Study design and ethical approval
This multiphase design study was conducted to develop the marine medicine syllabus for medical sciences students, which includes three phases. The first phase was a literature review to identify concepts and contents of marine medicine in published studies. The second phase includes a content analysis of interviews with marine medicine experts. The third phase was a Delphi method that includes two consecutive rounds to obtain consensus from the marine medicine expert panel. The graph of the study is shown in Fig. 1.
This research has been approved by the Ethics Committee of Baqiyatullah University of Medical Sciences (IR.BMSU.REC.1399.006) and conducted in Tehran, Iran from 2019 to 2022.
First Phase: Literature review
At this phase, a literature review was used to investigate the main subjects of marine medicine, its areas and important issue in this field in published studies.
The searches were conducted in the Web of Knowledge, Pubmed and Scopus and the Iranian databases SID and Magiran databases. The search terms include “marine”, “maritime”, “sea”, “seafarer”, “navy”, “naval”, “sailor”, “diving” and “medicine”, “medical”, “health”, “treatment”, “care”, “disease”, “sick”, “ill”, “disorder”, “poison” and “toxin”. The reference lists of articles were also reviewed using forward and backward citation tracking to identify other eligible documents. The search was limited to human studies between 1990 to 2021.
Inclusion and exclusion criteria
All studies that contained dimensions or concepts related to marine medicine were included in the study, regardless of the study design. Exclusion criteria included not being related to marine medicine, non-English language, and lack of access to the full text.
In the initial search by two researchers, 4314 possible articles related to marine medicine were found. A total of 589 studies were excluded due to being duplicates. Duplicate studies mean studies with the same title, names of authors and published journals. The title and abstract of 3725 studies were assessed and 3088 studies whose title and abstract were not related were excluded from the study. The full text of 637 studies was reviewed, of which 575 articles were excluded due to irrelevance or duplicate information. The remaining 62 studies were used to extract topics and concepts related to marine medicine. The flowchart of reviews process is shown in Fig. 2.
Two researchers (FH and AVA) independently examined the full text of the remaining selected papers for extracting data. All the data related to marine medicine that could be used in the main and sub topics of the marine medicine course were extracted. Any ambiguities or disagreements between the evaluators were resolved through discussion and consensus by a third person.
Second Phase: Content analysis
At this stage, a qualitative content analysis research method was conducted. Qualitative content analysis is the analysis of the content of narrative data, and it is a flexible method to identify prominent subthemes and patterns among themes. In the conditions of lack of enough knowledge about a phenomenon (such as marine medicine) or knowledge fragmentation, the inductive approach is recommended  to explain and interpret the data and elaborate the dominant and major themes of participants’ experiences .
The study participants consisted of experts and those who have experience and knowledge about marine medicine. The participants were recruited in the study using a purposeful sampling method. Also, maximum variation in sampling was considered in terms of age, working experience, degree and working field. Inclusion criteria were having experience in marine medicine, having a master’s degree or doctorate in the field of medical sciences, the ability to express experiences and desire to participate in the study. The exclusion criterion was the unwillingness to continue the study. Sampling continued until data saturation was reached.
Semi-structured interviews were used to collect the data. The research setting was the workplace of the participants, which the participants chose based on their preference and the researcher tried to help the participants express their experiences of marine medicine easily and freely by visiting at the chosen time of the participants and creating a safe environment and gaining their trust. The researcher obtained written informed consent from the patients before the interviews. The research questions were extracted by the discussion of the research team. The questions were open ended, neutral, clear, and relevant to the objective of the study. Four members of research team were working in the field of marine medicine which ensured the relevance of the interview questions to the research question and six members of the research team had the experience of conducting previous research interviews, which ensured the accuracy of the questions.. The Interview started with an grand tour question: what is marine medicine? Based on the answers, the interviews continued with probing and follow-up questions. All nonverbal behaviors, facial expressions and eye contact were paid attention during the interview. The duration of the interviews was between 43–72 min and an average of 56 min based on the physical and psychological conditions of the participants and the quality of their responses to the questions. All interviews were audiotaped and transcribed verbatim.
