The risk ratio was 5.32 for hospital admission among patients with PP, which exceeded this threshold. This is in contrast to the same analysis for the initial AT-PP measurement prior to standardized asthma treatment, in which the mean Wilcoxon AUC-ROC was 0.571 (95% CI, 0.27 to 0.87) [Fig 2, bottom, B, inset). The AT-PP threshold that maximized sensitivity and specificity was 9.6 mm Hg; patients’ AT-PP above this threshold had relative risk of 1.20 for hospital admission.
The specificity and sensitivity of the physicians in appropriately managing asthma in this study were 0.89 and 0.83, respectively (Table 2). There were eight cases in which physician management appeared to be correct on audit, but the AT-PP values failed to indicate a correct patient disposition. The specificity and sensitivity of AT-PP in appropriately managing asthma in this study were 0.78 and 0.78, respectively.
The overall accuracy of AT-PP and physician disposition were 0.78 and 0.87, respectively. Interestingly, there were only two overlapping cases in which inappropriate dispositions by both physicians and AT-PP occurred, suggesting that each may have their relative strengths and that a combinatory approach would prove more effective than either alone. This is also supported by the к-statistic, which showed incomplete overlap between AT-PP and physician disposition (Table 2). A total of five patients who were admitted to the hospital may have been admitted unnecessarily judging from an audit of the inpatient medical records. These records indicate treatment for asthma but at an intensity level that could have been accomplished on an outpatient basis. In each case, the length of the hospital admission was for 1 day. The mean AT-PP measurement posttreatment for these patients was 6.0 mm Hg (95% CI, 2.6 to 9.5 mm Hg) compared to 17.6 mm Hg (95% CI, 13.5 to 21.8 mm Hg) for the remaining patients who were appropriately admitted to the hospital (Student t test = 2.95; p = 0.007). A total of three patients relapsed; two of these patients had posttreatment AT-PP values of 21.3 and 20.7 mm Hg. The mean AT-PP measurement for all patients who were appropriately discharged from the hospital was 9.1 mm Hg (95% CI, 7.3 to 10.5), which was significantly different from that for patients who were appropriately admitted to the hospital (Students t test = 4.51; p < 0.001). Assuming that the AT-PP threshold of 11.3 mm Hg was adhered to in a prospective manner, the PP measurement may have prevented five unnecessary hospital admissions and two inappropriate hospital discharges.
Table 2—Comparison of AT-PP and Treating Physician-Assessed Disposition to Patient Chart Audit Data and Each Other
|Disposition||Patient Chart Audit||Physician Disposition|
|Admitted Patients!||Discharged Patients||Admitted Patients!||Discharged Patients|
|Cohen к-statistic||0.37 (0.14-0.61)|