Development of a Pediatric Asthma Predictive Index for Hospitalization
Monday, March 5, 2018
South Hall A2 (Convention Center)
Tiffany Jean, MD, Su-Jau Yang, PhD, William W. Crawford, MD FAAAAI, Scott H. Takahashi, PharmD, Javed Sheikh, MD FAAAAI

RATIONALE: Most asthma severity scores used in the emergency department (ED) focus only on clinical findings. Few have incorporated other risk factors such as epidemiology, co-existing atopy, and medication compliance, into the decision tree for asthma hospitalization.

METHODS: This was a retrospective chart review study which collected data from 12,066 pediatric patients, age 5 to 18 years, with an ED-coded asthma diagnosis from March 2014 to March 2015. Risk factors assessed included: O2 saturation, respiratory rate, demographics, asthma medication dispensing counts (inhaled corticosteroids, short-acting beta2-agonists, oral corticosteroids), atopy, family history of atopy, Medicaid insurance, and prior asthma emergency visits or hospitalizations. Patients hospitalized were compared to those discharged. Univariate and multivariate analysis were used to determine the best predictor variables for hospitalization. Our study then internally validated the hospitalization prediction index.

RESULTS: The factors associated with the greatest risk of asthma hospitalization from the ED are: O2 saturation <94%, respiratory rate >31 breaths/min, history of pneumonia, and prior asthma ED visit in past 12 months. With a reduced predictive model combining the 4 highest risk factors, the odds ratio for hospitalization was 44.9 (p<0.05). After measuring the discrimination power, our c-index was 0.773 with a validation c-index of 0.776, which demonstrates a solid prediction model.

CONCLUSIONS: We have developed and internally validated a pediatric hospitalization prediction index for acute asthma exacerbation with the aim of improving clinical outcomes. Further studies need to externally validate this hospitalization predictive model before its implementation into clinical practice.