Learner perception of oral and written examinations in an international medical training program

Abstract

Background

There are an increasing number of training programs in emergency medicine involving different countries or cultures. Many examination types, both oral and written, have been validated as useful assessment tools around the world; but learner perception of their use in the setting of cross-cultural training programs has not been described.

Aims

The goal of this study was to evaluate learner perception of four common examination methods in an international educational curriculum in emergency medicine.

Methods

Twenty-four physicians in a cross-cultural training program were surveyed to determine learner perception of four different examination methods: structured oral case simulations, multiple-choice tests, semi-structured oral examinations, and essay tests. We also describe techniques used and barriers faced.

Results

There was a 100% response rate. Learners reported that all testing methods were useful in measuring knowledge and clinical ability and should be used for accreditation and future training programs. They rated oral examinations as significantly more useful than written in measuring clinical abilities (p < 0.01). Compared to the other three types of examinations, learners ranked oral case simulations as the most useful examination method for assessing learners’ fund of knowledge and clinical ability (p < 0.01).

Conclusions

Physician learners in a cross-cultural, international training program perceive all four written and oral examination methods as useful, but rate structured oral case simulations as the most useful method for assessing fund of knowledge and clinical ability.

Introduction

Medical educators around the world have successfully used many different methods of assessing learners, both written and oral [1]. Multiple-choice and essay examinations have been a mainstay at every level of medical education in many countries. Additionally, there is a growing body of evidence that oral examinations, including case simulations in particular, can be important assessment tools in medical education [2–4].

Medical training programs in many countries use oral case simulations as assessment tools [5]. Many recognized clinical skills training programs such as basic life support (BLS), advanced cardiac life support (ACLS), and pediatric advanced life support (PALS) employ case simulations for teaching and assessment [6–8]. In emergency medicine (EM) and other specialties, oral case simulations are used extensively for teaching, assessment [9, 10], and certification [11–14].

Despite the evidence that the use of these and other methods leads to effective learner assessment in various countries [15–17], there has been little published on their use in cross-cultural, international medical training programs [18, 19]. Particularly with oral examinations, the question arises as to whether they can be useful in international programs in which teachers and learners may encounter language barriers or cultural differences.

In this paper we describe the learner perception of four common methods of testing (multiple-choice tests, essay tests, structured oral case simulations, and semi-structured oral examinations) used as part of the needs assessment (pre-testing) and qualification process (post-testing) in an international EM training program in Tuscany, Italy [20]. We also describe the techniques used and barriers faced in the examination process.

Objective

The aim of the study was to evaluate learner perception of four common examination methods in an international educational curriculum in EM: structured oral case simulations, multiple-choice tests, semi-structured oral tests, and essay tests.

Methods

Study design

This was a prospective, observational study using an assessment tool to evaluate learner satisfaction with four different examination methods used in a cross-cultural medical training program. This study was approved for exemption by the Institutional Review Board of the Azienda Ospedaliero-Universitaria Careggi, which is the University Hospital in Florence, Italy.

Study setting and population

The Tuscan Emergency Medicine Initiative (TEMI) is an international partnership involving the Tuscan Ministry of Health, the Tuscan University system, Harvard Medical International (HMI), and the Beth Israel Deaconess Medical Center (BIDMC) Department of Emergency Medicine in Boston, MA, USA. Its goal is to develop an EM training infrastructure for physicians working in the regional hospital system [21]. At the onset, 24 practicing Italian physicians participated in an EM train-the-trainers program based at the University Hospital in Florence, Italy from June 2003 to April 2004. Prior to the start of the program, participants were given written and oral pre-tests in order to evaluate their knowledge base in EM. At the end of the program they were given written and oral post-tests for summative assessment and qualification as EM educators in the region.

Examination methods

The oral case simulations were a pre-test which was used as part of the needs assessment for the project [22]. Ten structured oral case simulations based on clinical scenarios were prepared in a uniform format, with history, physical examination, radiological studies, laboratory results, and visual stimuli available when appropriate. The scenarios and questions asked were scripted in a uniform manner and there were critical actions which needed to be performed by examinees. All written materials were available in Italian and testing was conducted with a medical translator who, in addition to being fluent in Italian and English, was also an emergency physician and a content expert in the subject matter. Candidates were expected to complete three cases chosen randomly from the ten prepared cases. One case was selected for content with which they had adequate prior postgraduate training (internal medicine). Two cases were selected for content with which they had minimal postgraduate training (trauma, surgery, ophthalmology, wound care, etc.). Please see ESM Figure 1 for an example of one of the cases used. Twenty minutes were allotted for each case. For each test, one examiner administered the test and the other observed. The scores from both the examiner and the observer were used for the final score.

