Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

Prediction of Severe Asthma Outcomes in Children on EHR Data

Liu, Jiaqian (2023) Prediction of Severe Asthma Outcomes in Children on EHR Data. Master's Thesis, University of Pittsburgh. (Unpublished)

Download (894kB) | Preview


Background: Asthma is a leading chronic disease among children with nonnegligible numbers of Emergency Department (ED) visits and hospitalization annually. To effectively utilize real-world electronic health record (EHR) data, it is crucial to identify the best modeling approach that accounts for the unique features of EHR data. Additionally, identifying high-risk sub-populations susceptible to severe asthma outcomes can provide valuable insights for targeted
interventions and improved patient medical care.
Methods: Various statistical and machine learning models, including those with random effects such as linear (generalized) mixed effects models, and mixed effects random forests, were employed to develop a prediction model for length of stay (LOS) and asthma exacerbation using EHR data. Once the optimal prediction model was identified, it was further trained on the entire dataset to identify the risk factors that significantly contribute to severe asthma outcomes.
Results: Linear mixed effects model and generalized linear mixed effects model were the top-performing models for predicting inpatient LOS and asthma exacerbation risk, with an average RMSE of 0.53 and AUC of 0.87, respectively. Notably, patient age, action plan, and the patient health history such as inpatient visit (yes or no from the last encounter) and ED visit (yes or no from the last encounter) were the strongest predictors of severe asthma outcomes. Other statistically significant predictors included having chronic diseases, belonging to a minority race group, and during the pandemic period.


Social Networking:
Share |


Item Type: University of Pittsburgh ETD
Status: Unpublished
CreatorsEmailPitt UsernameORCID
Liu, Jiaqianjil289@pitt.edujil289
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Thesis AdvisorDing, Yingyingding@pitt.eduyingding
Committee MemberForno,
Committee MemberTang, Lulutang@pitt.edulutang
Date: 11 May 2023
Date Type: Publication
Defense Date: 20 April 2023
Approval Date: 11 May 2023
Submission Date: 28 April 2023
Access Restriction: No restriction; Release the ETD for access worldwide immediately.
Number of Pages: 63
Institution: University of Pittsburgh
Schools and Programs: School of Public Health > Biostatistics
Degree: MS - Master of Science
Thesis Type: Master's Thesis
Refereed: Yes
Uncontrolled Keywords: Mixed Effects, Random Forests, ED, Asthma Exacerbation, Length of Stay
Date Deposited: 11 May 2023 16:03
Last Modified: 11 May 2023 16:03


Monthly Views for the past 3 years

Plum Analytics

Actions (login required)

View Item View Item