Oral Presentation Australian Diabetes Society and the Australian Diabetes Educators Association Annual Scientific Meeting 2014

Hospital-wide HbA1c measurement identifies undiagnosed diabetes in 5% of patients aged 54 years and over: results of the Austin Health Diabetes Discovery Initiative (#5)

Natalie Nanayakkara 1 , Nyuk Pang 1 , Hang Nguyen 2 , Matthew Reynolds 1 , Benjamin Peake 1 , Soenarno Hoetomo 1 , Alyisha Tan 1 , Joel Smith 1 , Graeme Hart 3 , Elizabeth Owen-Jones 4 , Jane Ross 4 , Victoria Stevenson 1 , Raymond Robbins 5 , Leonid Churilov 6 , Que Lam 7 , Omar Farouque 8 , Douglas Johnson 9 , Scott Baker 1 9 , Jeffrey D Zajac 1 10 , Elif I Ekinci 1 10 11
  1. Department of Endocrinology, Austin Health, Melbourne, Victoria, Australia
  2. Department of General Medicine , Austin Health, Melbourne, Victoria, Australia
  3. Department of Intensive Care, Austin Health, Melbourne, Victoria
  4. Austin Centre for Applied Clinical Informatics, Austin Health, Melbourne, Victoria, Australia
  5. Department of Administrative Informatics, Austin Health, Melbourne, Victoria, Australia
  6. The Florey Institute of Neuroscience & Mental Health, Melbourne, Victoria, Australia
  7. Pathology Department, Austin Health, Melbourne, Victoria, Australia
  8. Department of Cardiology, Austin Health, Melbourne, Victoria, Australia
  9. Department of General Medicine, Austin Health, Melbourne, Victoria, Australia
  10. Department of Medicine, University of Melbourne, Heidelberg, Victoria, Australia
  11. Menzies School of Health Research, Darwin, Northern Territory, Australia


Hospital admission represents an opportunity to diagnose and manage diabetes mellitus.  As HbA1c measurement is superior to other tests for diabetes diagnosis in the hospital setting1, we aimed to investigate the prevalence of undiagnosed diabetes and poor glycaemic control using routine HbA1c testing in inpatients at Austin Health, a tertiary referral centre.  


Patients (aged ≥54 years) admitted between June 2013 and January 2014 had routine HbA1c testing via an automated order using the Cerner Millenniumⓒ Health IT System.  Patients were classified as having diabetes if the HbA1c was ≥6.5% and poor glycaemic control if the Hba1c≥8.5%1.  A history of diabetes was obtained from the hospital medical record.


Patients (n=5083, 6716 admissions, mean age 72.9±10.8 years) were divided into three categories: those with a prior history of diabetes (28.6%, CI 27.3-29.8%, n=1453), no prior history of diabetes and HbA1c<6.5% (66.1%, CI 64.8-67.4%, n=3359), and no prior history of diabetes (5.3%, CI 4.7-6.0%, n=271) and HbA1c ≥6.5%. Medical units had a higher rate of undiagnosed diabetes compared to surgical units (5.9%, [CI 5.1-6.9%] vs 4.3% [CI 3.4-5.3%] p=0.008). Readmissions during the study period were higher in patients with an HbA1c ³6.5% with and without a prior history of diabetes (26%, CI 23.6-28.2% and 23%, CI 18.4-28.7%, respectively), compared to patients with an HbA1c <6.5% and no prior history of diabetes (21%, CI 20.4-23.2%) (p=0.007).  No significant differences in rates of intensive care unit admission (9.1%, CI [8.1-10%], 10.6% [CI 9.1-12.3%] and 7.0% [CI 4.2-10.7%] respectively, p=0.105) or inpatient mortality (11.4%%, CI [10.3-12.5%], 11.8% [CI 10.2-13.5%] and 14.4% [CI 10.4-19.1%] respectively, p=0.324) were observed between the groups.


Approximately 5% of inpatients ≥54 years attending our tertiary hospital have undiagnosed diabetes. Routine inpatient HbA1c testing addresses a currently missed opportunity to identify patients with newly diagnosed diabetes and poor glycaemic control. The impact of early identification and treatment of poor glycaemic control on patient outcomes requires further study.

  1. International Expert Committee report on the role of the A1c assay in the diagnosis of diabetes. Diabetes Care 2009; 32: 1327-1334.