Kathleen Louden

September 24, 2013

CHICAGO — Of the less than 20% of patients with diabetes who meet the nationally recommended frequency for glycated hemoglobin (HbA1c) testing, more than a third are not prescribed treatment changes when their levels increase significantly, a new study has found.

"We're hearing about overuse of this test, but even when it is used appropriately, the result may not be," said lead investigator Ross Molinaro, PhD, from the Emory University School of Medicine in Atlanta.

He presented the findings here at the American Society for Clinical Pathology (ASCP) 2013 Annual Meeting.

Dr. Molinaro and his coworker, Jaime Noguez, PhD, also from Emory University, assessed clinician practice in ordering HbA1c testing, which is a measure of glycemic control over an average of 3 months.

They retrospectively reviewed the medical charts of 26,435 patients who underwent more than 32,000 HbA1c tests during a 1-year period at their laboratory. Only 4380 patients received more than 1 test in that period.

The American Diabetes Association (ADA) and other organizations recommend that people with diabetes have an HbA1c test every 6 months if their plasma glucose levels are within target range, and every 3 months if they are not achieving glycemic control or if their therapy has changed.

"From our data, there appears to be an approximate equal distribution of patients who were tested too often and those who were not tested often enough," Dr. Molinaro reported.

The researchers analyzed data from the 737 patients who underwent HbA1c testing at the recommended frequency. They used reference change value — the change in HbA1c level between 2 tests — to assess the statistical significance of differences in serial laboratory results from individual patients, and determined that a change in HbA1c of 0.6% or more in their population was statistically significant, according to Dr. Molinaro.

Nineteen patients with initial glycemic control had a statistically significant increase in HbA1c level on subsequent testing, the researchers report.

Table. Patients Meeting the Recommended Frequency for HbA1c Testing

Outcome Patients %
Met frequency recommendations 737/4380 17
Initially in glycemic control (HbA1c < 7%) 410/737 56
Subsequently out of glycemic control (HbA1c ≥ 7%) 64/410 16
Significant increase in HbA1c on subsequent test (≥0.6%) 19/64 30
No treatment change after significant increase in HbA1c 7/19 37


The investigators surveyed the 15 treating endocrinologists to determine whether they had adjusted their patients' treatment regimen, as recommended in the ADA guidelines. The 5 responding physicians indicated that they had recommended a change in medication or lifestyle in 63% of cases.

Dr. Molinaro said the survey results show that endocrinologists are unclear about what constitutes a statistically significant reference change value and would find that information useful. An informal survey of the nonresponding endocrinologists elicited similar comments, he noted.

"Our study suggests that better communication between the lab and the clinician is needed on when to follow up with a patient who has a significant change in HbA1c," Dr. Molinaro said. "I think the onus is on the lab to find a better way to communicate test results."

Although pathologists and laboratory professionals are required to report the reference range, they are not required to report reference change values, Dr. Molinaro said. He noted that his institution plans to add this information to reports of serial HbA1c test results, along with an interpretation of the significance of the change.

"We'd like to see if there is any change in the way physicians manage their patients because of this information," he said.

Better communication between the lab and the clinician is needed on when to follow up with a patient who has a significant change in HbA1c.

There is no universally agreed-upon algorithm to determine whether an increase in HbA1c level is statistically significant, explained Daniel Einhorn, MD, president of the American College of Endocrinology, who was asked by Medscape Medical News to comment on the study.

"I'm not aware of any clinician using reference change values," he said.

Dr. Einhorn, who is medical director of the Scripps Whittier Diabetes Institute in San Diego, stressed that a statistically significant change in HbA1c does not necessarily mean it is clinically meaningful. Even when it is, a clinician must consider many variables when deciding whether to change a patient's treatment, he said.

"The decision to add a drug or more education time is not without consequence, whether it is cost or potential side effects," he said.

"There is concern that some patients are not getting HbA1c testing often enough, but I believe this study overstates the problem," Dr. Einhorn noted.

"We are aware of the many assumptions made in this study, and the fact that, currently, the reasons behind the lack of treatment observed aren't well understood, " Dr. Molinaro said. "However, establishing an open line of communication about reference change values, along with subsequent HbA1c results, could possibly assist clinicians in their treatment decisions. "

It is possible that the study data do not provide the complete picture, said ASCP president-elect William Finn, MD, who was not involved in the study. Some patients who did not receive HbA1c testing at the recommended interval might have been tested at another institution that did not communicate the results to Emory, he noted.

Still, the study highlights a well-known issue, Dr. Finn told Medscape Medical News. "We can be efficient at feeding information to clinicians, but the question is how to close the communication loop to make sure the physician is counseled on what the test result means."

Dr. Finn, who is medical director of the Warde Medical Laboratory in Ann Arbor, Michigan, said the need to communicate a significant change in HbA1c results can apply to other laboratory tests as well. The ASCP, he noted, is championing a greater exchange of information between laboratory personnel and clinicians, largely through education to its members.

"More and more pathologists realize they need to be firmly implanted in the process of best practices and communication with the clinical teams," Dr. Finn said.

Dr. Molinaro reports receiving reagents and funding for diabetes research from Bio-Rad Laboratories, Sebia Electrophoresis, and Siemens Corp. Dr. Einhorn and Dr. Finn have disclosed no relevant financial relationships.

American Society for Clinical Pathology (ASCP) 2013 Annual Meeting: Scientific poster 2-37. Presented September 20, 2013.


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.