**2.3 Data analysis**

In order to ensure the validity of coding for the analysis, the two researchers were supplemented by another researcher who was not present during the focusgroup interview. To systematize the focus-group data and open statements from the diaries, we used a qualitative inductive content analysis inspired by Georgi's Interpretative Phenomenological Approach as described by Malterud [19]. The decision to use this approach was based on our aim of gathering knowledge about the occupational therapists' experiences with the HE app. Georgi describes four iterative stages: stage 1—encounter with the text; stage 2—identify units of analysis; stage 3—code the units of analysis; and stage 4—categorize and summarize themes and points [19]. In line with these stages, the three researchers first read and reread the interview transcript and the statements from the diaries in order to acquire a good grasp of the data. The researchers also took notes on observed points. Then all three researchers individually identified units of analysis and made notes in the text about the various themes. Thereafter, the researchers met to discuss their interim findings and to reach a consensus on several temporary themes. In the next stage, the researchers individually coded the text into the temporary themes. Subsequently, the codes were again discussed, leading to a final agreement on three main themes: 1) the need to collect multiple types of data, 2) the need for structure versus the need for situated data collection, and 3) the application's influence on the interaction between the occupational therapist and the client. In the final stage, the content of the themes was summarized in writing to provide an overview of the essence and points [19].
