The asthma prediction rule to decrease hospitalizations for children with asthma

Defining Treatment Response for Clinical Trials of Pediatric Acute Asthma

BACKGROUND

An agreed-upon definition of treatment response for clinical trials of pediatric acute asthma does not exist, limiting meaningful comparisons among therapeutic interventions and advances in asthma management.

OBJECTIVE

To develop a consensus definition of treatment response for clinical trials of pediatric acute asthma.

METHODS

A multidisciplinary panel of 22 experts participated in a Web-based modified Delphi process to achieve consensus on a definition of treatment response. Round 1 consisted of closed- and open-ended questions in which panelists ranked measures of treatment response developed by literature review, suggested additional measures, and explained their responses. In rounds 2 and 3, panelists reviewed summary statistics of the panel's rating from prior rounds and reconsidered their rankings. In round 3, pairwise ranking was performed to determine the ranked importance of components. Consensus was defined as 70% or greater agreement among panelists choosing Likert-scale values of 1 to 6 (extremely unimportant to extremely important) and an interquartile range less than 2.

RESULTS

Drawing on results from the expert panel, we developed a definition of treatment response that includes Clinical Severity Score, need for additional therapies, and hospitalization. Clinical Severity Score encompassed most ranked criteria (eg, respiratory distress, wheeze) for a treatment response definition. Panelists recommended that a valid and pragmatic severity score be used consistently across institutions. Panelists also achieved consensus on the top 10 criteria that appropriately classify need for hospitalization.

CONCLUSIONS

This consensus definition of treatment response can be used in clinical trials of children with acute asthma to standardize outcome measurement and report meaningful outcomes.

Risk Factors and Outcomes Associated With Antibiotic Therapy in Children Hospitalized With Asthma Exacerbation

OBJECTIVE

Despite lack of benefit, antibiotics are overused in management of asthma exacerbation in children. In this study, data from a single children's hospital were analyzed to identify factors and outcomes associated with antibiotic use in children hospitalized with asthma.

METHODS

The study population was identified by using administrative data from 2012 to 2015, with subsequent verification of asthma. We analyzed factors associated with antibiotic use (demographic, seasonal, clinical) and outcome (length of stay [LOS]) with respect to: 1) disposition to pediatric floor (PF) or pediatric intensive care unit (PICU); and 2) evidence of coexisting bacterial infection and/or fever. Statistical analysis included univariate and controlled regression models. Data are presented as median and IQR for continuous variables and OR and regression coefficient (β) with 95% CIs for regression analyses.

RESULTS

Of 600 patients, 28.8% were admitted to PICU, 14.8% had verified bacterial infection, and 53.8% received antibiotic, mainly azithromycin. Nearly all PICU patients were treated with antibiotic, irrespective of coexisting bacterial infection or fever. Among PF patients, nearly 30% without bacterial infection or fever and 40% with fever alone received antimicrobials. Overall risk for antibiotic treatment was associated with older age, female sex, desaturation events, oxygen supplementation, and PICU admission. Additionally, antibiotic treatment was associated with 13- to 19-hour increased LOS for PF patients without bacterial infection and/or fever.

CONCLUSIONS

Almost half of pediatric patients admitted with asthma exacerbation received antibiotic therapy with no clear indication, which was associated with prolonged LOS.

Machine learning approaches to personalize early prediction of asthma exacerbations

Patient telemonitoring results in an aggregation of significant amounts of information about patient disease trajectory. However, the potential use of this information for early prediction of exacerbations in adult asthma patients has not been systematically evaluated. The aim of this study was to explore the utility of telemonitoring data for building machine learning algorithms that predict asthma exacerbations before they occur. The study dataset comprised daily self-monitoring reports consisting of 7001 records submitted by adult asthma patients during home telemonitoring. Predictive modeling included preparation of stratified training datasets, predictive feature selection, and evaluation of resulting classifiers. Using a 7-day window, a naive Bayesian classifier, adaptive Bayesian network, and support vector machines were able to predict asthma exacerbation occurring on day 8, with sensitivity of 0.80, 1.00, and 0.84; specificity of 0.77, 1.00, and 0.80; and accuracy of 0.77, 1.00, and 0.80, respectively. Our study demonstrated that machine learning techniques have significant potential in developing personalized decision support for chronic disease telemonitoring systems. Future studies may benefit from a comprehensive predictive framework that combines telemonitoring data with other factors affecting the likelihood of developing acute exacerbation. Approaches implemented for advanced asthma exacerbation prediction may be extended to prediction of exacerbations in patients with other chronic health conditions.