Involving patient partners as appraisers of shared decision making: an explanatory sequential mixed-methods study

Abstract

Abstract Background Patient and public involvement (PPI) is increasingly recognised as a valuable component of health research, yet patients are rarely included in data analysis within shared decision making (SDM) studies. We aimed to assess the feasibility and value of involving patient partner appraisers of SDM using the OPTION5 instrument. Specific objectives were to examine interrater reliability between patient partners and researchers, and to explore patient partners’ perspectives on observed differences in interrater reliability. Methods An explanatory sequential mixed-methods design was used. Four patient partners and three researchers used the validated Danish version of OPTION5 to independently score 102 audio-recorded oncology consultations from two randomised trials evaluating patient decision aids delivered before versus during consultations. Each consultation was scored by one trained patient partner and two researchers. Scores were converted to a 0–100 scale. Agreement was assessed using two-way mixed-effects, absolute-agreement intraclass correlation coefficients (ICC) for single and average measures. Item-level analyses included unadjusted and Bonferroni-adjusted p-values. Three patient partners participated in a semi-structured focus group interview to discuss differences in interrater reliability and to reflect on their assessments based on lived experience. Interview data were analysed using a primarily deductive thematic approach informed by hermeneutic–phenomenological principles. Results The four patient partners appraised 25–27 consultations each. Total OPTION5 scores were median 60.0 for patient partners and 52.5 for researchers. Overall single-rater agreement was moderate (ICC = 0.60, 95% CI 0.46–0.71), and good when averaging raters (ICC = 0.75, 95% CI 0.63–0.83). Item-level differences were small. Although item 2 showed a nominal difference before correction for multiple testing, no item-level differences remained statistically significant after Bonferroni correction. Qualitative analysis revealed three overarching themes: emotion-driven scoring (atmosphere, tone, empathy and recognition), navigating the OPTION5 instrument (growing confidence and calibration over time), and evolving roles as co-researchers. Conclusions Involving patient partners as appraisers of SDM using OPTION5 was feasible, and their appraisal ratings showed moderate reliability with researcher ratings. Findings suggest that, while using the same structured OPTION5 scoring framework, patient partners may draw on broader relational cues when interpreting observable SDM behaviours in clinical consultations. Trial registration Ethical approval was obtained from the Research Ethics Committee at the University of Southern Denmark (approval ID 23/555) for the involvement of patient partners as co-researchers.Plain English summary We wanted to find out whether patients can help assess how well doctors involve patients in decisions about their care during cancer consultations. In this study, four patient partners and three researchers listened to recordings of 102 cancer consultations from two studies of patient decision aids, which are tools designed to help patients take part in decisions. The patient partners and researchers each rated how well the cancer patients in the consultations were involved. Overall, the patient partners’ ratings were fairly close to the researchers’ ratings. This suggests that it is possible for patients to take part in this kind of research assessment. When talking to the patient partners about their experience with the assessment task, they explained that they often noticed things that went beyond the checklist of shared decision making used in the study. For example, they paid close attention to the atmosphere of the conversation, the doctor’s tone of voice, empathy, and whether the patient seemed recognised and understood. They also said that they became more confident in using the rating tool over time. These findings show that involving patient partners in analysing research data is both possible and valuable. Patients may notice important parts of shared decision making that researchers or doctors do not always capture, especially the human and emotional side of the consultation. This suggests that patient involvement in research can add a broader and more real-world understanding of what good shared decision making looks like in clinical practice.