RESUMEN
BACKGROUND: Asthma affects a large proportion of the population and leads to many hospital encounters involving both hospitalizations and emergency department visits every year. To lower the number of such encounters, many health care systems and health plans deploy predictive models to prospectively identify patients at high risk and offer them care management services for preventive care. However, the previous models do not have sufficient accuracy for serving this purpose well. Embracing the modeling strategy of examining many candidate features, we built a new machine learning model to forecast future asthma hospital encounters of patients with asthma at Intermountain Healthcare, a nonacademic health care system. This model is more accurate than the previously published models. However, it is unclear how well our modeling strategy generalizes to academic health care systems, whose patient composition differs from that of Intermountain Healthcare. OBJECTIVE: This study aims to evaluate the generalizability of our modeling strategy to the University of Washington Medicine (UWM), an academic health care system. METHODS: All adult patients with asthma who visited UWM facilities between 2011 and 2018 served as the patient cohort. We considered 234 candidate features. Through a secondary analysis of 82,888 UWM data instances from 2011 to 2018, we built a machine learning model to forecast asthma hospital encounters of patients with asthma in the subsequent 12 months. RESULTS: Our UWM model yielded an area under the receiver operating characteristic curve (AUC) of 0.902. When placing the cutoff point for making binary classification at the top 10% (1464/14,644) of patients with asthma with the largest forecasted risk, our UWM model yielded an accuracy of 90.6% (13,268/14,644), a sensitivity of 70.2% (153/218), and a specificity of 90.91% (13,115/14,426). CONCLUSIONS: Our modeling strategy showed excellent generalizability to the UWM, leading to a model with an AUC that is higher than all of the AUCs previously reported in the literature for forecasting asthma hospital encounters. After further optimization, our model could be used to facilitate the efficient and effective allocation of asthma care management resources to improve outcomes. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): RR2-10.2196/resprot.5039.
Asunto(s)
Asma , Adulto , Asma/epidemiología , Asma/terapia , Atención a la Salud , Predicción , Hospitales , Humanos , Estudios RetrospectivosAsunto(s)
Infecciones por Bacterias Gramnegativas/diagnóstico , Infecciones por Bacterias Gramnegativas/terapia , Infecciones por Bacterias Grampositivas/diagnóstico , Infecciones por Bacterias Grampositivas/terapia , Neumonía Bacteriana/diagnóstico , Neumonía Bacteriana/terapia , Cuidados Posteriores/métodos , Atención Ambulatoria/métodos , Antibacterianos/uso terapéutico , Niño , Terapia Combinada , Infecciones Comunitarias Adquiridas/complicaciones , Infecciones Comunitarias Adquiridas/diagnóstico , Infecciones Comunitarias Adquiridas/terapia , Drenaje , Infecciones por Bacterias Gramnegativas/complicaciones , Infecciones por Bacterias Grampositivas/complicaciones , Hospitalización , Humanos , Pediatría , Neumonía Bacteriana/complicaciones , RadiografíaRESUMEN
Asthma puts a tremendous overhead on healthcare. To enable effective preventive care to improve outcomes in managing asthma, we recently created two machine learning models, one using University of Washington Medicine data and the other using Intermountain Healthcare data, to predict asthma hospital visits in the next 12 months in asthma patients. As is common in machine learning, neither model supplies explanations for its predictions. To tackle this interpretability issue of black-box models, we developed an automated method to produce rule-style explanations for any machine learning model's predictions made on imbalanced tabular data and to recommend customized interventions without lowering the prediction accuracy. Our method exhibited good performance in explaining our Intermountain Healthcare model's predictions. Yet, it stays unknown how well our method generalizes to academic healthcare systems, whose patient composition differs from that of Intermountain Healthcare. This study evaluates our automated explaining method's generalizability to the academic healthcare system University of Washington Medicine on predicting asthma hospital visits. We did a secondary analysis on 82,888 University of Washington Medicine data instances of asthmatic adults between 2011 and 2018, using our method to explain our University of Washington Medicine model's predictions and to recommend customized interventions. Our results showed that for predicting asthma hospital visits, our automated explaining method had satisfactory generalizability to University of Washington Medicine. In particular, our method explained the predictions for 87.6% of the asthma patients whom our University of Washington Medicine model accurately predicted to experience asthma hospital visits in the next 12 months.
RESUMEN
OBJECTIVES: Manual clinical scoring systems are the current standard used for acute asthma clinical care pathways. No automated system exists that assesses disease severity, time course, and treatment impact in pediatric acute severe asthma exacerbations. WORKING HYPOTHESIS: machine learning applied to continuous vital sign data could provide a novel pediatric-automated asthma respiratory score (pARS) by using the manual pediatric asthma score (PAS) as the clinical care standard. METHODS: Continuous vital sign monitoring data (heart rate, respiratory rate, and pulse oximetry) were merged with the health record data including a provider-determined PAS in children between 2 and 18 years of age admitted to the pediatric intensive care unit (PICU) for status asthmaticus. A cascaded artificial neural network (ANN) was applied to create an automated respiratory score and validated by two approaches. The ANN was compared with the Normal and Poisson regression models. RESULTS: Out of an initial group of 186 patients, 128 patients met inclusion criteria. Merging physiologic data with clinical data yielded >37 000 data points for model training. The pARS score had good predictive accuracy, with 80% of the pARS values within ±2 points of the provider-determined PAS, especially over the mid-range of PASs (6-9). The Poisson and Normal distribution regressions yielded a smaller overall median absolute error. CONCLUSIONS: The pARS reproduced the manually recorded PAS. Once validated and studied prospectively as a tool for research and for physician decision support, this methodology can be implemented in the PICU to objectively guide treatment decisions.