Electronic health record phenotyping improves detection and screening of type 2 diabetes in the general United States population: A cross-sectional, unselected, retrospective study
نویسندگان
چکیده
OBJECTIVES An estimated 25% of type two diabetes mellitus (DM2) patients in the United States are undiagnosed due to inadequate screening, because it is prohibitive to administer laboratory tests to everyone. We assess whether electronic health record (EHR) phenotyping could improve DM2 screening compared to conventional models, even when records are incomplete and not recorded systematically across patients and practice locations, as is typically seen in practice. METHODS In this cross-sectional, retrospective study, EHR data from 9948 US patients were used to develop a pre-screening tool to predict current DM2, using multivariate logistic regression and a random-forests probabilistic model for out-of-sample validation. We compared (1) a full EHR model containing commonly prescribed medications, diagnoses (as ICD9 categories), and conventional predictors, (2) a restricted EHR DX model which excluded medications, and (3) a conventional model containing basic predictors and their interactions (BMI, age, sex, smoking status, hypertension). RESULTS Using a patient's full EHR or restricted EHR was superior to using basic covariates alone for detecting individuals with diabetes (hierarchical X(2) test, p<0.001). Migraines, depot medroxyprogesterone acetate, and cardiac dysrhythmias were associated negatively with DM2, while sexual and gender identity disorder diagnosis, viral and chlamydial infections, and herpes zoster were associated positively. Adding EHR phenotypes improved classification; the AUC for the full EHR Model, EHR DX model, and conventional model using logistic regression, were 84.9%, 83.2%, and 75.0% respectively. For random forest machine learning out-of-sample prediction, accuracy also was improved when using EHR phenotypes; the AUC values were 81.3%, 79.6%, and 74.8%, respectively. Improved AUCs reflect better performance for most thresholds that balance sensitivity and specificity. CONCLUSIONS EHR phenotyping resulted in markedly superior detection of DM2, even in the face of missing and unsystematically recorded data, based on the ROC curves. EHR phenotypes could more efficiently identify which patients do require, and don't require, further laboratory screening. When applied to the current number of undiagnosed individuals in the United States, we predict that incorporating EHR phenotype screening would identify an additional 400,000 patients with active, untreated diabetes compared to the conventional pre-screening models.
منابع مشابه
Economic Evaluation of Screening for Type 2 Diabetes: Case Study of Iran
Abstract Objective: In health economics, one of the important methods of economic evaluation is cost-effectiveness analysis. The usual procedure in this method is to consider effectiveness of Disability-Adjusted Life Years (DALYs) as a measure criterion. Material and Methods: This was a cross sectional study of economic assessment type or Cost-Effectiveness. All over 30 years old population o...
متن کاملAssessment of Risk of Type 2 Diabetes Among Adults of Banepa Municipality, Nepal: Community Based Cross-Sectional Study
Introduction: The prevalence of type 2 diabetes has been escalating worldwide, including low- and middle-income countries such as Nepal. Early detection of individuals at risk is of the utmost importance to prevent the escalating condition. This study used a simple, cost-effective screening tool known as the Indian Diabetes Risk Score (IDRS) in order to assess the proportion of...
متن کاملComparing Mental and Physical Health of Full-Service Sex Workers in the United States to the Health of the General Population in 2019-2020
Introduction: Full-service sex workers (FSSWs) are relatively prevalent in the U.S. and are known to face criminalization, stigma, and other factors relating to poor health. The main aim of this study was compare the mental and physical health of full-service sex workers in the United States in 2019-2020 Methods: In this prospective cohort study, participants were recruited through national c...
متن کاملTrend of Changes in Blood Sugar and Glycated Hemoglobin in Electronic Health Records, Sari, Iran 2012- 2018
Background and purpose: High prevalence of Type 2 Diabetes Mellitus (T2DM) and its demand for regular care have resulted in a massive amount of electronic health records (EHRs). Hemoglobin A1c (HbA1c) and Fasting Blood Sugar (FBS) levels are two widely-used diagnostic measures for evaluation of diabetes care. This study aimed at assessing Family Medicine Plan (FMP) and exploring the trend of ch...
متن کاملContextualizing Obesity and Diabetes Policy: Exploring a Nested Statistical and Constructivist Approach at the Cross-National and Subnational Government Level in the United States and Brazil
Background This article conducts a comparative national and subnational government analysis of the political, economic, and ideational constructivist contextual factors facilitating the adoption of obesity and diabetes policy. Methods We adopt a nested analytical approach to policy analysis, which combines cross-national statistical analysis with subnational case study comparisons to examine...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید
ثبت ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- Journal of biomedical informatics
دوره 60 شماره
صفحات -
تاریخ انتشار 2016