Before data aggregation, we assigned equal weights to all objective indicators, expert judgments, and patient feedback. We used MATLAB to develop CBB1007 computerized program to calculate and aggregate all indicators and subjective evaluations automatically. For illustration, the ER-based Intelligent Decision System (IDS) (Xu et al., 2006) was used to show the hierarchical structure of the medical quality assessment framework based on objective indicators, expert judgments and patient feedback. Fig. 1 shows the hierarchical structure of the quality assessment framework in IDS.
Fig. 1. The medical quality assessment framework modeled by IDS.Figure optionsDownload full-size imageDownload as PowerPoint slide
After combining all objective indicators and subjective evaluations, we got distributed assessments about the yearly medical quality of the studied hospitals from 2006 to 2010 (see Table 6).Step 5: Rank medical quality of the three hospitals.
To rank medical quality of different hospitals, bolus is desirable to generate numerical values that are equivalent, in terms of expected utility, to the distributed assessments. For this purpose, the utilities of individual assessment grades need to be defined first (Yang & Xu, 2002). In our study, we used quality scores to define such utility values of different assessment grades. More specifically, we assigned a quality score of 100 to excellent quality, 80 to good, 60 to average, 40 to poor, and 20 to worst. In this way, a distributed assessment can be transformed to a quality score. Finally, we ranked medical quality of the three studied hospitals on the basis of the computed combined quality scores.