Understanding the importance of key risk factors in predicting chronic bronchitic symptoms using a machine learning approach. Deng, H., Urman, R., Gilliland, F. D., & Eckel, S. P. BMC medical research methodology, 19(1):70, March, 2019. ![Understanding the importance of key risk factors in predicting chronic bronchitic symptoms using a machine learning approach. [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex BACKGROUND: Chronic respiratory symptoms involving bronchitis, cough and phlegm in children are underappreciated but pose a significant public health burden. Efforts for prevention and management could be supported by an understanding of the relative importance of determinants, including environmental exposures. Thus, we aim to develop a prediction model for bronchitic symptoms. METHODS: Schoolchildren from the population-based southern California Children's Health Study were visited annually from 2003 to 2012. Bronchitic symptoms over the prior 12 months were assessed by questionnaire. A gradient boosting model was fit using groups of risk factors (including traffic/air pollution exposures) for all children and by asthma status. Training data consisted of one observation per participant in a random study year (for 50% of participants). Validation data consisted of: (1) a random (later) year in the same participants (within-participant); (2) a random year in participants excluded from the training data (across-participant). RESULTS: At baseline, 13.2% of children had asthma and 18.1% reported bronchitic symptoms. Models performed similarly within- and across-participant. Previous year symptoms/medication use provided much of the predictive ability (across-participant area under the receiver operating characteristic curve (AUC): 0.76 vs 0.78 for all risk factors, in all participants). Traffic/air pollution exposures added modestly to prediction as did body mass index percentile, age and parent stress. CONCLUSIONS: Regardless of asthma status, previous symptoms were the most important predictors of current symptoms. Traffic/air pollution variables contribute modest predictive information, but impact large populations. Methods proposed here could be generalized to personalized exacerbation predictions in future longitudinal studies to support targeted prevention efforts.
@article{deng_understanding_2019,
title = {Understanding the importance of key risk factors in predicting chronic bronchitic symptoms using a machine learning approach.},
volume = {19},
issn = {1471-2288 1471-2288},
url = {https://bmcmedresmethodol.biomedcentral.com/articles/10.1186/s12874-019-0708-x},
doi = {10.1186/s12874-019-0708-x},
abstract = {BACKGROUND: Chronic respiratory symptoms involving bronchitis, cough and phlegm in children are underappreciated but pose a significant public health burden. Efforts for prevention and management could be supported by an understanding of the relative importance of determinants, including environmental exposures. Thus, we aim to develop a prediction model for bronchitic symptoms. METHODS: Schoolchildren from the population-based southern California Children's Health Study were visited annually from 2003 to 2012. Bronchitic symptoms over the prior 12 months were assessed by questionnaire. A gradient boosting model was fit using groups of risk factors (including traffic/air pollution exposures) for all children and by asthma status. Training data consisted of one observation per participant in a random study year (for 50\% of participants). Validation data consisted of: (1) a random (later) year in the same participants (within-participant); (2) a random year in participants excluded from the training data (across-participant). RESULTS: At baseline, 13.2\% of children had asthma and 18.1\% reported bronchitic symptoms. Models performed similarly within- and across-participant. Previous year symptoms/medication use provided much of the predictive ability (across-participant area under the receiver operating characteristic curve (AUC): 0.76 vs 0.78 for all risk factors, in all participants). Traffic/air pollution exposures added modestly to prediction as did body mass index percentile, age and parent stress. CONCLUSIONS: Regardless of asthma status, previous symptoms were the most important predictors of current symptoms. Traffic/air pollution variables contribute modest predictive information, but impact large populations. Methods proposed here could be generalized to personalized exacerbation predictions in future longitudinal studies to support targeted prevention efforts.},
language = {eng},
number = {1},
journal = {BMC medical research methodology},
author = {Deng, Huiyu and Urman, Robert and Gilliland, Frank D. and Eckel, Sandrah P.},
month = mar,
year = {2019},
pmid = {30925901},
pmcid = {PMC6441159},
keywords = {Air pollution, Bronchitic symptoms, Gradient boosting model, Machine learning, Prediction model},
pages = {70},
}
Downloads: 0
{"_id":"FouqSiWG9Kqntvr7r","bibbaseid":"deng-urman-gilliland-eckel-understandingtheimportanceofkeyriskfactorsinpredictingchronicbronchiticsymptomsusingamachinelearningapproach-2019","author_short":["Deng, H.","Urman, R.","Gilliland, F. D.","Eckel, S. P."],"bibdata":{"bibtype":"article","type":"article","title":"Understanding the importance of key risk factors in predicting chronic bronchitic symptoms using a machine learning approach.","volume":"19","issn":"1471-2288 1471-2288","url":"https://bmcmedresmethodol.biomedcentral.com/articles/10.1186/s12874-019-0708-x","doi":"10.1186/s12874-019-0708-x","abstract":"BACKGROUND: Chronic respiratory symptoms involving bronchitis, cough and phlegm in children are underappreciated but pose a significant public health burden. Efforts for prevention and management could be supported by an understanding of the relative importance of determinants, including environmental exposures. Thus, we aim to develop a prediction model for bronchitic symptoms. METHODS: Schoolchildren from the population-based southern California Children's Health