The methodology applied in this study was of a qualitative phenomenological nature, given our interest in the emic approach to the study subjects. This research adheres to coreq guidelines [46]. Our analysis was based on information gathered from 21 in-depth interviews and two discussion groups staged between January and April 2021, with the aim of knowing the influence of Covid-19 on medication reconciliation. Focusing on this dimension in the interviews enabled us to analyze in greater depth the plots that were subsequently superimposed on the discourses generated in the two discussion groups, and with that, the ideas that condition specific ways of acting, facilitating understanding of difficulties and strategies. If, as Mottier says [47],"stories are the means through which actors try to see the reality of others, suggest certain social positions and impose practices, and criticize alternative social arrangements", then the arguments used to understand each of the reactions of the primary care professionals to the pandemic represent the tools by which each of the actors justifies their position and articulates their self-generated discursive strategy to the rest of society. Thus, an inductive analysis of the arguments and the affinity relationships between them was used to obtain the main arguments deployed in the context of this study phenomenon.
2.1 Participants and study environments
The sampling strategy was theoretical sampling, a technique that was developed by Glaser and Strauss [48] and where the sample is selected through the use of a successive strategy, progressive incorporation of informants, and evidence of similar studies [49]. The theoretical population considered the range of healthcare professionals (primary care doctors and nurses) in relation to medication reconciliation in patients over 65, and the locations where they performed their professional activity, which yielded a sample of healthcare professionals from both rural and urban settings. Our final selection was 13 nurses and 8 doctors, of whom 8 belongedto rural areas and 13 to urban areas.
The choice of these groups was driven only by the search for actors with primary care experience and who had been confronted by situations related to medication reconciliation in patients aged over 65 during the health crisis caused by the Covid-19 pandemic.
For the selection of interviews within each group, a snowball sampling method was used [50]. The sample size was determined progressively during the course of the incorporating research informants until reaching the saturation of the information [51], obtaining a total of 21 interviews.
The interview script covered dimensions associated with the location where the primary care professional works, their professional career, type of healthcare activities performed in their regular primary care practice, the particularities of the geographical environment where they work, the tools and practices applied in medication reconciliation before a patient is admitted to hospital, the medication reconciliation procedures that follow admission, the changes caused by the Covid-19 health crisis in their usual healthcare practice, the perceived difficulties in medication reconciliation, and the influence of Covid-19 on medication reconciliation and the functioning of health services.
2.2 Analysis
We analyzed the discourse in line with the phenomenological paradigm inspired by repetitive "decontextualization" and "recontextualization" processes, in order to understand the experiences and perspectives of the interviewees following the coreq criteria regarding qualitative analysis [46]. This model differentiated three different phases in the processing of the data collected. First, the action or discovery phase, which focused mainly on data collection and reiterated reading of the interviews in order to extract the most repeated topics; this enabled us to obtain different categories and develop related theoretical concepts. Once completed, a new reading of the bibliographic content related to the topic used in this study was carried out to develop an interview guide
The next phase was coding. With the different categories and emerging topics in the interviews now identified, we proceeded to separate the various data obtained. According to the categories established, only the most useful data were selected to help us refine the analysis and extract the greatest potential from the interviews.
Finally, the “data relativization” phase, which not only took into account the theoretical data extracted in the interviews, but also variables such as whether or not these interviews were requested, how our presence could affect the conduct of the interview, the characteristics of the environment in which the interviews took place, and the interviewer's (the authors) own assumptions.
We used the Atlas.ti data management program for the coding and recoding process, to identify all the arguments expressed by the interviewees. This program quantifies the citations for each of the codes assigned during the analysis of the interviews.
2.3 Ethical considerations
The study was approved by the ethical health research committee of the province of xxxxxx, which was assigned code xxxxx. Verbal consent was granted by the participants in the interviews and focus group discussions. Throughout the interview process, the unintended consequences of interviews and focus group discussions were always taken into consideration