All participants participated in the survey after providing written informed consent. This study was approved by the Institutional Review Board of Hallym University (HIRB-2018-049).

In order to identify the key character elements that doctors should have and to create a character education model for Korean doctors, we conducted a literature review and case analysis. Based on this, a 3-round Delphi survey and consultant group (CG) meetings were conducted (Fig. 1).

The participants of the Delphi survey were Korean professors of medical education, physicians, experts from nursing schools, and a head nurse. The first round of the Delphi survey consisted of 5 open questions: the necessity of character education in medical education, the personal character required by doctors, what is considered “character,” the problems and failures of character education in current medical education, and the key character elements for doctors in the fourth industrial age. Participants were asked to list 10 or more key elements, numbered from 1 to 10 in order of importance.

The second round of the Delphi survey was conducted among respondents who participated in the first round. The questions focused on assessing respondents’ degree of satisfaction with the core elements of character education extracted from the first round and with the level of graduates regarding those skills. The respondents could also suggest corrections or deletions of the core elements, and provide additional comments. The third round of the Delphi survey was only conducted among respondents who reported discordant responses in the second round of the Delphi analysis, and consensus on the second-round responses was reached.

For the Delphi survey, the manual for Delphi participants and consent form were distributed together. The content included the background and purpose of the research, subjects, research methods, duration of the research, a statement that participants could withdraw from the study, the disadvantages if the study was not done, personal information, and a statement on confidentiality.

To create the Delphi questionnaire, discuss the results of the rounds of the Delphi survey, and set directions for future initiatives, a CG was assembled with a total of 17 participants, including 2 character education specialists, 3 medical education professors, 2 professors from nursing school, 2 physicians, 2 members of the general public, 1 head nurse, and 6 medical students. Some of them (3 medical education professors, 2 physicians) also participated in the Delphi survey. The questionnaires used are presented in the Appendix 1.

The medical education experts selected for the Delphi survey were drawn from professors of medical education at 40 medical schools in Korea, to ensure the greatest possible representativeness. The medical education experts were also selected from the past and present board members of the 2 major representative institutions leading medical education in Korea, the Korean Society of Medical Education and the Korean Institute of Medical Education and Evaluation. Therefore, 67 medical professors and 63 members of the institute were selected for the Delphi survey. Eleven physicians and 2 experts from nursing schools were added to the list based on suggestions from the CG. A total of 143 subjects were selected for the Delphi survey. The number of Delphi experts is generally between 30 and 100; however, a higher number was chosen given the busy schedules of medical professors and to prepare for possible drop-out of participants during the multiple survey rounds. Three rounds of the Delphi survey were conducted between September 2018 and February 2019. In the first round of the Delphi survey, 47 (32.9%) of the 144 invited subjects replied. In the second round of the Delphi survey, 38 (80.8%) of the 47 respondents in the first round of the Delphi survey participated. In the third round of the Delphi survey, only 26 respondents participated, corresponding to 100% of those who disagreed in the second survey (Table 1).

Descriptive statistics were calculated, such as the frequency, mean, and standard deviation.

The first round of the Delphi survey asked open questions about the necessity and problems of character education in medical education. There was 1 non-response, and all 46 respondents stated that character education was essential in medical education. Regarding the problem of character education, there was a considerable number of opinions based on factors such as a knowledge-oriented educational system, lack of an appropriate curriculum, lack of concern with individuals’ character as an evaluation factor in the course of becoming a doctor, inadequate awareness of the need for character education among professors and students, and insufficient teaching ability to handle character education properly.

Some respondents also expressed the opinions that the concept of character is ambiguous, and that there is a lack of research on what content to teach. In particular, many respondents pointed out that medical school professors or students do not value character education, but only academic performance. The raw data are available in Supplement 1.

In order to define the concept of character required by doctors, this question was included in the first round of the Delphi survey. Forty-seven respondents gave a variety of comments, which could be summarized as follows. The character that a doctor requires is the basic attitude, values, and mindset that must be present to perform his or her duties. These include respect for human beings, empathy and consideration for patients, a sense of calling, honesty, ethics, and responsibility.

In the first round of the Delphi survey, a total of 420 elements were collected from respondents’ lists of the 10 key elements required for medical students. The content of each element was analyzed, and similar elements were grouped into 17 items, the top 10 of which were selected for inclusion in the major classification.

