Assessing the validity of EQ-5D-5L in people with head & neck cancer: Does a generic quality of life measure perform as well as a disease-specific measure in a patient population?

Background: Head and neck cancer (HNC) is an important cause of morbidity and mortality globally. Radical treatment methods may result in facial disfigurement and/or functional difficulties with subsequent adverse impacts on health-related quality of life (HRQoL). Guidelines suggest that HRQoL should be measured repeatedly throughout treatment to enable refined treatment protocols and tailored follow-up support but questionnaires are often long and burdensome. We compared condition-specific and generic metrics to assess HRQoL for people with this condition.

Methods: We used data from the prospective Head and Neck 5000 clinical cohort study - 5511 participants with a new diagnosis of HNC between 2011 and 2014. HRQoL data were collected at baseline from 2065 people who completed both the condition-specific EORTC-QLQ-C30 and the shorter, generic EQ-5D-5L questionnaires.

Results: There was strong evidence of association between comparable scales on each questionnaire at baseline: higher levels of functioning and lower levels of reported symptoms assessed with EQ-5D-5L were associated with lower EORTC-QLQ-C30 symptom scores. A moderate relationship (0.61) was found between overall QoL in the EQ-5D-5L index and self-perceived health (EQ VAS).

Conclusions: HRQoL data collected from the generic EQ-5D-5L and cancer-specific EORTC-QLQ-C30 questionnaires are comparable at baseline for people diagnosed with HNC. This would allow a reduced burden of data collection but the EQ-5D-5L may not be sensitive to some condition-specific symptoms. Clinicians and researchers must clarify their aims and outcomes of interest before choosing their HRQoL measures. Further work is required to examine the ability to detect change in these measures over time.