Outcome measures for research in adult women with symptoms of lower urinary tract dysfunction

The purpose of this communication is to offer the clinical and research communities an initial attempt at incorporating outcome measurements within identifiable domains, as well as providing initial information as to the reliability of those measurements most commonly used. Scientific evaluation of the outcome of therapeutic interventions is based on assessment before and after treatment. However, the reliability of methods and measurements used in the evaluation is often poor or unclear, which makes interpretation of outcome difficult. The reliability of a test depends on the accuracy and reproducibility of the method. The diagnostic accuracy is determined by verifying test results against a reference (‘‘gold’’) standard that defines true disease status. The predictive value of a measure is considered the most important. Since it may be impossible to establish a final true diagnosis, reliability must in some cases be measured by reproducibility, which is determined by comparing results of repeated examinations of the same patient.1 Reproducibility for tests where the result is given on a continuous scale may be expressed as bias with the 95% confidence limit [Bland and Altman, 1986], while for binary tests the kappa coefficient, which adjusts the observed agreement for expected chance agreement, is used [Gjorup, 1997]. Urinary incontinence is defined as ‘‘a condition in which involuntary urine loss is a social or hygienic problem and is objectively demonstrable.’’ It represents a multidimensional phenomenon with wide-reaching effects which may be quantified within various areas or domains. These areas or domains include: