New Research Suggests Diabetes Overtreatment in Older Adults

Miriam E Tucker

January 12, 2015

A substantial proportion of elderly adults with diabetes and in poor health are being treated to tight glycemic targets with insulin or sulfonylureas, potentially placing them at increased risk for severe hypoglycemia, a new study reveals.

Results from the analysis of 2001–2010 National Health and Nutrition Examination Survey (NHANES) data are published online January 12, 2015, in JAMA Internal Medicine by Dr Kasia J Lipska (Yale School of Medicine, New Haven, Connecticut) and colleagues.

"Our findings suggest that many older adults with diabetes were potentially overtreated. In other words, they reached a hemoglobin A1c target below 7% despite questionable benefits and potential harms of this approach," Dr Lipska told Medscape Medical News.

Since 2003, guidelines from the American Geriatrics Society (AGS) have endorsed higher glycemic targets for older patients with multiple comorbidities and/or limited life expectancy, due to the concern about severe hypoglycemia in the short term and the lack of expected long-term benefit (J Am Geriatr Soc 2003;51:265–280).

In 2010 the Veterans Administration and the Department of Defense also advised a more cautious approach to diabetes management in the elderly, as did a 2012 joint consensus statement from the AGS and the American Diabetes Association (ADA).

"The problem is that we haven't put this into practice," Dr Lipska said.

Treating to the Wrong Target?

In their study, Dr Lipska and colleagues classified older adults into the three categories endorsed in the AGS/ADA guidelines: the "relatively healthy"; those with complex medical histories for whom self-care may be difficult ("complex/intermediate"); and those with very significant comorbid illness and functional impairment ("very complex/poor").

Among 1288 NHANES patients aged 65 and older who reported having diabetes and for whom at least one HbA1c value during the study period was available, 51% were classified as relatively healthy, 28% had complex/intermediate health, and 21% had very complex/poor health.

Of the total 1288, 62% had an HbA1c value of less than 7% and 42% had HbA1c levels less than 6.5%. There were no significant differences in HbA1c levels across the three health-status groups.

Notably, over a third (38%) of the patients with very complex/poor health had HbA1c levels of less than 6.5%, as did nearly half (45%) of those with complex/intermediate health.

Among those with HbA1c values below 7%, more than half (55%) were treated with either insulin or sulfonylureas (4% were treated with both). And again, this didn't differ significantly by health status; fully 60% of the least healthy patients were using those potentially hypoglycemia-inducing agents, as were 51% of the relatively healthy group and 59% of the intermediate group.

Similarly, among those with HbA1c values less than 6.5%, the proportions using insulin/sulfonylureas were 44% for the relatively healthy, 52% for the intermediate group, and 56% for the least healthy.

And while the use of insulin/sulfonylureas among patients with HbA1c less than 7% declined over time among the relatively healthy (P = .05), it remained stable among both groups with complex health issues (P = .65).

Harms vs Benefits

Dr Lipska acknowledges that the study is limited by the fact that no data on actual rates of hypoglycemia or other harm were available from NHANES.

However, in the paper she and her colleagues point to previous research — including from their group — showing that older people derive less benefit from tight control and are particularly susceptible to the effects of severe hypoglycemia.

"It is a [study] limitation. But we already know that 'tight' glycemic control increases the risk for hypoglycemia. And older people are especially susceptible to hypoglycemia," she told Medscape Medical News.

She added, "We could not tell whether or not anyone was actually harmed by diabetes treatment. What we tried to do is to look at treatment among people who were less likely to benefit and were more likely to be harmed by reaching a 'tight' glycemic target, using drugs that increase that risk for harm."

Because current guidelines focus on reaching HbA1c levels below upper thresholds and don't provide limits at the low end, an individualized approach is needed in older, medically complex patients, she says.

Some experts have suggested adding red flags to electronic health records to warn when a patient at high risk for hypoglycemia reaches an HbA1c target below 7%.

Another approach "is to sensitize both patients and clinicians to this concept of going too low. They need to have ongoing discussions about what intensity of treatment makes sense on a case-by-case basis," Dr Lipska added.

She continues to work on this topic. "I'm interested in more precisely defining harms and benefits of diabetes treatment for older adults. I'd like to be able to tell my patients what they can expect from treatment given who they are, their other medical problems, their history of glucose control, etc. And I want to develop better tools to communicate this information with patients, so they can make informed decisions."

This study was supported by the Pepper Center Career Development Award, the National Institute on Aging, Yale Center for Investigation, and the American Federation for Aging Research. Dr Lipska reports receiving support from the Centers for Medicare & Medicaid Services to develop performance measures that are used for public reporting; disclosures for the coauthors are listed in the article.

JAMA Intern Med. Published online January 12, 2015. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.