The Uniscale Assessment of Quality of Life: Applications to Oncology

Abstract

The measurement of quality of life (QOL) is an important challenge in clinical research, not only because QOL is one of the two main endpoints of effectiveness of treatments, the other being overall survival, but also for taking clinical decisions shared with the patient. Today, QOL is generally assessed using psychometric questionnaires; nevertheless, these latter suffer from several shortcomings that often lead to unreliable results. About 14 years ago, the Italian Group for the Evaluation of Outcomes in Oncology (IGEO) planned a research program articulated in two phases. In the first phase, domains of QOL and problems connected with it were defined performing a content analysis of the interviews of 248 Italian cancer patients, based on four areas related to the foundations of quality of life. The domains/problems referred by the patients were submitted to a large population of more than 6,000 Italian cancer patients so as to assign a frequency to the relevance of each domain and to the presence/absence of each problem. In this study, a uniscale evaluation of QOL, using a Visual Analogue Scale (VAS), was also obtained. The rating of each patient was classified in “bad QOL” and “good QOL,” if the chosen point fell in the 0–30 or 70–100 interval, respectively; the other scores (30–70 interval) were not considered. The relationship between the uniscale assessment of QOL and the presence of each problem was investigated. The impact of each problem, adjusted for the presence of the others, and for the patient’s characteristics, was detected by a unifactorial analysis using logistic additive models, where “good” and “bad” QOL were in turn considered as dependent variables. Thirteen of 19 problems were significant, and this confirms the external validity of the uniscale assessment of QOL. A VAS can be considered a reference point in multidimensional QOL scales and should still be regarded as a useful and synthetic tool to investigate phenomena related to the patient’s QOL. In this perspective, more research on the psychometric properties of this instrument, in the context of the assessment of QOL, is still needed.