiatry mainly select their patients several hours before the lesson (82.14%), while most surgery teachers select their patients not until shortly before the lesson (57.90%). The majority of teachers (83.33%) across all departments stated that their patient selection is guided by predefined learning objectives, especially disease patterns, but beyond that, no further requirements for patient selection seem to exist (68%).

With regard to the patient selection for BST, two criteria were prioritized across all three departments by far (see table 2 [Tab. 2]): patient consent for participation in the BST as most important criterion (median=1), followed by the fitting of the patient’s symptoms or disease, respectively, with the learning objectives (median=2). Compared with surgeons, psychiatrists put significantly more emphasis on good German language skills with medium effect size (p=.036, r=.042), and teachers from internal medicine compared with surgeons almost significantly did so too with medium effect size (p=.061, r=.38). Furthermore, internal medicine teachers attached greater importance to patients’ consent than teachers from surgery (p=.040, r=.41). With regard to the other four aspects, no significant differences could be detected and pairwise effect sizes only reached the size of a small effect or less.

When selecting patients for BST, non-consent for participation constitutes the most important exclusion criterion for teachers (see table 3 [Tab. 3]). Further criteria which were at least rated with a mean probability of “4” are the expected non-availability of patients due to examinations at the time of BST, an acute deterioration of the patient’s state of health as well as patients’ spontaneous non-consent upon arrival of the student group. Comparing departments, the only discovered significant difference was that an acute deterioration of the patient’s state of health was indicated as a likelier exclusion criterion by surgeons compared with internists (p=.015, d=.66). An approximately significant difference, also of medium effect size (p=.066, d=.53), hints at the possibility that surgeons possibly exclude patients from BST more likely who are not yet aware of their disease’s severity and its prognosis. With regard to the other four aspects, no significant differences could be found and effect sizes only reached the size of a small effect or less.

Three main categories emerged from open answers regarding aspects teachers had most difficulties with regarding BST (see table 4 [Tab. 4]): lack of time and human resources (27 entries), problems on the behalf of students (17 entries), and problems regarding patient selection (16 entries). As the most important subcategory additional “time constraint” resulting from the integration of BST into already existing clinical routines was mentioned 12 times.

In our sample, BST in internal medicine was carried out to a significant greater extent by older teachers (attending physicians), while in surgery mainly (younger) residents taught BST. The latter was also mentioned as one of the problems in the open answers. Although BST was traditionally taught by consultants and attending physicians, the increasing clinical workload is associated with a shift of lessons towards younger residents in the US [15]. In order to come up against this trend without loss of quality, programs for teacher training were developed by several American groups which already address medical students from higher semesters as well as interns and residents [15], [16], [17]. The fact that BST in internal medicine, in contrast to surgery, is still taught mainly by attending physicians at the German university hospital in this study despite developments observed in other countries, could also be due to the fact that in this subject aspects of clinical reasoning are of great importance, which can be better structured by more experienced teachers [18].

Regarding patient selection, patient consent to participation as well as the fitting of the patient’s symptoms or disease with the learning objectives proved to be by far the most important criteria across departments from the teachers’ point of view. The latter confirms a previous mere qualitative result [11]. Predominantly, teachers seem to follow predefined learning objectives and aim to comply with them, which results in better learning effects [8]. Although asking for patient consent to participate in BST is not mentioned explicitly in teacher guidelines for BST, it has such a high priority for teachers that patients’ non-consent constitutes the most important exclusion criterion across departments. Accordingly, the situation in which a patient, despite prior consent, refuses to participate in BST at the arrival of the student group also proved to be an important exclusion criterion. Given the right of self-determination towards medical actions, BST without patient consent would constitute a massive offense against law and ethics. Given the on-going debate, especially in North America, e.g. concerning the physical examination of narcotised patients, a particular reference to this aspect appears to be essential despite its professional implicitness [19], [20].

