Home / Conditions / Type 1 Diabetes / Reclassifying Diabetes: is it Time for an Expanded Definition?

Reclassifying Diabetes: is it Time for an Expanded Definition?

Oct 15, 2016

eden-millerGuest Editorial by Dr. Eden Miller

For those healthcare providers who treat diabetes with any kind of regularity, it has become increasingly obvious that several of our patients fall outside the current diabetes classification system. With individualization of patient care being the new focus, it can be challenging to accomplish this, when certain diabetic individuals don’t “conform” to the current definitions.  At the 2014 American Diabetes Association Session in San Francisco, Dr. Stan Schwartz posed the question “Isn’t it time for a new approach to the diagnosis of Diabetes?” Recently, Dr. Schwartz introduced, “Beta-cell classification of Diabetes and the Egregious Eleven.”miller-callout-schwartz2

If we start by asking a few questions, the need for a reclassification or enhanced definition of diabetes becomes more apparent.

Such as…

  •       Is a patient with type 1 diabetes immune to developing type 2 diabetes in their lifetime, or can they possess features of both, such as dysmetabolic syndrome?
  •       Is it possible for the patient with type 2 diabetes to progress to insulinopenia, become ketosis prone, or even possess immune markers during the course of their disease?
  •       Are there type 1 diabetic patients who never test positive for antibodies, and are thus non-autoimmune yet insulin dependent?
  •       Is it unusual for a patient with type 1 diabetes to have issues with high BMI and insulin resistance?   
  •       What are the special care considerations for type 2 diabetes patients who don’t present with insulin resistance features or are not obese?
  •       What are the key factors of type 2 diabetes, and shouldn’t we utilize the medications that address these physiologic elements that are beyond glucose management?

and the questions keep coming….

Diabetic specialists routinely use their knowledge of the etiology, contributory features, and variations in disease presentation to guide individualized treatment for their patients. With all the variations in the presentation of diabetic patients, a comprehensive classification system that considers not only phenotypic presentation but genomic and immunologic characteristics as well needs to be developed. This new Variance Classification (VC) system could provide a better understanding of the landscape we call diabetes. It would lay a foundation for more directive and potentially effective clinical trials, leading to revolutionary treatment options.  Added treatment choices would truly individualize care, and would be based on presenting variations in disease expression.

By expanding the classification system for diabetes, we will equip healthcare professionals to better address the unique safety needs of the diabetic patient, as well as allow us to keep pace with the individuals’ disease progression, emerging treatment options, and future innovations in genetic testing.

Currently, similar classification systems are utilized in other diseases such as breast cancer’s TNM staging. Advances made in identifying estrogen receptors, genetic testing (i.e. BRCA genes) and nodal involvement were incorporated into the diagnosis and then into clinical trials resulting in revolutionary treatment protocols and improvements in survival rates. In addition, the Chronic Kidney Disease classification system is considered a major advancement that has led to better outcomes, treatments, and investigational trials.   

This need for an expanded classification has become more apparent with the recent FDA report citing cases of diabetic ketoacidosis in patients using Sodium Glucose Co-Transporter 2 Inhibitors (SGLT2-I) for diabetes treatment.

Further review of these cases revealed several of the patients were antibody positive and thus may have been undiagnosed type 1 DM patients.  There were additional cases of type 2 DM individuals who were insulin deficient, had increased ketone production due to recent surgery, illness or dietary changes.  The glucose threshold for ketosis identification appears to be lowered in these DKA patients and it is imperative that the “at risk patients” be identified and provided individualized treatment recommendations.

There is probably a high level of consensus for the necessity to expand the definition of diabetes, but the real challenge will be the definitions or modifiers that are chosen to achieve the desired outcome.

Where do we begin?

This process will undoubtedly evolve over time, but there are a few areas of consideration that are foundational. This Aauthor doesn’t believe the classification will be served best with a type 1, 1 ½, 2, 3, 4, and 5 approach.  Rather, a suggested nomenclature would include information about what is currently known regarding the etiology of diabetes and then incorporate “modifiers” that describe immunologic, genomic or phenotypic.

All patients with diabetes share one commonality: that is elevated glucose. From there, it varies greatly.  One variance category would include immunologic antibody status against GAD, Islet, Zinc, and a host of others revealed in the future.  Positive findings of antibodies suggest an autoimmune chronic condition that is ketosis prone, insulin requiring, with inflammatory undercurrents.