The information obtained from the interviews was analyzed by conventional content analysis with Geranheim’s method. This method includes the following steps: 1-Writing down the interviews 2-Choosing the unit of analysis 3- Getting a general view of the entire interview by reading the units of analysis 4- Reading word by word and line by line and coding each meaningful unit (obtaining primary codes) of related concepts 5- Placing similar codes in a subcategory by constantly comparing codes. 6- Collecting similar subcategories in a category and naming them .
In this study, Guba and Lincoln’s criteria were used to increase trustworthiness, which are known as the gold standard. For this purpose, long-term involvement with the data and spending enough time to collect and analyze the data were used. Also, all the interviews were conducted under the supervision of the supervisor, and during the data analysis, the codes and categories that emerged were presented to the supervisors and advisors to review the cases and present and apply their revisions. To increase reliability, all documents and evidence related to the data were securely stored to help readers and external observers to trust the data. Also, a number of coded interviews along with the extracted categories were given to other researchers familiar with qualitative research methods to check the accuracy of coding and classification. In order to increase transferability, the researcher tried to provide in-depth and rich descriptions of the research context and word-for-word quotes of the participants in order to increase the fit of the theory with the context under study.
The found topics in the literature review and content analysis of interviews were combined and formed the initial draft of the marine medicine syllabus, which was validated with the Delphi method in the next phase.
Third Phase: Delphi method
In the third phase, a Delphi method was used to obtain the consensus of experts in the syllabus of marine medicine. Delphi is a valid and scientific method to provide comprehensive and reliable information about a controversial issue or to reach a consensus on an issue, which is done through quantitative and qualitative processes . The purpose of this method is to make an expert judgment about a question. This method is based on the assumption that a group of experts with different attitudes create more valid results than the judgment of one person, even if he is the best in his field . One of the applications of the Delphi method in health care is in the field of education and curriculum development .
The Delphi method requires at least two rounds, and if the first round is open-ended, the required number of rounds is three . In this research, the initial draft of the marine medicine syllabus was compiled in the previous phases, and there was no need to do the first round, therefore, this study was conducted in two rounds.
The Delphi panel consisted of 18 experts from health system policy makers, managers, university educators, health care providers and researchers who were well informed and experienced in the field of the marine medicine. Inclusion criteria were having at least five years work experience in marine medicine field, having a master’s degree or doctorate in the field of medical sciences and desire to participate in the study. The exclusion criterion was not completing the questionnaire and unwillingness to continue the study.
The main goal of the Delphi method is to reach a consensus . Consensus should be determined before starting Delphi . The most common method of reporting consensus is the percentage of agreement, which is defined as a proportion of participants who agree with a certain rating range . The consensus is not fixed  and has been determined from 51 to 100% in different studies . Like many studies [22,23,24,25,26,27,28], we also considered the cut-off point of 80% agreement as reaching consensus, and the main and sub topics that after each round of Delphi, by less than 80% of experts obtain very agree or agree, were removed from the syllabus.
At first, the researchers contacted the experts and obtained their consent to participate in the study. Then the Delphi questionnaire was sent to them. The Delphi questionnaire was structured from concepts found from previous phases for a rating on a 5-point Likert scale (1-strongly disagree, 2-disagree, 3-neutral, 4-agree, 5-strongly agree). The face validity of the questionnaire was confirmed by faculty members. Interrater reliability between panelists in first round was estimated based on the Intraclass Correlation Coefficient and 0.944 Cronbach’s alpha indicated an excellent interrater agreement. Intraclass correlation coefficient (ICC) is a standard tool to determine the interrater reliability of more than two raters based on interval scaled data . The five-point rating scale used in this study is interval-scaled. Interrater reliability between panelists in the second round was estimated based on the Intraclass Correlation Coefficient and 0.810 Cronbach’s alpha indicated a good interrater agreement.
After the completion of the first round of Delphi, items that had less than 80% consensus between the panelists were removed from the syllabus and topics suggested by the experts were added to the list for judging in the second round. In the second round of Delphi, topics that achieved 80% or above consensus and suggested topics consisted of the second-round questionnaire and was given to the expert panel. After the completion of the second round, the items that had less than 80% consensus among the participants were removed and the remaining topics formed the final syllabus of the marine medicine.