K. Edwards, E. Defranco, Taylor Griffith, H. Masters
{"title":"Initiation of Oral Agents Versus Insulin as First-Line Therapy for Gestational Diabetes Mellitus [ID: 1360154]","authors":"K. Edwards, E. Defranco, Taylor Griffith, H. Masters","doi":"10.1097/01.AOG.0000931012.09612.75","DOIUrl":null,"url":null,"abstract":"INTRODUCTION: Although the American College of Obstetricians and Gynecologists recommends insulin as first-line pharmacologic treatment for gestational diabetes mellitus (GDM) since 2017, use of oral agents as first-line remains common. This study aims to identify patient-level factors associated with choice of initial pharmacologic treatment of GDM with oral agents compared to insulin. METHODS: Case–control study of 617 patients with White's class A2 GDM treated at a single academic center in a comprehensive GDM treatment program, 2011–2018. Pregestational diabetes and GDM not requiring medication (A1 GDM) were excluded. Case group was GDM pregnancies treated with any oral agent first-line and control group was treated with insulin as first-line treatment. Logistic regression was performed to identify patient factors associated with oral agents as first-line pharmacologic treatment. Receiver-operator characteristic (ROC) curve estimated the ability of these factors to predict initial medication choice. RESULTS: Among those with GDMA2, 80% were started on an oral agent and 20% were started on insulin. Those started initially on oral agents were more likely to be Hispanic or non-Hispanic White, be non-English speakers, have private insurance, and have normal body mass index (BMI). Factors strongly associated with starting on an oral agent rather than insulin were Hispanic ethnicity, treatment initiated later in pregnancy (>28 weeks), normal BMI, and 1-hour glucose challenge test (GCT) greater than 200. The ROC derived from the adjusted logistic regression model found a predictive value of these variables for initiating oral agents of 79% (area under the curve 0.79). CONCLUSION: Patient-level factors are predictive of pharmacological agent chosen for first-line treatment of GDM. Factors that are typically associated with more difficult to control GDM, such as obesity, high values on the GCT screening test and earlier gestational age at diagnosis are more common in patients treated initially with insulin. However, some demographic factors such as Hispanic ethnicity and non-English primary language are more common among patients initiated on oral agents for treatment of GDM.","PeriodicalId":19405,"journal":{"name":"Obstetrics & Gynecology","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Obstetrics & Gynecology","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1097/01.AOG.0000931012.09612.75","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
INTRODUCTION: Although the American College of Obstetricians and Gynecologists recommends insulin as first-line pharmacologic treatment for gestational diabetes mellitus (GDM) since 2017, use of oral agents as first-line remains common. This study aims to identify patient-level factors associated with choice of initial pharmacologic treatment of GDM with oral agents compared to insulin. METHODS: Case–control study of 617 patients with White's class A2 GDM treated at a single academic center in a comprehensive GDM treatment program, 2011–2018. Pregestational diabetes and GDM not requiring medication (A1 GDM) were excluded. Case group was GDM pregnancies treated with any oral agent first-line and control group was treated with insulin as first-line treatment. Logistic regression was performed to identify patient factors associated with oral agents as first-line pharmacologic treatment. Receiver-operator characteristic (ROC) curve estimated the ability of these factors to predict initial medication choice. RESULTS: Among those with GDMA2, 80% were started on an oral agent and 20% were started on insulin. Those started initially on oral agents were more likely to be Hispanic or non-Hispanic White, be non-English speakers, have private insurance, and have normal body mass index (BMI). Factors strongly associated with starting on an oral agent rather than insulin were Hispanic ethnicity, treatment initiated later in pregnancy (>28 weeks), normal BMI, and 1-hour glucose challenge test (GCT) greater than 200. The ROC derived from the adjusted logistic regression model found a predictive value of these variables for initiating oral agents of 79% (area under the curve 0.79). CONCLUSION: Patient-level factors are predictive of pharmacological agent chosen for first-line treatment of GDM. Factors that are typically associated with more difficult to control GDM, such as obesity, high values on the GCT screening test and earlier gestational age at diagnosis are more common in patients treated initially with insulin. However, some demographic factors such as Hispanic ethnicity and non-English primary language are more common among patients initiated on oral agents for treatment of GDM.