Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


43043.pdf (3.33 MB)

ETD Abstract Container

Abstract Header

Predicting Lung Function Decline and Pulmonary Exacerbation in Cystic Fibrosis Patients Using Bayesian Regularization and Geomarkers

Peterson, Clayton
http://rave.ohiolink.edu/etdc/view?acc_num=ucin165951875213978

Abstract Details

2022, MS, University of Cincinnati, Medicine: Biostatistics (Environmental Health).
BACKGROUND: Environmental exposures and community characteristics have been linked to rapid lung function decline and other adverse pulmonary outcomes in people with cystic fibrosis (CF). Geomarkers, the measurements of these exposures, have been linked to patient outcomes in other respiratory diseases, though broad-based geomarker studies are lacking and it is unknown which geomarkers will have the greatest predictive potential for rapid decline and pulmonary exacerbation (PEx) in CF. OBJECTIVE: A retrospective longitudinal cohort study was performed to determine whether and which geomarkers would be chosen via novel Bayesian joint covariate selection approaches and to compare the predictive performance of the resultant models for onset of PEx. METHODS: Non-stationary Gaussian linear mixed effects models were fitted to data from 151 cystic fibrosis patients aged 6 – 20 receiving care at the Cincinnati Children’s Hospital Cystic Fibrosis Center (2007-2017). The outcome of interest was forced expiratory volume in 1 second of percent predicted (FEV1pp). Target functions were used to predict PEx onset according to an established definition based on drops in FEV1pp. Covariates included 11 clinical/demographic characteristics (age, sex, number of PEx-defined events within previous year, F508del mutation, pancreatic insufficiency, MEDICAID insurance use, BMI percentile, PA infection, MRSA infection, CF-related diabetes mellitus, and the number of hospital visits within the previous year), and 45 geomarkers comprising 8 categories (socioeconomic status, access to care, roadway proximity, crime, land cover, impervious descriptors, weather, and air pollution). Joint selection of covariates for predictive models was achieved using four Bayesian penalized regression models (elastic-net, adaptive lasso, ridge, and lasso). Unique covariate selections at both the 95% and 90% credible intervals (CIs) were fit to a linear mixed effects model with non-stationary stochastic processes and the predictive performance of these for PEx was assessed using 5-fold cross-validation repeated 100 times coupled with receiver operating characteristics (ROC) analysis. MAIN RESULTS: Six unique covariate selections were achieved. Resultant models included one geomarker to six geomarkers (air temperature, percentage of tertiary roads outside urban areas, percentage of impervious nonroad outside urban areas, fine atmospheric particulate matter, fraction achieving high school graduation, and motor vehicle theft) which were derived from weather, impervious descriptor, air pollution, socioeconomic status, and crime categories. The adaptive lasso method had the lowest information criteria of the four selection methods. For predictive accuracy of PEx, the covariate selection from the 95% CI elastic-net method had the highest area under the receiver-operating characteristic curve (mean ± standard deviation; 0.780 ± 0.026) along with the 95% ridge & lasso method selection (0.780 ± 0.027). The covariate selection from the 95% CI elastic-net method had the highest sensitivity (0.773 ± 0.083), and the covariate selection from the 95% CI adaptive lasso method had the highest specificity (0.691 ± 0.087). CONCLUSION: This novel application produced real-time prediction of PEx informed by certain localized geomarkers, which have the potential to positively impact care decisions for people with CF according to neighborhood/community-level exposures.
Marepalli Rao, Ph.D. (Committee Member)
Rhonda Szczesniak, Ph.D. (Committee Member)
34 p.
Statistics
Cystic Fibrosis; Pulmonary Exacerbation; Bayesian Regularization; Geomarkers; Rapid Lung Function Decline; Linear Mixed Effects Model

Recommended Citations

Citations

  • Peterson, C. (2022). Predicting Lung Function Decline and Pulmonary Exacerbation in Cystic Fibrosis Patients Using Bayesian Regularization and Geomarkers [Master's thesis, University of Cincinnati]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=ucin165951875213978

    APA Style (7th edition)

  • Peterson, Clayton. Predicting Lung Function Decline and Pulmonary Exacerbation in Cystic Fibrosis Patients Using Bayesian Regularization and Geomarkers. 2022. University of Cincinnati, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=ucin165951875213978.

    MLA Style (8th edition)

  • Peterson, Clayton. "Predicting Lung Function Decline and Pulmonary Exacerbation in Cystic Fibrosis Patients Using Bayesian Regularization and Geomarkers." Master's thesis, University of Cincinnati, 2022. http://rave.ohiolink.edu/etdc/view?acc_num=ucin165951875213978

    Chicago Manual of Style (17th edition)

ucin165951875213978
82
© 2022, all rights reserved.
This open access ETD is published by University of Cincinnati and OhioLINK.