24 Chapter 2 Data analysis The first author verified all transcripts. We used thematic analysis to analyse the collected data [10]. We used an inductive approach to analyse the data, starting with line-by-line coding of the transcripts, using Atlas-ti (version 7.5.18) computer software. During this open coding process, we developed an initial list with codes. All data were coded by the main researcher, HJB, and two members of the research team (BC and FA), and codes were ultimately grouped and combined into subthemes in an iterative manner. We held weekly meetings to discuss disagreements in the coding and grouping processes, until reaching consensus. The last stage consisted of discussions among members of the research team (HJB, BC, FA, SB, PR, TH) until consensus was reached on the final themes. Data saturation was not the aim of this study, as we wanted to explore themes among representatives from major disease groups. All members of the research team work at the University Medical Center Groningen and are affiliated with the Research Center for Insurance Medicine. The first author, HJB, is an insurance physician and PhD candidate; FA has a background in work and organizational psychology; TH and SB have backgrounds in health sciences; and PR has a background in health sciences and occupational physiotherapy. FA, PR and SB have PhDs in the domain of work and health research, and are experienced in conducting qualitative research. Additionally, BC, who played an important role in analyzing the data, was an insurance physician and a senior researcher at the Research Center for Insurance Medicine, with a PhD in work and health. The mixed backgrounds of the team members enriched the analysis by introducing different perspectives. Analyses were influenced by the first author’s experience in conducting actual work disability assessments, and his extensive knowledge on the topic inability to work fulltime. We summarized and searched the texts underlying the themes and codes to find quotes that best illustrate the views and experiences of the interviewees. Quotes from interviewees were selected by two authors (HJB, FA), translated into English by a professional translator, and discussed with all co-authors. To indicate the diversity of opinions while maintaining anonymity, we indicate quotes from physicians with Ph1-Ph10, and from patient representatives with Pa1-Pa9. In the final iteration, we formed a conceptualization based on emerging themes describing the key elements. We used these key elements to operationalize the concept into relevant characteristics, dimensions and measurable indicators, and inventory methods for assessing these indicators. Although the interview guide contained no questions regarding the International Classification of Functioning, Disability and Health (ICF) [11], we were able to identify and categorize responses to this framework. Other sources of categorization were national guidelines on prescribing adequate,
RkJQdWJsaXNoZXIy MTk4NDMw