After the second and third rounds of the Delphi survey and the CG meeting, we were able to identify 8 core categories of character education: service and sacrifice, empathy and communication, care and respect, honesty and humility, responsibility and calling, collaboration and magnanimity, creativity and positivity, and patience and leadership. These core elements, together with their sub-elements, are presented in Table 2.

Fig. 2. shows the 8 core categorical elements for the character education of medical students derived from the Delphi survey and the CG. The sub-elements with similar concepts were integrated and the final sub-elements were selected to fit the 8 core categorical elements, and their definitions are shown in Table 3. The final names and definitions of the 8 core categorical elements and the sub-factors of each core element were selected through expert meetings.

In the third round of the Delphi survey, medical graduates’ levels of the 8 categorical core elements were rated on a scale ranging from very low (1 point) to very high (5 points) (Fig. 3). The average level of graduates for the 8 categorical core elements was 2.45–3.46. The core factors that were rated highest were ‘responsibility and calling’ (mean=3.46, standard deviation=0.57), while ‘patience and leadership’ (mean=2.45, standard deviation=0.55) was the lowest.

The appropriate academic year for instruction on the 8 core categorical elements was investigated, and the results are shown in Table 4. Many respondents indicated that the core elements could be instructed not in any particular year, but in any or all academic years of the medical education process.

The majority of the respondents said that ‘service and sacrifice’ are necessary throughout the premedical years (n=31) and that ‘honesty and service’ are appropriate for all years of medical education (n=24). Many also said that ‘honesty and humility’ and ‘patience and leadership’ would be appropriate to focus on in the senior years of medical education, with the general opinion that education on all 8 core elements should be available in all years.

Some of the respondents to the first round of the Delphi survey failed to respond to the second round, resulting in a decrease in the number of subjects. Although the Delphi survey itself was conducted among a small number of specialists, the loss of 9 of the 47 primary Delphi respondents could be a limitation for generalizing the results of the study. In particular, only a model of the key factors of character education for medical students was developed. Furthermore, research was not conducted on educational content or methods that could be used at actual instructional sites.

In the future, it will be possible to develop various research initiatives and programs using the core factor model of personality education developed in this study. Our suggestions for follow-up research and applications are as follows:

When developing and implementing medical education, medical schools refer to standardized guidelines, such as “Learning outcomes of basic medical education” issued by the Korean Association of Medical Colleges [15]. Although each university has certain goals and objectives, it is necessary to include specific levels of education and essential content in order for students to qualify for the medical licensure examination and for medical schools to receive accreditation. It is expected that this research will provide basic guidelines for character education by identifying the core competencies of character that are essential for future doctors, providing a basis for each university to develop specific character education courses suitable for that institution.

This study provides a basis for diverse related studies. For example, if core competencies for character education are identified, a variety of educational content could be developed for the competency of “altruism” alone. In other words, since educational content, instructional methods, and evaluation methods may vary depending on the core competencies of character education, it will be possible to investigate which method is more suitable and to examine their effects when such programs are actually implemented. The study investigated the level of these competencies in graduates and the appropriate academic level of education for each core competency, and further studies could be conducted among more faculty members of medical schools, physician groups, and patient groups. By doing so, it will be possible to determine which core competencies are lacking among medical school graduates, to investigate when it is appropriate to provide education on those competencies, and to use those findings to develop character education programs.

Mentoring methods for character education can be considered, because mentoring has already been found to be a desirable educational methodology for fostering personality traits, community awareness, a sense of responsibility, service spirit, interpersonal skills, and moral reflection, while bringing about overall growth [16]. Under the Character Education Promotion Act in Korea [17], personality education is emphasized for college students, as well as for students in primary and secondary schools, and various character education programs have been developed. An analysis of the literature on medical humanities curriculum for college students in Korea [18] found that the main components were self-understanding and socializing, emotional control and self-esteem, and connection with others.

However, medical school students have different personality traits from those required by general college students. For example, among the factors found in this study, respect for life and vocational consciousness were identified as personality traits that are more important for medical students than for general college students. Therefore, the program developed is not suitable for general college students.

In conclusion, based on the core elements of character education identified in this study, we expect to be able to develop character education programs or mentoring modules suitable for medical students.