ASSESSMENT OF QUALITY-OF-LIFE IN HEAD AND NECK-CANCER PATIENTS

Seventy-five consecutive patients were selected to evaluate a disease- specific quality-of-life questionnaire (UW QOL). The new test was comp ared to two established equality of life evaluation tools, the Karnofs ky scale and the Sickness impact Profile (SIP). Each test was administ ered on three separate occasions: (1) several days preoperatively; (2) immediately postoperatively; and (3) 3 months postoperatively. The Ka rnofsky scale is relatively crude and lacks the ability to measure sub tle changes. The SIP is a detailed questionnaire that is quite sensiti ve to change. However, due to its length, the SIP is inefficient and e xpensive to administer, and patient noncompliance is often a problem. The three questionnaires were compared according to the following fact ors: Acceptability. 97% of the patients favored the UW QOL scale compa red with the SIP because it was more concise and easier to complete. V alidity: Validity indicates the ability of the test under investigatio n to measure what it was intended to measure. Using the SIP as a gold standard, the UW QOL scale demonstrated an average criterion validity of 0.849, whereas the Karnofsky average criterion validity was 0.826. Reliability: Reliability is a measurement of the reproducibility of th e data. The UW QOL questionnaire scored >0.90 on reliability coefficie nts versus 0.80 for the Karnofsky and 0.87 for the SIP scale. Responsi veness: Responsiveness is the ability of the test to measure clinical change. The UW QOL scale faired better than the Karnofsky and the SIP scale in detecting change. The UW QOL scale is comparable to the Karno fsky and SIP scales when tested for validity and reliability. It was t he preferred test format of 97% of patients and provided the greatest responsiveness to clinical change. (C) 1993 John Wiley & Sons, Inc.