Use of machine learning to identify characteristics associated with severe hypoglycemia in older adults with type 1 diabetes: a post-hoc analysis of a case-control study.
Nikki L B Freeman, Rashmi Muthukkumar, Ruth S Weinstock, M Victor Wickerhauser, Anna R Kahkoska
{"title":"Use of machine learning to identify characteristics associated with severe hypoglycemia in older adults with type 1 diabetes: a post-hoc analysis of a case-control study.","authors":"Nikki L B Freeman, Rashmi Muthukkumar, Ruth S Weinstock, M Victor Wickerhauser, Anna R Kahkoska","doi":"10.1136/bmjdrc-2023-003748","DOIUrl":null,"url":null,"abstract":"<p><strong>Introduction: </strong>Severe hypoglycemia (SH) in older adults (OAs) with type 1 diabetes is associated with profound morbidity and mortality, yet its etiology can be complex and multifactorial. Enhanced tools to identify OAs who are at high risk for SH are needed. This study used machine learning to identify characteristics that distinguish those with and without recent SH, selecting from a range of demographic and clinical, behavioral and lifestyle, and neurocognitive characteristics, along with continuous glucose monitoring (CGM) measures.</p><p><strong>Research design and methods: </strong>Data from a case-control study involving OAs recruited from the T1D Exchange Clinical Network were analyzed. The random forest machine learning algorithm was used to elucidate the characteristics associated with case versus control status and their relative importance. Models with successively rich characteristic sets were examined to systematically incorporate each domain of possible risk characteristics.</p><p><strong>Results: </strong>Data from 191 OAs with type 1 diabetes (47.1% female, 92.1% non-Hispanic white) were analyzed. Across models, hypoglycemia unawareness was the top characteristic associated with SH history. For the model with the richest input data, the most important characteristics, in descending order, were hypoglycemia unawareness, hypoglycemia fear, coefficient of variation from CGM, % time blood glucose below 70 mg/dL, and trail making test B score.</p><p><strong>Conclusions: </strong>Machine learning may augment risk stratification for OAs by identifying key characteristics associated with SH. Prospective studies are needed to identify the predictive performance of these risk characteristics.</p>","PeriodicalId":9151,"journal":{"name":"BMJ Open Diabetes Research & Care","volume":"12 1","pages":""},"PeriodicalIF":3.7000,"publicationDate":"2024-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10900355/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"BMJ Open Diabetes Research & Care","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1136/bmjdrc-2023-003748","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENDOCRINOLOGY & METABOLISM","Score":null,"Total":0}
引用次数: 0
Abstract
Introduction: Severe hypoglycemia (SH) in older adults (OAs) with type 1 diabetes is associated with profound morbidity and mortality, yet its etiology can be complex and multifactorial. Enhanced tools to identify OAs who are at high risk for SH are needed. This study used machine learning to identify characteristics that distinguish those with and without recent SH, selecting from a range of demographic and clinical, behavioral and lifestyle, and neurocognitive characteristics, along with continuous glucose monitoring (CGM) measures.
Research design and methods: Data from a case-control study involving OAs recruited from the T1D Exchange Clinical Network were analyzed. The random forest machine learning algorithm was used to elucidate the characteristics associated with case versus control status and their relative importance. Models with successively rich characteristic sets were examined to systematically incorporate each domain of possible risk characteristics.
Results: Data from 191 OAs with type 1 diabetes (47.1% female, 92.1% non-Hispanic white) were analyzed. Across models, hypoglycemia unawareness was the top characteristic associated with SH history. For the model with the richest input data, the most important characteristics, in descending order, were hypoglycemia unawareness, hypoglycemia fear, coefficient of variation from CGM, % time blood glucose below 70 mg/dL, and trail making test B score.
Conclusions: Machine learning may augment risk stratification for OAs by identifying key characteristics associated with SH. Prospective studies are needed to identify the predictive performance of these risk characteristics.
期刊介绍:
BMJ Open Diabetes Research & Care is an open access journal committed to publishing high-quality, basic and clinical research articles regarding type 1 and type 2 diabetes, and associated complications. Only original content will be accepted, and submissions are subject to rigorous peer review to ensure the publication of
high-quality — and evidence-based — original research articles.