Inpatient Glucose Values: Determining the Nondiabetic Range and Use in Identifying Patients at High Risk for Diabetes

Mary K. Rhee, Sandra E. Safo, Sandra L. Jackson, Wenqiong Xue, Darin E. Olson, Qi Long, Diana Barb, J. Sonya Haw, Anne M. Tomolo, Lawrence S. Phillips

Research output: Contribution to journalArticlepeer-review

6 Scopus citations


Background: Many individuals with diabetes remain undiagnosed, leading to delays in treatment and higher risk for subsequent diabetes complications. Despite recommendations for diabetes screening in high-risk groups, the optimal approach is not known. We evaluated the utility of inpatient glucose levels as an opportunistic screening tool for identifying patients at high risk for diabetes. Methods: We retrospectively examined 462,421 patients in the US Department of Veterans Affairs healthcare system, hospitalized on medical/surgical services in 2000-2010, for ≥3 days, with ≥2 inpatient random plasma glucose (RPG) measurements. All had continuity of care: ≥1 primary care visit and ≥1 glucose measurement within 2 years before hospitalization and yearly for ≥3 years after discharge. Glucose levels during hospitalization and incidence of diabetes within 3 years after discharge in patients without diabetes were evaluated. Results: Patients had a mean age of 65.0 years, body mass index of 29.9 kg/m2, and were 96% male, 71% white, and 18% black. Pre-existing diabetes was present in 39.4%, 1.3% were diagnosed during hospitalization, 8.1% were diagnosed 5 years after discharge, and 51.3% were never diagnosed (NonDM). The NonDM group had the lowest mean hospital RPG value (112 mg/dL [6.2 mmol/L]). Having at least 2 RPG values >140 mg/dL (>7.8 mmol/L), the 95th percentile of NonDM hospital glucose, provided 81% specificity for identifying incident diabetes within 3 years after discharge. Conclusions: Screening for diabetes could be considered in patients with at least 2 hospital glucose values at/above the 95th percentile of the nondiabetic range (141 mg/dL [7.8 mmol/L]).

Original languageEnglish (US)
Pages (from-to)443.e11-443.e24
JournalAmerican Journal of Medicine
Issue number4
StatePublished - Apr 2018

Bibliographical note

Funding Information:
Authorship: All authors had access to the data and a role in writing the manuscript. MR researched the data, MR, WX, SES, and QL performed the statistical analyses, MR drafted the manuscript, and AT, SJ, DO, LP, and all of the other authors contributed to the discussion and reviewed/edited the manuscript. The authors gratefully acknowledge the contributions of Christine Jasien, Atlanta VAMC who assisted the authors with obtaining VINCI data. Her contribution was without compensation. Mary Rhee (Atlanta VA Medical Center and Emory University School of Medicine) had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Drs. Rhee, Phillips, Olson, and Tomolo are supported in part by the VA. This work is not intended to reflect the official opinion of the VA or the US government government. Clinical Research Study

Funding Information:
This work was supported in part by the US FDA (Federal Drug Administration) award RO1FD003527 (LSP), Veterans Affairs Health Services Research & Development Investigator-Initiated Research award 07-138 (LSP, SLJ), National Institutes of Health awards R21DK099716 (LSP, QL, and SLJ), DK066204 (LSP), U01 DK091958 (LSP and MKR), and U01 DK098246 (LSP and DEO), and Cystic Fibrosis Foundation award PHILLI12A0 (LSP). It is also supported in part by the National Center for Advancing Translational Sciences of the National Institutes of Health under award number UL1TR000454 . The sponsors had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript.


  • Diabetes mellitus
  • Epidemiology
  • Screening

Fingerprint Dive into the research topics of 'Inpatient Glucose Values: Determining the Nondiabetic Range and Use in Identifying Patients at High Risk for Diabetes'. Together they form a unique fingerprint.

Cite this