Childhood asthma is a heterogeneous disease. This is evident given the broad range of clinical characteristics and treatment responses for children defined as asthmatic. Current classification systems for asthma phenotypes are inadequate to fully understand the pathogenesis, pathophysiology, and treatment outcomes of asthma. Given the heterogeneous nature of asthma, novel ensemble clustering of childhood asthma using data from five clinical trials is proposed. It is hypothesized these discovered phenotypes (or clusters) will have defining characteristics that suggest unique etiologies and appropriate treatment strategies.
The aims are to (1) perform ensemble clustering of childhood asthma and describe characteristics for the observed phenotype, (2) develop a model to classify childhood asthma into the phenotypes previously defined in Aim (1), and (3) retrospectively determine whether childhood asthma phenotypes identified by ensemble clustering correlate with treatment responses in clinical trials. We expect (1) fewer than ten phenotypes to be identified with some attributes over- represented in each phenotype, (2) to develop a simple, easily interpretable classification model, and (3) to identify phenotypes for which clinical trial treatments were effective or ineffective. On the path toward personalized medicine, novel research with ensemble clustering and a large dataset with clinical trial outcomes will advance the field by validating and hypothesizing childhood asthma pathogenesis, pathophysiology and treatment responses.

Public Health Relevance

Childhood asthma is likely not a single disease, but rather a heterogeneous disease with many phenotypes (or subtypes). Computational analysis can identify and elucidate the characteristics of these phenotypes. These phenotypes may provide insight into the causes of and appropriate treatments for asthma.

National Institute of Health (NIH)
National Heart, Lung, and Blood Institute (NHLBI)
Individual Predoctoral NRSA for M.D./Ph.D. Fellowships (ADAMHA) (F30)
Project #
Application #
Study Section
Special Emphasis Panel (ZRG1-F16-B (20))
Program Officer
Tigno, Xenia
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of Wisconsin Madison
Biostatistics & Other Math Sci
Schools of Medicine
United States
Zip Code
Chang, Timothy S; Gangnon, Ronald E; David Page, C et al. (2015) Sparse modeling of spatial environmental variables associated with asthma. J Biomed Inform 53:320-9
Tomasallo, Carrie D; Hanrahan, Lawrence P; Tandias, Aman et al. (2014) Estimating Wisconsin asthma prevalence using clinical electronic health records and public health data. Am J Public Health 104:e65-73
Chang, Timothy S; Lemanske Jr, Robert F; Mauger, David T et al. (2014) Childhood asthma clusters and response to therapy in clinical trials. J Allergy Clin Immunol 133:363-9
Chang, Timothy S; Jensen, Matthew B (2014) Haemodilution for acute ischaemic stroke. Cochrane Database Syst Rev 8:CD000103
Chang, Timothy S; Lemanske Jr, Robert F; Guilbert, Theresa W et al. (2013) Evaluation of the modified asthma predictive index in high-risk preschool children. J Allergy Clin Immunol Pract 1:152-6