The multiple-choice examination was a written pre-test composed of 75 multiple-choice questions selected from various test preparation materials used in the USA and Europe and modified to cover the intended curriculum [22]. The questions were translated into Italian and edited by an Italian clinician for accuracy and local clinical relevance.

The oral semi-structured examination was a post-test that was similar to the oral pre-test, but was not as rigidly structured. The same basic format and testing procedures were used except for the following: Since these examinations were used for qualification purposes, highly experienced examiners not directly affiliated with our training program were brought in as experts to examine the participants. The beginning of each case was structured in a similar fashion as the pre-test with scripted clinical scenarios and prepared materials, but the examiners were allowed more flexibility to ask unstructured follow-up questions to assess elements of the examinees’ fund of knowledge, points of management, and decision-making logic.

The essay test was a written post-test that was composed of four short answer essay questions. Examinees were informed of the general topics to be covered ahead of time (which consisted of the major topics covered in the curriculum during the training program), but the actual specific questions were unknown to the examinees until the time of the test. The test was graded according to whether they addressed the major critical topics correctly.

A learner satisfaction survey was obtained at the end of the training program that asked the Italian physicians to rank the four different examination methods in order of preference according to “usefulness in assessing fund of knowledge” and “usefulness in assessing clinical abilities.” They were also asked to rate the oral and the written examination difficulty levels on a 1–5 Likert scale (anchors of 1 = extremely difficult, 2 = too difficult, 3 = appropriate, 4 = too easy, 5 = extremely easy). They were asked if the written and oral examinations were useful in measuring fund of knowledge and clinical ability (yes/no answers) and whether they should be used in future programs or for accreditation to practice EM in their region (yes/no answers). This was a written survey that was conducted as part of the end-of-the-year course evaluation (ESM Figure 2). Participants were asked if they would give honest feedback to help improve the process for future learners. As a result, the physician learners were blinded to the purpose of the learner satisfaction survey and all responses were anonymous.

Statistical analysis

Because the study outcomes were not normally distributed, comparisons were made using the following nonparametric tests: the Wilcoxon rank sum test was used to compare the difference in the mean content difficulty ratings of oral versus written examinations; the Fisher exact test (due to the small samples) was used to compare survey responses that were in yes/no format; and the Friedman repeated measures test was used to compare mean rankings of examination usefulness in cases when there was one categorical independent variable (examination type) and one continuous variable (mean rank score for fund of knowledge and clinical abilities). All statistical analyses were performed using SPSS version 14.0 (Chicago, IL, USA).

Results

There was a 100% response rate. All 24 participants responded to the survey, but there were data missing from 2 participants (6 answers total). All areas of missing data are noted in the text or tables.

The respondents found the oral examinations slightly more difficult than the written examinations. The mean difficulty rating was 2.75 for oral examinations and 3.00 for written examinations, with a standard deviation of ± 0.45 (Z = −2.45; p < 0.01). There was one respondent who did not answer the question on the difficulty of the oral examinations.

In general, learners liked all testing methods, with the majority of learners responding in the affirmative when asked whether each examination method was valuable for use in future programs and accreditation, and for measuring fund of knowledge or clinical abilities. Only the perceived usefulness in measuring clinical abilities was found to be significantly higher in oral (83%) versus written (67%) examinations (p < 0.01). Please see Table 1.

Table 1 Learner perception of examination usefulness

Using the Friedman repeated measures test, we found a significant difference among the four types of examinations for assessing the fund of knowledge (chi-square 16.42, p < 0.001) and clinical abilities (chi-square 14.23, p < 0.01). Post hoc Wilcoxon (Bonferroni corrected) indicated that the structured oral pre-test was ranked significantly higher than the other three examinations on both measures: fund of knowledge (p < 0.01) and clinical abilities (p < 0.01). No other pairwise comparisons among the other three types of examinations were found to be significant. Please see Table 2.