Study were visited annually from 2003 to 2012. Bronchitic symptoms over the prior 12 months were assessed by questionnaire. A gradient boosting model was fit using groups of risk factors (including traffic/air pollution exposures) for all children and by asthma status. Training data consisted of one observation per participant in a random study year (for 50% of participants). Validation data consisted of: (1) a random (later) year in the same participants (within-participant); (2) a random year in participants excluded from the training data (across-participant). RESULTS: At baseline, 13.2% of children had asthma and 18.1% reported bronchitic symptoms. Models performed similarly within- and across-participant. Previous year symptoms/medication use provided much of the predictive ability (across-participant area under the receiver operating characteristic curve (AUC): 0.76 vs 0.78 for all risk factors, in all participants). Traffic/air pollution exposures added modestly to prediction as did body mass index percentile, age and parent stress. CONCLUSIONS: Regardless of asthma status, previous symptoms were the most important predictors of current symptoms. Traffic/air pollution variables contribute modest predictive information, but impact large populations. Methods proposed here could be generalized to personalized exacerbation predictions in future longitudinal studies to support targeted prevention efforts.","language":"eng","number":"1","journal":"BMC medical research methodology","author":[{"propositions":[],"lastnames":["Deng"],"firstnames":["Huiyu"],"suffixes":[]},{"propositions":[],"lastnames":["Urman"],"firstnames":["Robert"],"suffixes":[]},{"propositions":[],"lastnames":["Gilliland"],"firstnames":["Frank","D."],"suffixes":[]},{"propositions":[],"lastnames":["Eckel"],"firstnames":["Sandrah","P."],"suffixes":[]}],"month":"March","year":"2019","pmid":"30925901","pmcid":"PMC6441159","keywords":"Air pollution, Bronchitic symptoms, Gradient boosting model, Machine learning, Prediction model","pages":"70","bibtex":"@article{deng_understanding_2019,\n\ttitle = {Understanding the importance of key risk factors in predicting chronic bronchitic symptoms using a machine learning approach.},\n\tvolume = {19},\n\tissn = {1471-2288 1471-2288},\n\turl = {https://bmcmedresmethodol.biomedcentral.com/articles/10.1186/s12874-019-0708-x},\n\tdoi = {10.1186/s12874-019-0708-x},\n\tabstract = {BACKGROUND: Chronic respiratory symptoms involving bronchitis, cough and phlegm in children are underappreciated but pose a significant public health burden. Efforts for prevention and management could be supported by an understanding of the relative importance of determinants, including environmental exposures. Thus, we aim to develop a prediction model for bronchitic symptoms. METHODS: Schoolchildren from the population-based southern California Children's Health Study were visited annually from 2003 to 2012. Bronchitic symptoms over the prior 12 months were assessed by questionnaire. A gradient boosting model was fit using groups of risk factors (including traffic/air pollution exposures) for all children and by asthma status. Training data consisted of one observation per participant in a random study year (for 50\\% of participants). Validation data consisted of: (1) a random (later) year in the same participants (within-participant); (2) a random year in participants excluded from the training data (across-participant). RESULTS: At baseline, 13.2\\% of children had asthma and 18.1\\% reported bronchitic symptoms. Models performed similarly within- and across-participant. Previous year symptoms/medication use provided much of the predictive ability (across-participant area under the receiver operating characteristic curve (AUC): 0.76 vs 0.78 for all risk factors, in all participants). Traffic/air pollution exposures added modestly to prediction as did body mass index percentile, age and parent stress. CONCLUSIONS: Regardless of asthma status, previous symptoms were the most important predictors of current symptoms. Traffic/air pollution variables contribute modest predictive information, but impact large populations. Methods proposed here could be generalized to personalized exacerbation predictions in future longitudinal studies to support targeted prevention efforts.},\n\tlanguage = {eng},\n\tnumber = {1},\n\tjournal = {BMC medical research methodology},\n\tauthor = {Deng, Huiyu and Urman, Robert and Gilliland, Frank D. and Eckel, Sandrah P.},\n\tmonth = mar,\n\tyear = {2019},\n\tpmid = {30925901},\n\tpmcid = {PMC6441159},\n\tkeywords = {Air pollution, Bronchitic symptoms, Gradient boosting model, Machine learning, Prediction model},\n\tpages = {70},\n}\n\n","author_short":["Deng, H.","Urman, R.","Gilliland, F. D.","Eckel, S. P."],"key":"deng_understanding_2019","id":"deng_understanding_2019","bibbaseid":"deng-urman-gilliland-eckel-understandingtheimportanceofkeyriskfactorsinpredictingchronicbronchiticsymptomsusingamachinelearningapproach-2019","role":"author","urls":{"Paper":"https://bmcmedresmethodol.biomedcentral.com/articles/10.1186/s12874-019-0708-x"},"keyword":["Air pollution","Bronchitic symptoms","Gradient boosting model","Machine learning","Prediction model"],"metadata":{"authorlinks":{}}},"bibtype":"article","biburl":"https://api.zotero.org/users/3649949/collections/52M3HD2M/items?key=kvw05jEWpV9zO4gNkD1KQFRV&format=bibtex&limit=100","dataSources":["qkN2F4hKQojRGQeTy","cR2bQCnuvgoCQwTEh","Zv8utRXNjhXZcJdZX","zguJ5LkMpKLhRDgWX","x94sDkjv6sHRisXm3","Xngq2WDk97aWqouA5","SbR97nY87cwp9QrPt"],"keywords":["air pollution","bronchitic symptoms","gradient boosting model","machine learning","prediction model"],"search_terms":["understanding","importance","key","risk","factors","predicting","chronic","bronchitic","symptoms","using","machine","learning","approach","deng","urman","gilliland","eckel"],"title":"Understanding the importance of key risk factors in predicting chronic bronchitic symptoms using a machine learning approach.","year":2019}