Developing a risk stratification model for predicting future health care use in asthmatic children

Area of Science:

• Pediatric Pulmonology
• Health Services Research
• Clinical Epidemiology

Background:

• Asthma management often involves risk stratification for exacerbations and healthcare utilization.
• Existing models typically integrate multiple clinical parameters.
• A need exists for simpler predictive tools in pediatric asthma care.

Purpose of the Study:

• To evaluate if historical acute care visits (ACVs) alone can predict future healthcare use in pediatric asthma patients.
• To determine the predictive value of ACV frequency for future asthma-related healthcare needs.
• To assess the utility of a single metric for risk stratification.

Main Methods:

• A retrospective cohort study identified children with asthma outpatient visits over three years.
• The number of ACVs in the 12 months preceding the outpatient visit was quantified.
• Logistic regression models, adjusted for demographic factors, assessed the odds of future ACVs.

Main Results:

• 75.2% of 28,047 outpatient visits had no prior ACVs.
• The probability of future ACVs increased substantially with historical ACVs, from 30% to 87% for 1 to 5+ visits.
• Patients with 5+ historical ACVs, representing 1.1% of the sample, accounted for a disproportionately high number of future ACVs.
• Each additional historical ACV significantly increased the odds of future ACVs (adjusted OR up to 58.71 for 5+ visits).

Conclusions:

• The historical count of ACVs is a strong predictor of future healthcare utilization in pediatric asthma.
• Risk stratification based solely on the number of past ACVs is effective.
• A small subset of pediatric asthma patients with frequent historical ACVs drives a significant portion of future healthcare use.