Table 2 Rank preferences by type of examination: usefulness in assessing fund of knowledge and clinical ability (1 = most useful and 4 = least useful)

Discussion

Learner assessment is a complex process and there are various methods which can be used [1]. Each of the methods described in the literature have their own strengths and weaknesses [15]. Many authors believe that the use of multiple methods of assessment in any one training program can overcome the limitations of individual methods and enhance the overall validity and effectiveness of learner assessment [16, 17]. Furthermore, different methods may be more effective at assessing the different levels of Miller’s framework of clinical assessment [23]. Structured case simulations may provide educators with a better assessment of a learner’s behavior (and therefore predicted clinical performance), rather than simply his cognition [1, 23, 24]. It is important to distinguish the structured and semi-structured nature of the oral examinations we used in this training program from the traditional unstructured oral examinations used in the past in many places, including Italy. Most authors agree that structured examinations have better validity and reliability, with less susceptibility to gender or cultural bias than unstructured examinations [20, 25, 26].

There are multiple examples of training programs in countries throughout the world using case-based oral pre-tests for needs assessment and oral post-tests for assessment of learners or accreditation of physicians [27, 28]. However, there are few descriptions of cross-cultural international educational programs utilizing these same methods. We found oral pre- and post-tests extremely valuable to educators and well-received by learners in this training program and offer our perspective in the hopes that others will be encouraged to incorporate this methodology into similar ventures. International medical education presents a host of unique opportunities: the opportunity to learn, to teach, and to share knowledge beyond preconceived boundaries or borders. With these opportunities come unique challenges: the tasks of bridging language, cultural, knowledge, and experiential differences [29].

In considering the question of whether case-based oral testing can be used in an international training program, we found several barriers that needed to be addressed. The potential language barriers were mitigated by having all written materials translated into the examinees’ native language and having translators knowledgeable in the necessary medical terminology present for oral simulations. It was important to eliminate any possible miscommunication or confusion due to language barriers when assessing the examinee’s fund of knowledge and ability to manage complex case scenarios.

Another potential barrier was cultural. We attempted to create an environment that was professional, conducive to participation, and likely to be culturally acceptable to the examinees. Although the actual examinations were conducted by visiting physicians, we encouraged participation whenever possible by physician leadership from the host country. They gave input into the content covered and procedures used in the examination process. We also used information gathered by observing their actual clinical practice to guide creation of materials and scenarios that were as realistic as possible and sensitive to their cultural expectations. Moreover, after the first class of trainees was qualified, they were trained and used as examiners for the next class. Now that there are enough graduates who have been trained, the examiners are entirely Italian, which has made the administration of the examination easier, without need for translators, and more acceptable to the candidates.

Other potential barriers that needed to be addressed were the preconceived expectations of the learners and teachers involved. It was imperative to understand and appreciate the prior training experience and knowledge base of the physicians participating in the program. The learners in this case were highly trained adult learners, with a significant existing skill set that needed augmentation in specific areas (as they often are in similar projects). In this case they were well-trained in internal medicine and cardiology, but required further training in the acute care aspects of trauma surgery, orthopedics, wound care, pediatrics, ophthalmology, otolaryngology, obstetrics, and gynecology related to the practice of EM. In constructing the examinations for the program, we made every effort to address the areas that we felt needed to be covered in the greatest depth to ensure that the curriculum met the needs of the training program as defined by both the host and visiting country’s leadership.

Limitations

This study has several limitations. Learner perception is by definition subjective and an incomplete measure of an assessment tool. Learner satisfaction with structured oral examinations indicates only that these examinations were well-received, and we cannot make conclusions as to the efficacy or validity of these examinations in assessing the learner. Although there were several statistically significant findings, this study also had a small sample size and larger, more in-depth studies are needed to further investigate the topic. The examinations used in the curriculum were tailored to the specific needs of the program and therefore had not been previously validated. Another problem is that questions on the survey referring to oral versus written examinations sometimes did not distinguish between pre-test or post-test, which could have led to imprecise results. Finally, the external validity of this study depends in part on the applicability of our findings to other cultural environments. Learners who are not familiar with oral examinations may be less ready to accept novel (to them) methods of assessment. Since the learners in this study were primarily physicians working at an academic center, they may have been more receptive to these assessment methods than physicians working in a different setting would have been.

Conclusions

Physician learners participating in a cross-cultural, international training program in EM perceived all examination methods as useful, including structured oral case simulations, multiple-choice tests, semi-structured oral examinations, and essay tests. Learners ranked the structured oral case simulations as the most highly rated testing method and felt that oral examinations were better at assessing their clinical abilities when compared to written examinations. Oral case simulations can be useful assessment tools in an international medical training program. The results of this study may be useful in guiding the development of training programs in countries with similar educational goals and clinical practice environments.