Methodological issues in evaluating measures of health as outcomes for COPD

Abstract

The selection of an appropriate outcome measure depends on the aspect of the disease being addressed and the purpose of the study being conducted. The most fundamental property of any outcome is its ability to measure the biological variable under question. Other properties such as discriminative and evaluative abilities are also important. Specific outcomes measure a single biological variable, such as forced expiratory volume in one second or depression. The specificity of such measures is attractive but requires precise definition of what is being measured and why. Other, summative, outcomes are used to quantify the overall effect of a number of different biological processes. The simplest summative measures are global questions such as "How would you rate your health overall?" Others are complex with many items. If designed and used correctly, these questionnaires can provide an estimate of the overall impact of disease or response to therapy and an index of whether that response was clinically worthwhile. Standardisation of measurements is important to permit comparisons between patients and studies, which makes the measurement of an individual's "quality of life" difficult. The term "health-status measurement" may be better when referring to the use of standardised questionnaires. Utility-based measures help address concerns regarding clinical versus statistical improvement and place outcomes for chronic obstructive pulmonary disease treatment trials in the context of all healthcare treatments.