Reliability and validity of an observer-rated disfigurement scale for headand neck cancer patients

Background. Facial disfigurement is considered to be one of the most distre ssing aspects of head and neck cancer and its treatment, but it has been th e focus of little systematic study. Existing studies have yielded conflicti ng results about the psychosocial impact of disfigurement. No studies to da te have examined disfigurement using a valid and reliable observer-rated me asure. The purpose of the current study was to examine the validity (conver gent and discriminant) and the inter-rater reliability of a novel nine-poin t observer-rated disfigurement scale. Methods. The sample consisted of 74 ambulatory head and neck cancer patient s more than 6 months post treatment. Ratings of disfigurement were assigned independently by surgical and non-surgical raters. Validity was assessed b y comparing the association between disfigurement ratings and sociodemograp hic and illness treatment variables, Reliability was assessed by examining the concordance between the surgical and nonsurgical ratings. Results. Disfigurement ratings were not associated with several sociodemogr aphic variables, supporting the discriminant validity of the scale. Disfigu rement was significantly related to a diagnosis of oral cancer, a history o f adjunctive radiation, the type of surgical procedure performed, the degre e of physical dysfunction, and the presence of postoperative complications. Observer ratings of disfigurement were significantly related to patient ra tings of disfigurement. These findings support the convergent validity of t he disfigurement scale. Inter-rater reliability of the scale was high (intr aclass correlation coefficient = .91). Conclusion. The study provides preliminary evidence for the validity and in ter-rater reliability of a novel nine point observer-rated disfigurement sc ale that may be useful in evaluating the impact of disfigurement on quality of life in head and neck cancer. (C) 2000 John Wiley & Sons, Inc.