A Comparative Analysis of Five Instruments to Measure Control of Asthma

Abstract

The control of asthma is a very important part of an asthmatic's life. Decreasing control can lead to asthma attacks, which can be fatal. For this reason, the researchers have set out to create an instrument to measure control of asthma, and have suggested five possible instruments. Before an instrument can be used, it must be shown to be reliable, valid, and responsive. Reliability will be shown using various intraclass correlation coefficients, depending on the model being used for the data. Construct validity will be shown by how well the instrument's correlation coefficients with other instruments correspond to a priori predictions. Responsiveness will be shown by three methods, t tests comparing the change in changers and stable subjects, a responsiveness index, and Receiver Operating Characteristic (ROC) curves. All five instruments are shown to be highly reliable. No conclusions can be drawn as of yet about the validity, as the a priori predictions have yet to made. For all three methods of assessing responsiveness, the five instruments were ranked identically. In choosing the best instrument, no final decisions can be made, as validity has yet to be shown, but at this time it appears as though the simplest instrument (involving only five questions on the patient's asthma symptoms asked at a visit to a clinic) is also the best as it has high reliability and is highly responsive.