Email not displaying correctly? View it in your browser.

Vol. 9 No. 26, December 19, 2012

Fact and Stats

Algorithms applied to electronic health record (EHR) data could help identify cases of diabetes, according to a study published in the journal Diabetes Care

For the study, researchers from Harvard Medical School and Harvard Pilgrim Health Care Institute in Boston analyzed EHR data from a large, multi-specialty ambulatory provider serving about 700,000 patients. The data included: laboratory test results, diagnosis codes and prescriptions.

Researchers found that the algorithm provided more comprehensive information about possible diabetes cases than claims codes alone. In total, the algorithm identified 43,177 possible cases of diabetes. Of those cases:

  • 78% had diabetes claims codes in their charts
  • 66% had lab results suggestive of diabetes; and
  • 46% had prescriptions suggestive of diabetes.

When researchers sought to differentiate between cases of Type 1 and Type 2 diabetes, the algorithm correctly classified 35 out of 36 patients with Type 1 diabetes. 

According to the researchers, the algorithm can identify more diabetes cases than claims codes alone and can effectively differentiate between Type 1 and Type 2 diabetes.

Read more

Source: Automated Detection and Classification of Type 1 Versus Type 2 Diabetes Using Electronic Health Record Data

Date: November 27, 2012

Around the World

News from Canada



Industry Speak

Canadian Market Research

Access up-to-date market intelligence on the Canadian healthcare IT landscape based on data from the HIMSS Analytics Canadian DatabaseSM. Research that can help healthcare IT vendors and consultants expand into the Canadian market and become more competitive in the space. ITAC members receive a 15% discount. For more information click here.

Infoway Update

Find Out What's New and Happening at Canada Health Infoway

Worth Noting

Digital Health Summit 2013

Health 2.0 Middle East 2013

2013 Annual HIMSS Conference & Exhibition

eHealth 2013 (Ottawa, ON)

Know. Learn. Implement. Subscribe to the Longwoods eLetter. brings you the latest news, best practices, policy and ideas in eHealth. The subject is salient to advancing the quality of care; our objective is to provide you with clarity and to keep you current. is published every two weeks. You can, of course, opt in [or out] anytime; just click here. Send this to a colleague who might be interested; they too can just sign up. Tell us what you think or what you would like included: email the editor.
ElectronicHealthcare | Longwoods Publishing | HIMSS Analytics | Canada Health Infoway

Copyright 2012 Longwoods Publishing Corporation, publishers of academic, scientific and professional journals. 

Longwoods Publishing Corporation, 260 Adelaide Street, East, #8, Toronto, Ontario, Canada M5A 1N1. Please read our Privacy Statement.

Home | About Us | Subscriptions | | Contact | Unsubscribe Here