Research methodology
We adopted the general inductive approach [16, 17]. A feature of this approach is methodological flexibility: it is not guided by any established qualitative methodologies, but is rather a systematic set of procedures for analyzing qualitative data which can provide reliable and valid findings. It aims to allow research findings to emerge from frequent, dominant, or significant themes inherent in the raw data in the context of focused evaluation questions. To meet this aim, thematic analysis of the data was conducted.
Theoretical framework
This study is focused on the knowledge-cognitive aspects of role conception and role expectation based on Role Theory. We deliberately did not include “role enactment,” which is a cognition-behavior aspect. Although Role Theory considers that a normative role affects role conception and role expectation, and that the mutual interaction of factors with the addition of role enactment makes the roles apparent (Fig. 1) [7], role enactment is greatly influenced by individual behaviors based on their character as well as the history, culture and values of the community, which therefore makes it possible for local characteristics or ‘color’ to be overly represented. Accordingly, we focused here on the knowledge-cognitive aspects of multiple healthcare professionals by elucidating intersubjective facts.
Research method
We adopted FGs, which are intended to elicit open-ended responses conveying thoughts or feelings, as the research design [18, 19]. FGs represent a qualitative data collection approach aimed at uncovering people’s thoughts and values [20]. However, unlike interviews, the researcher plays a peripheral role, acting as a “facilitator” [21]. We therefore adopted the FG approach as consistent with our aim of revealing the credible thoughts and feelings of the participants.
Population and setting
We selected a community in X Prefecture, Japan. JH had worked in this community’s 30-bed public hospital, which provides outpatient services and home visits in the community, between 2015 and 2017. He established relationships with healthcare professionals in this hospital during this time, but no longer worked there. The community is located in a city with a population of 70,000 at the center of the prefecture. The community public hospital provides primary healthcare services that meet community needs, and also cooperates with an advanced treatment hospital located 10 min away by car, as well as other clinics and welfare institutions. In total, the community has around ten medical establishments, including a 36-department, 500-bed prefectural central hospital, a 100-bed hospital with a mix of acute and chronic/rehabilitation beds, a hospital that combines general wards and long-term care facilities, and outpatient-only clinics. In addition, the community also has five home nursing stations and fifty care managers working as healthcare professionals, who primarily support the at-home needs of patients/users within the community-based integrated care system.
Data collection
FGs for healthcare professions in the community were held in January, 2019. The healthcare participants were divided into groups of seven and given 30 min to hold discussions. One of the authors (JH) presented the groups with prompting questions without joining the groups. The other authors (SO, RG, and TM) did not participate in the FGs; instead, staff working at the community general support center played the role of facilitator. Generally, a facilitator is someone who helps a group of people understand their common objectives and assists them to plan to achieve those goals without taking a particular position in the discussion [21]. JH asked the facilitators to closely listen to the participants’ opinions with the aim of finding the right moment to flexibly ask prompting questions. Prompting questions were developed collaboratively by JH, RG, and SO according to the research question and theoretical framework. Thus, the participants were asked main prompting questions on “what image they had of the physicians who they work with” and “what they expect of physicians in a community-based integrated care system”. The FGs were divided so that the participants were discrete in terms of main profession, sex, and affiliation. Since studies using semi-structured guides report that at least three to six focus groups are likely to identify 90% of the themes captured, we asked seven cooperating staff members to divide the group into seven FGs. Because the number of staff facilitators who cooperated in the research was limited to seven [22], seven groups consisting of 7–8 people each were established. FGs were conducted in a quiet room in a community center. All audio records of the FGs were transcribed verbatim, which the researchers commissioned to agencies that specialize in transcribing. Researchers took field notes while observing FGs from the outside and asked the cooperating staff what they noticed during the FGs with regard to non-verbal data [21]. Points noted by the researchers in their observation of participants and by the cooperating staff were also shared during the analysis [21].
JH, SO and TM are general practitioners, and RG is a physical therapist. JH has received training in qualitative research as part of a PhD program, while SO, RG, and TM received this training after obtaining their PhD degrees.
Participants
In this local community, care conferences are held every 2 months to help attendees solve challenges they are facing or learn about new topics. Attendees are mainly welfare staff, with relatively few medical staff. The present study provided these professionals with the opportunity to exchange information with physicians working in the community hospital and clinics. To recruit participants, we informed previous care conference attendees in person or by postal mail that the next care conference 2 months hence would be held as part of the study. We also directly encouraged healthcare professionals who attended a care conference in January 2019 to take part in the study as a local accessibility-based convenience sample [23].
Participants comprised a wide variety of professions, including nurses and pharmacists, as well as administrative staff (including care workers and social workers), occupational therapists, physical therapists, care managers, and medical social workers. Administrative staff (including care workers and social workers) often propose community-wide policies and take care of patients who are welfare recipients or have mental disorders and who have severe care requirements. Care managers are responsible for planning care services provided under Japan’s long-term care insurance system [24]. Facility caregivers assist facility residents with activities such as meals, toileting, and bathing.
At the start of the FGs, they were briefed on the research being conducted and informed that they would be put at no disadvantage if they did not agree to participate. First, the participants completed a list of questions about their personal background, including sex and main profession in the community. They were then divided into seven FGs such that the professions and workplaces of participants in each group differed. The researchers did not participate in the FGs, but managed the whole project. One researcher (JH) contacted the participants via collaborating staff in the community general support center who managed the care conferences. These staff also maintained lists of individual FG participants and confirmed their details.
Analysis
A verbatim record of each FG was analyzed by JH using theme analysis [25]. The validity of the findings was discussed with SO and RG based on “Role theory”. In accordance with the framework of “Role Theory” [6, 7, 9, 26], we focused on the thoughts and values concerning normative roles, and on the exhibited behaviors of physicians witnessed by the participants which influenced their role conception and role expectations of doctor-patient and physicians-other professional relationships. This allowed the identified themes to then emerge consistently across the series of data derived from the FGs. The main focus was on the analysis of verbal data, but observational data obtained by researchers and facilitators were also used as complementary explanatory material. TM then confirmed the consistency of the presented data and the findings, as well as the transferability of the findings. All researchers finally reviewed the data, including the appropriateness of sampling, and critically examined the themes to ensure the robustness of the data and analyses [27]. To ensure the validity of the analysis findings, member checking [28] was conducted among multi-profession participants who attended a care conference in April, 2019. There, JH shared the emerged themes of role conception and role expectation that the researchers had analyzed, and the verbal data associated with these themes. If there was any disagreement or doubt about the analysis, participants were also able to provide opinions through the collaborating staff [28].
Ethical approval
This study was reviewed and approved by the research ethics committee of the University of Tsukuba (No 1353–1). All study participants provided informed consent prior to participation. To protect the anonymity of the participants, quotes in this paper are identified by randomly assigned professions and alphabet codes rather than participant names or initials.