Staging and Stratification

The challenge of developing outcomes management systems rests in adequately defining and stratifying starting points and in accurately quantifying endpoints. In sinusitis, for example, a great deal of attention has been focused on the appropriate method to stage a preoperative computed tomography (CT) scan of the paranasal sinuses. Yet, the factors that comprise a good staging system (i.e., validity, reliability, statistical distribution, comprehensiveness, and the ability to predict outcome) are rarely considered or objectively evaluated. In two studies of clinician-based staging systems,1'2 we have found wide variations in reliability (both intra-rater and inter-rater) and the lack of substantial correlations to postoperative outcome. Nevertheless, for consistency in reporting results, it is necessary to choose a staging system. Table 18-1 describes several proposed staging systems that have been evaluated. In the comprehensive evaluation by Metson et al.,2 which included staging systems by Kennedy,3 Friedman et al.,4 and

May et al.,5 the Harvard system6 and the Lund-MacKay staging systems7 were found to be the most reliable, whereas the Harvard system was the easiest to use. In 1996, in a report of the Rhinosinusitis Task Force Committee Meeting, Drs. Lund and Kennedy recommended the modified Lund-MacKay staging system for further outcomes research.8 Ultimately, the best staging system will be chosen by evaluating outcomes studies to determine which factors are most predictive of the result. That will not be possible until additional data are collected.

0 0

Post a comment