Perspective from W. Timothy Garvey, MD
September 01, 2014
6 min read

Prediabetes criteria could be costly for patients, health care system

Perspective from W. Timothy Garvey, MD
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

Tagging people who have moderately high blood glucose levels with a prediabetes label may do more harm than good, according to an analysis by a pair of endocrinologists published in the British Medical Journal.

John S. Yudkin, MD, of the University College London, United Kingdom, and Victor M. Montori, MD, of the Mayo Clinic, Rochester, Minn., contend the American Diabetes Association cut points that define prediabetes are not clinically relevant, are costly for health care systems and stigmatize the patient.

“There are real adverse effects of the prediabetes diagnosis,” Yudkin told Endocrine Today. “The diagnosis should be made only if there is realistic evidence of possible benefits to balance that out.”

Impetus and assessment

The results of a study published in JAMA — showing 50.1% of Chinese adults, or 493.4 million people, had prediabetes based on ADA criteria — prompted the analysis, Yudkin said.

“It just seemed crazy, especially when comparing it to the 8.3% who had impaired glucose tolerance, which is something that is clearly a risk factor for diabetes and heart disease,” Yudkin said.

The ADA changed the diagnosis guidance for prediabetes in 2010, according to the analysis, creating an HbA1c threshold of 6% to 5.7% for the intermediate category. The definition for IGT, however, was established in 1979 and has not changed.

Yudkin called it “bizarre” that it would be possible to expand a category sixfold by changing tests and dropping cut points; he said it is diluting numbers with people at lower risk.

Further, substantial biological and assay variability in diagnosing glucose intolerance, as noted in the analysis, further confound the categorization. With borderline HbA1c measurements twice as high in blacks than whites and glucose tolerance deteriorating with age, Yudkin said demographics deserve more care.

“To many people, this is a US body steamrolling out recommendations that have been poorly thought through in terms of public health, applicability to other countries and other populations, and even to all ethnic groups within the US,” Yudkin said.

John S. Yudkin

With the ADA in disagreement with the International Diabetes Federation (IDF) and WHO on diagnostic criteria, Yudkin said the value of the prediabetes category is under debate.

Since 2006, IDF and WHO have recommended using “intermediate hyperglycemia” to describe levels between normal glucose tolerance and diabetes. WHO adopted HbA1c >6.5% as a top cut point in 2001, but did not define a lower cut point.

“Endocrinologists outside the US should probably start listening to what the IDF and WHO say,” Yudkin said. “Endocrinologists inside the US will probably need to engage with their representative organization, the ADA, in the hope there’s going to be some common sense thinking done.”

Paradigm shift

With diabetes prevention at the forefront of minds — in terms of patient health, clinician concern and public health — the desire to identify people who could be affected in the coming years is a logical thought process, Montori said, but lacks a broader view.

“That narrative is nice and elegant and makes for a compelling case for doing something about it,” Montori said. “What’s striking about it is the very narrow focus on this as being a problem of individuals.”

Under this paradigm, Montori said, people who are overweight, sedentary and at risk for diabetes would appear to have something “wrong” that needs to be “fixed.”

Montori reasons there exists an ecology inside of which people are vulnerable to developing diabetes. He likened it to a school of fish: Certain fish are being asked to swim in the opposite direction but nothing is being done about the “current” — issues including income inequality, social isolation and provision of food for underserved communities.


“By intervening on the individual and leaving the ecology unchanged, we are for the most part committing that patient to failure — unless of course we do something like bariatric surgery, in which we just delay the failure for many years,” Montori said.

Trying to determine whether HbA1c levels are reproducible or track important outcomes are distractions, Montori said, and not the correct conversation for health care specialists and diabetes-focused organizations.

Victor M. Montori

“The discussion we should be having is whether this is a clinical problem or whether this is a societal problem,” Montori said. “This idea of blaming individuals for what is, in fact, a societal problem, and then giving individuals the responsibility for fixing it one person at a time, seems unfair to clinicians, and health care systems and patients.”

Costs and benefits

The psychological toll associated with a prediabetes label may not be quantifiable, but it is something clinicians must thoroughly consider, Montori said.

“The general approach of diagnosing individuals as being at risk is thinking of their situation as a manifestation of a personal failure or a flaw in their character or constitution or biology,” Montori said. “We don’t need a diagnosis to make patients feel afraid of their outcome, or scare people into changing their behaviors.”

Of 100 people diagnosed with prediabetes, fewer than 50 are likely to develop diabetes in the next decade, according to the analysis.

Montori noted that feeling threatened by consequences of future complications may even keep patients from returning to see clinicians, meaning valuable opportunities to develop respectful relationships would be missed.

More tangible burdens of the diagnosis range include medical care and treatment, along with insurance and employment challenges, according to the analysis.

With 86 million Americans, or more than one-third of the US population, satisfying the prediabetes diagnosis, Yudkin said he fears what that would do to health care systems, as well as patients’ out-of-pocket payments and co-payments.

“Could you imagine? Even if 10% of them were put on medication?” Yudkin said. “Making a diagnosis is supposed to be of benefit, or should be done and is likely to be of benefit. If the benefit is unproven, then there are disadvantages in terms of exposure to health care professionals, insurance premiums and possible exposure to side effects of drugs.” — by Allegra Tiver

Xu Y. JAMA. 2013;310:948-958.
Yudkin JS. BMJ. 2014;349:g4485.
John S. Yudkin, MD, can be reached at the Academic Unit of Diabetes and Endocrinology, Department of Medicine of the University College and Middlesex School of Medicine, Whittington Hospital, Archway Road, London N19 5NF, UK; email:
Victor M. Montori, MD, can be reached at Knowledge and Evaluation Research Unit, Division of Endocrinology and Diabetes, 200 First Street SW, Plummer 3-35, Rochester, MN 55905; email: