The Utility of Pediatric Asthma Severity Scores in Disposition Decisions for Status Asthmaticus
Poster May 09, 2018
Ryan KS, McKinley S, Bartlett J, Son S, Wilsey M, Nakagawa TA, Sochet AA
The Pediatric Asthma Severity Score (PASS) is a commonly used clinical and research metric for assessing pediatric asthma exacerbation severity. PASS is not validated for practice beyond distinguishing appropriate initial emergency room (ER) disposition. We aimed to determine if PASS identifies patients with hospital readmission, PICU bounce-back, or late rescue. We performed a retrospective, cohort study from May 2015 to May 2017 in children ≥5 years of age admitted for status asthmaticus. We excluded patients with pulmonary hypertension, chronic lung disease, cystic fibrosis, or tracheostomy. In our facility PASS data were noted prior to all respiratory treatments by trained therapists. Primary outcomes were hospital readmission (<30d of discharge), PICU bounce-back (<24hr readmission after PICU to floor transfer), and late rescue (<24hr PICU transfer after ER to floor admission). PASS data, general outcomes, and cohort characteristics were assessed with Fisher's exact, student's t, and Wilcoxon rank-sum tests. Receiver operator characteristic (ROC) curves yielded area under the curve (AUC) values and PASS cutoffs to discriminate unplanned hospital readmission, PICU bounce-back, and late rescue. Type I error was set at 0.05. Data were analyzed using Stata v13.1.
Of the 227 children who met study criteria, 88 (39%) were PICU admissions and 139 (61%) floor. We noted 2.3% had hospital readmission, 6.5% late rescue, and 5.7% PICU bounce-back. General sample characteristics, asthma classification, comorbidities, applied therapeutics, and routine outcomes did not differ between research cohorts. Discharge PASS did not differ for those w/wo hospital readmission (1 [IQR:1-2] vs 2 [IQR:1-3]). PASS prior to PICU to floor transfer were no different for those w/wo PICU bounce-back (6 [IQR:5-7] vs 4 [IQR:2-5]). PASS just prior to ER to floor admission were greater in those with late rescue (8 [IQR:7-8] vs 6 [IQR:4-8], p=0.03). ROC curve for discharge PASS on hospital readmission yielded an AUC of 0.24. ROC curve for PASS prior to PICU to floor transfer on PICU bounce-back yielded an AUC of 0.74 (95% CI: 0.45-0.99; cutoff PASS 8: 20% SE, 100% SP). ROC curve for PASS prior to ER to floor admission on late rescue yielded an AUC of 0.71 (95% CI: 0.57-0.85; cutoff PASS 8: 78% SE, 62% SP).
PASS may provide insight for acute disposition planning (i.e., selecting an appropriate level of care) rather than determining hospital discharge readiness. Prior to adopting PASS criteria in these scenarios, further study is warranted.