A second variance indicator would establish if the patient was predisposed to ketosis or is insulinopenic with regards to their physiologic needs.  Is this tendency inherently positive or could it be acquired over time with disease progression?

Now we shift to body habitus or phenotype. Is the patient obese by definition or presents with markers of insulin resistance such as elevated lipids, or central adiposity?  Possessing these features in the course of a patient’s disease can alter the treatment response, and in some cases require a therapeutic change in direction.

A final category would be family history or genetic expression.  Frequently a family history of diabetes or its inherited disease modifiers are often discarded or not considered in a patient’s individual care.  In the future, advancements in genetic testing for diabetic markers may further clarify the similarities, unique features, and areas of overlap of the disease presentation.

These features could be integrated into a new classification system that would leave room for new advances in testing and discoveries.  It needs to be simple enough to be accepted and utilized, yet broad enough to identify the disease’s components.

One possible way is to add on the current system of type 1 and type 2 by including modifiers.  The current classification system would be expanded upon by adding the term variance when an individual’s disease presentation or progression falls outside of the current system.


       Type 1                        Area of Variance                      Type 2
Antibody positiveKetosis ProneInsulin Requiring

Phenotype Lean

Insulin Sensitive

Antibody Negative

Ketosis Unlikely

Exogenous Insulin Varia

Phenotype Obese

Insulin Resistance

    —————-Family History/Genetic Typing—————-

When a patient presents or progresses in their disease with characteristics beyond the current definition, A V+ for variance would be added to the T1 (Type 1) or T2 (Type 2) nomenclature.

Then the modifier that the patient exhibits would be added to the variant. Such as Ab +/- for antibody status, K +/- for Ketosis prone, IR +/- for insulin resistance, P 1/2/3 for phenotypic features, and in the future G typing for Genetic Allele expression.

A few examples of this system in a patient may look like the following.

40-year-old Female presents with elevated glucose of 322

BMI is 26

GAD, Islet, Zinc Antibody Negative  (Ab-)

C peptide is .4 (Insulinopenic) Urinary Ketones; moderately positive

Family history Positive for Type 2 DM in Father.

Laboratory Serum Fasting Glucose year prior was 89.

Initial treatment of elevated glucose with 8 units of basal

Insulin QD showing a sensitive responsiveness to insulin resulted in AM fasting glucose average of 152 after 3 days

The patient seems to share the essential features for treatment with Type 1 but still varies since she is antibody negative.

This 40-year-old female would be a Type 1 V Ab-

In other words, she is a most similar to a type one diabetic, who is ketosis prone, is sensitive but needs insulin support, but antibody negative, with a family history of Type 2 diabetes and genetic markers to be determined its impact in the future.

Another instance could present like this.

27 year old Male Type 1 DM from age 13

GAD, Islet, Zinc Antibody Positive

BMI 38.4  (P3)

Elevated LDL, TG’s and Low HDL, and

24 hours total insulin dose 120 units (IR+)

A1c 8.0

Family History Grandfather with Type 1 DM and Mother with Type 2 DM

We would modify his diagnosis to include this insulin resistance features and elevated BMI.  His variance would be Type 1 V P3, IR+

In order to customize and improve individualized patient care, a better system needs to be created (a container to pour the information of the future into it).  We need to start treating the patient as a whole, and address these diabetic variances. This new classification system could be foundational for the future in diabetes prevention, treatment, and research.   The question is not whether it is needed but rather how is it best accomplished.

Eden M. Miller D.O.
Family Practice Sisters, Oregon for High Lakes Health Care
Diabetes Nation – CEO     


Schwartz, Stanley S “Isn’t it time for a new approach to the diagnosis of Diabetes? Presentation 2014 American Diabetes Association Session San Francisco, California

Schwartz, Stanley S. “The Time Is Right for a New Classification System for Diabetes: Rationale and Implications of the B-Cell-Centric Classification Schema.” Diabetes Care 39 (2016): n. page Web. 18 Jan. 2016.


Dr. Eden Miller graduated from Kirksville College of Osteopathic Medicine in Missouri in 1997. In 2001, she completed her residency at Eastmoreland Hospital in Portland, Oregon. Eden was a board member of the Osteopathic Physician and Surgeon Board of Oregon and past vice-president of the American College of Osteopathic Family Physicians of Oregon. Her practice is family oriented with a subspecialty in diabetes care. “I focus on diabetes because in medical school I contracted the disease and I am determined to teach patients how to become experts on their own disease.” Eden is also a national lecturer for diabetes treatment and prevention.