Comorbidities in head and neck cancer: agreement between self-report and chart review.

OBJECTIVES: To determine the accuracy of self-reported comorbidities compared with medical record review and the clinical and sociodemographic characteristics associated with accuracy of self-reported comorbidities. STUDY DESIGN: We conducted a prospective study of 458 newly diagnosed head and neck cancer patients using self-administered questionnaire and medical chart review data. Overall and itemwise consistency between self-report and chart review was evaluated. Social, clinical, and demographic characteristics of consistent versus inconsistent responders were analyzed. RESULTS: Seventy-four percent of patients had at least one comorbidity. There was good overall consistency between self-report and chart review (kappa = 0.50). Compared with consistent responders, inconsistent responders were found to be older (P < 0.05), have lower sleep (P < 0.05) and physical activity scores (P < 0.05), be more depressed (P < 0.05), and have more severe comorbidities (P < 0.05). CONCLUSIONS AND SIGNIFICANCE: Self-report may be considered as an alternative to chart review for comorbidity assessment in head and neck cancer patients. Younger patients, those with good general health, fewer depressive symptoms, and mild comorbidities, are more likely to give responses consistent with chart review.