OBJECTIVES: To examine the content validity of six generic and condition-specific preference-based measures (PBMs) for measuring outcomes of people treated for cancer.

METHODS: A qualitative study design was employed using semi-structured interviews to assess three generic PBMs (EQ-5D-5L, EQ-HWB short form, CS-Base), two cancer-specific PBMs (QLU-C10D, FACT-8D), and one early development measure of positively framed health items (PosiPRO).

Study participants included 96 patients with a diagnosis of haematological or solid tumour cancer from the US (n=49) and UK (n=47). Participants were also classified as currently or recently (within the past year) receiving drug treatment for cancer (n=49) or having received a first drug treatment for cancer more than two years ago but not more than 10 years ago (n=47). Interviews lasted up to an hour and were audio recorded, pseudonymized, and transcribed verbatim.

RESULTS: All measured were seen by patients as useful for capturing appropriate aspects of health-related quality of life. Condition specific measures were evaluated as having more relevant items that generic measures. The FACIT-8D was evaluated as having better content validity in terms of relevancy, and the EQ-5D-5L in terms of comprehensibility. For generic measures, the EQ-HWB-S received the most favourable evaluation in terms of relevancy. PosiPRO items saw only marginally less favourable evaluations of content validity than other measures.

All measures held comparable comprehensiveness, with no measure being seen as not missing any aspects. Recall periods were frequently raised as problematic due to the substantial variation in cancer symptoms in a given timeframe. Patients suggested language and format changes to multiple items across all measures.

CONCLUSIONS: All study questionnaires were generally appropriate and relevant for measuring cancer outcomes, with more favourable evaluations for the FACT-8D, QLU-C10D, and EQ-HWB-S in terms of relevancy. Positively framed items have the potential for enhancing outcomes measurement, subject to further development.