Qualitative study to inform the design and contents of a patient-reported symptom-based risk stratification system for patients referred from primary care on a suspected head and neck cancer diagnostic pathway.

Objectives: This study aims to inform the development of a patient-reported symptom questionnaire for head and neck cancer and outline the requirements for a patient-reported symptom-based risk stratification system. The study objectives are to explore how clinicians ask questions and decide subsequent steps for patients referred with suspected head and neck cancer; the language patients and clinicians use to describe symptoms; how clinicians reassure and discharge low-risk patients; and identify clinician and patient experiences of the head and neck cancer diagnostic pathway and their views on a novel diagnostic pathway using patient-reported symptom-based risk stratification.

Design: The study employed qualitative methods including observation and recordings of clinic consultations and semistructured interviews with clinicians and patients. Analysis proceeded concurrently with data collection using a rapid qualitative analysis approach.

Setting: Three acute UK National Health Service Trusts with variation in service delivery models. Data collection took place between April and October 2023.

Participants: One hundred and fifty-six adults referred for suspected head and neck cancer, and 21 clinicians from different subspecialties were recruited. A subset of recruited patients (n=16) and clinicians (n=13) were interviewed. One observation of a general head and neck clinic was conducted.

Results: The findings highlight types of symptoms and the language used by patients and clinicians to describe these symptoms in clinic consultations. During interviews, patients described the need for in-person support and human clinical decision-making, an accessible system for reporting their symptoms and reassurance regarding the security of patient data. Clinicians discussed the need for risk scores to be sufficiently validated to be trusted, the potential clinical usefulness of a risk score-based system, for example, to support triage by discriminating symptoms, and accessibility for patients. The observation highlighted inconsistent and sometimes unclear referral information and the limited time clinicians have to read referral information.

Conclusion: The findings have implications for the development of a patient-reported symptom-based risk stratification system. As well as ensuring patients can understand the language used, it will be important to consider how their emotional needs can be met. The findings also have wider implications for understanding the impact of language on emotionally evocative healthcare interactions.