The request for patient consent to participate in medical teaching corresponds with recommended guidelines [21] and also leads to a higher satisfaction among participating patients [22]. However, at the same time missing consent can lead to difficulties in BST planning if not enough patients are willing to participate, which was mentioned as a problem by several teachers. With regard to this problem alternative teaching methods with equal learning objectives could be considered for curricular planning which prospectively lead to similar learning results and evaluations [9].

As a subject-specific difference regarding patient selection especially psychiatrists, but also internists put greater emphasis on patients’ German language skills. This could be due to a stronger BST focus on exploration as learning objective in psychiatry, history taking in internal medicine and physical examination in surgery. With respect to increasing numbers of migrants, exploration of and history taking with non-German-speaking patients should be integrated in the curriculum in order to acquire the special requirements in the contact with this patient group [23]. Apart from the mentioned aspects, further differences could not be found neither for the inclusion nor the exclusion criteria. This suggests that the importance of the majority of criteria is valued to the same extent by teachers across disciplines and therefore training concepts only have to be adapted to specific subjects regarding the differentiating aspects mentioned above.

With regard to difficulties with BST, time constraint proved to be the most frequently mentioned individual subcategory. This has also been reported for other clinical fields of teaching where teachers have to provide clinical care at the same time [24]. Although the double challenge of clinical tasks and teaching does not necessarily lead to a loss of teaching quality [25], a permanently scheduled BST program where teachers are relieved from patient care during lessons can lead to a higher satisfaction among teachers and students [12]. Non-compliance with time schedules and organisational rules as well as a lack of motivation were mentioned among others as problems regarding students’ side. Interestingly, in another study similar interpersonal and communicative factors in their positive occurance in teachers were stated by students to be important for the successful acquisition of the physical examination [26]. For this matter, as well as regarding different levels of the students’ previous knowledge which were indicated as a problem although they constitute an integral part of our peer-teaching-based curricular model of the clinical years, further analyses should be conducted in order to eliminate this problem perceived in BST. A major part of further denominated difficulties, for example, missing rooms for briefing and debriefing or the ambiguous communication of teachers’ schedules, are important identified parameters in order to modify organisational and communication structures for BST. In combination with a teacher training course specifically tailored to BST requirements [6] this could contribute to a reduction of mentioned problems.

For the interpretation of this study’s significant findings, a potential source of error, especially for multiple group comparisons, results from a lack of correction on the α-level. When calculating the cumulated α-error according to the formula α_cum = 1- (1- α)^k, the probability to receive at least one significant result by mistake is 0.37 for the sociodemographic comparisons, 0.26 for the priorities regarding patient selection and 0.58 for the exclusion criteria. Therefore, special relevance has been given to the effect sizes for the appraisal of the results.

Furthermore, with regard to teachers’ answers, it cannot be excluded that aspects were not valued according to their own assumptions but according to social acceptability. As participation in this study was voluntary, a selective sample containing more dedicated teachers cannot be excluded. In addition, this study was only conducted at one university hospital which makes generalizing of the results more difficult.

As very often only keywords were given as open answers concerning problems regarding BST, a completely unambiguous allocation to one of our generated categories was not possible in three cases. These had to be excluded from analysis.

For this study, except for surgery, only teachers were contacted who actually teach BST. As lessons in the department of surgery are distributed ad hoc, all colleagues had to be informed about this study, so that the total response rate has to be estimated higher.

Regarding the design of BST, the findings suggest that general as well as subject-specific aspects for its execution and for patient selection can be identified which could be relevant for a successful and learning objective oriented implementation of BST. With regard to the planning of content and organisation of BST and also for the design of faculty development programs structures of organisation and communication as well as subject-specific characteristics should be considered especially. Some of the parameters identified as problematic suggest that further studies which also include students’ and patients’ perspectives on BST could contribute additional aspects to the elimination of structural, organisational and difficulties with regard to contents, for an optimal design of BST regarding curriculum and didactics.

The authors thank all teachers from Hamburg University Medical Center who participated in this study.

This project was approved by the Vice President of the Ethics Committee of the State of Hamburg Physicians’ Ethics Board and complies with the ethical standards of the Declaration of Helsinki.

The authors declare that they have no competing interests.