Methods

Study design

This qualitative exploratory study, situated within an interpretivist paradigm, was designed to generate in-depth understanding of a process that has received limited empirical attention. To this end, individual semi-structured interviews were conducted to examine how PROMs are currently experienced in oncology care in the Netherlands, which bottlenecks are encountered, and where in the implementation process they emerge. This study is reported according to the Standards for Reporting Qualitative Research [23]. Ethics approval was obtained from the Research Ethics and Data Management Committee of Tilburg Universtiy (approval code: REDC 2024.83) and the study was conducted in compliance with the ethical and data management regulations of the school. Written informed consent was obtained from participants prior to data collection.

Interview guide

The semi-structured interviews followed a predefined interview guide, allowing the interviewers to follow core topics while remaining flexible to capture additional insights raised spontaneously by participants. The interview guide was based on the communication-centered framework of De Ligt et al. [19], grounded in Lasswell’s classical communication model [24], which conceptualizes communication processes through the questions “who does what, to whom, in what form, and with what effect.” (See Fig. 1 for an overview of the framework¹). Building on these principles, our interview guide asked participants to identify the steps they have experienced in PROMs-related activities, and reflect on who was involved, what actions were taken at different points in the process, in what form information was communicated, and with what intended and perceived effects. Participants were encouraged to discuss their experience related to PROMs, challenges encountered, and desired solutions or opportunities at each step of the process. Prior to data collection, two interviewers (LH and AT) conducted practice sessions to test the clarity of the interview guide to ensure that it elicited the type and depth of information needed for the study. The interview guide is provided in Online Resource 1.

Fig. 1: A communication-centered framework of PROMs implementation [19]. Reproduced from De Ligt et al., J Med Internet Res 2025; 27: e60777 (https://www.jmir.org/2025/1/e60777). © 2025 De Ligt et al. Licensed under CC BY 4.0

Participants

Participants included three groups: (1) clinicians working in oncology departments of hospitals and cancer centers in the Netherlands; (2) PROMs implementation staffs, such as policy advisors and coordinators; and (3) cancer patients or survivors. Clinicians and staffs were recruited through the research team’s professional networks using purposive and convenience sampling to ensure variation in institutional type, oncology specialties, and roles in PROMs implementation. Patients were recruited through a Dutch online platform (Kanker.nl) for individuals with lived experience of cancer.

A total of 7 professionals and 10 patients participated in the interview. Professional participants represented a range of oncology specialties, including head and neck oncology (n = 2), gynae-oncology (n = 1), neuro-oncology (n = 3), and gastrointestinal oncology (n = 2)², working across four different organizations in the Netherlands: two general hospitals, one comprehensive cancer center, and one specialized radiotherapy institute. This group also included experts involved in PROMs implementation (n = 1) and policy advisory work (n = 1). Years of clinical or implementation experience ranged from 4 to 20 years. In these settings, PROMs were typically administered automatically through electronic health record (EHR) portals or separate online platforms. Clinicians used PROMs results during consultations when accessible; the implementation specialist coordinated PROMs use and dashboard management, and the policy advisor coordinated instrument selection and consulted with national quality registries.

The patient sample included individuals at different stages of cancer journey: some were in active treatment (n = 6), others had completed treatment (n = 3), and one participant did not specify their treatment status. Participants represented diverse cancer types, including breast cancer (n = 6), prostate cancer (n = 1), rectal cancer (n = 1), and thyroid cancer (n = 2). They were treated in at least seven different institutions across six provinces, with no systematic overlap with professional participants’ settings. Time since diagnosis ranged from 4 months to 10 years. Patients varied in their prior experience with PROMs. Most had experience with quality-of-life or symptom questionnaires on multiple occasions as part of routine care, while two patients recalled only pre-treatment screening forms or general health questionnaires. When patients did not spontaneously distinguish PROMs from other questionnaires, we probed them to reflect on these specifically.

Sample size was guided by the concept of information power, a theory for determining sample size in qualitative research that emphasizes the specificity and diversity of information rather than numerical saturation [25]. In this study, information power was supported by the combination of a focused study aim, a guiding theoretical framework that structured the inquiry, the inclusion of participants with complementary perspectives on the same implementation process (patients, clinicians, and implementation/policy professionals), and the specificity of the PROMs implementation context in oncology care.

Data collection

Interviews were conducted between January 2025 and July 2025 by two interviewers (LH and AT). Interviews with professionals were conducted in English, interviews with patients in Dutch. The interviews lasted 30 to 60 min each and were conducted online via Microsoft Teams.

Interviews were audio-recorded and transcribed. Dutch transcripts were translated into English for analysis. Participants were offered the opportunity to receive a copy of their interview transcript to check for accuracy. One participant reviewed and confirmed the accuracy of the transcript; no changes were requested.

Data analysis

Transcribed data were analyzed using reflexive thematic analysis [26, 27]. We adopted a hybrid approach combining deductive and inductive coding [28]. The deductive component entailed that when coding, we used the framework’s stages [19] as coding categories where applicable. For example, content about how purpose was communicated or whether results were discussed was coded in relation to the corresponding stage. This did not involve a predetermined codebook. Alongside this, inductive codes emerged freely for content that did not map onto the framework. Themes were then developed inductively by grouping codes, and subsequently mapped (but not forced) onto the framework to examine where they aligned with, or extended beyond, the sequential stages.

Two researchers independently coded an initial subset of transcript and compared their coding to align on the analytical approach. Subsequent coding was then performed in an iterative process over multiple reads of the transcripts. Coding and data management were performed using ATLAS.ti (version 26) [29]. The first author developed a preliminary thematic structure by grouping codes with shared meanings, which was discussed and refined with all co-authors across multiple rounds. In line with the principles of reflexive thematic analysis [23], we acknowledge the researchers’ positioning in shaping the analysis. The interviewers (LH and AT) have backgrounds in health communication and PROMs communication in oncology, which informed their sensitivity to communicative processes but may also have oriented attention toward communication-related challenges. The author team includes expertise in PROMs implementation in clinical practice (KdL), whose perspectives helped balance the analytical lens during the collaborative discussions in which themes were refined.

Footnotes

1. The horizontal and vertical dimensions of this framework informed the overall structure of the interview guide. The specific content presented within the cells was not used as predefined themes or codes; interview remained open-ended. ↩

2. Some professionals work across multiple oncology domains; therefore, these categories are not mutually exclusive and do not add up to n = 7. ↩