Simple Tool Identifies Hypoglycemia Risk in Type 2 Diabetes

Liam Davenport

August 23, 2017

A tool that uses standard clinical parameters can identify patients with type 2 diabetes who are at high risk of hypoglycemia-related emergency-department or hospital usage over the next year, a large US study reveals.

In independent validation cohorts, patients categorized as high risk by the tool have a 20- to 23-fold higher risk of healthcare utilization than those categorized as low risk. Moreover, patients in the high-risk groups were significantly more likely to self-report a severe hypoglycemic episode than their low-risk peers (49.7% vs 9.2%; < .001).

"Hypoglycemia is often preventable with the proper clinical attention, and we believe this tool will help focus that attention on the patients who most need it," lead author Andrew J Karter, PhD, division of research, Kaiser Permanente Northern California, Oakland, said in a press release.

The findings, published online August 21 in JAMA Internal Medicine, are now being disseminated with the aid of the Centers for Medicare and Medicaid Services (CMS). Several healthcare systems and organizations, including CMS, the Mayo Clinic, and Kaiser Permanente, are determining how they can apply the risk-stratification tool.

Dr Karter told Medscape Medical News that the tool is highly transportable and could be used in "just about any healthcare setting" because the factors used in the tool are regularly included in patient charts and evaluations.

The tool could help "alert providers or healthcare systems that they have a patient who is at a high risk, and hopefully that would initiate some kind of shared decision making and discussion between the patient and their provider about ways that they can reduce risk," he said.

Developing, Validating the Tool

The risk of iatrogenic hypoglycemia has been underappreciated in type 2 diabetes. However, it is one of the most common adverse events, particularly in older patients, and accounts for one in four emergency hospitalizations for drug-related adverse events.

Moreover, recent estimates suggest that it is associated with direct medical costs in the United States of approximately $1.8 billion every year.

As there are currently no validated methods to identify patients at high risk for hypoglycemia, the researchers conducted a cohort study using prospectively collected data to develop a tool that could identify the 12-month risk of hypoglycemia-related healthcare usage, defined as emergency-department or hospital utilization.

They initially conducted a literature review for clinical risk factors associated with hypoglycemia in type 2 diabetes patients, selecting 122 categorical and 34 continuous clinical, demographic, and behavioral factors that could be useful for developing the model.

Then, using data from the Kaiser Permanente Northern California healthcare system at a baseline date of January 1, 2014, the team identified 206,435 individuals aged ≥21 years with type 2 diabetes. They randomly split the patients into a derivation sample of 165,148 patients for tool development and 41,287 who made up the internal validation sample.

With recursive partitioning, they created a risk-classification tree, with the classification nodes subsequently simplified into high-, intermediate- and low-risk groups.

They then validated the risk-stratification tool using two independent samples of patients with type 2 diabetes: 1,335,966 patients from the Veterans Administration Diabetes Epidemiology Cohort (VA cohort) and 14,972 patients from the Group Health Cooperative in Seattle, Washington (GH cohort).

The final classification tree yielded six patient-specific variables to stratify patients according to their risk of hypoglycemia-related healthcare usage:

  • Total number of prior hypoglycemia-related healthcare use episodes (0, 1–2, ≥3 times).

  • Number of emergency-department encounters for any reason in the prior 12 months (<2, ≥2 times).

  • Insulin use (yes/no).

  • Sulfonylurea use (yes/no).

  • Presence of severe or end-stage kidney disease (yes/no).

  • Age younger than 77 years (yes/no).

Based on the resulting risk stratification tool, the estimated annual risk of hypoglycemia-related healthcare usage was >5% for high-risk patients, 1% to 5% for intermediate-risk patients, and <1% for low-risk patients.

The overall rate of hypoglycemia-related healthcare utilization during the 12 months postbaseline in the derivation sample was 0.49%, compared with 0.47% in the internal validation sample (= .53), 0.30% in the GH cohort (= .008), and 0.51% in the VA cohort (= .27).

When stratified according to the tool, 2.0% of patients in the derivation sample were classified as high risk, 10.7% as intermediate risk, and 87.3% as low risk. Across the internal and external validation samples, the proportion of high-risk patients ranged from 1.5% to 2.6%, while the proportion of low-risk individuals was 82.1% to 87.3%,

The authors note that the tool had high discrimination and good calibration in both internal and external samples, although it overestimated hypoglycemia-related healthcare usage in the intermediate-risk and higher-risk categories in the external samples.

Specifically, the odds ratio of hypoglycemia-related utilization in high-risk vs low-risk patients was 34.6 in the internal validation sample, compared with 23.3 in the VA cohort and 20.7 in the GH cohort (< .001).

In the interview with Medscape Medical News, Dr Karter said, "The inputs needed for the tool are so simple that any healthcare setting or, for that matter, an individual clinician with adequate records on their patients, would be able to classify them as low, intermediate, or at high risk for a hypoglycemic event."

The tool could also be used as a platform to teach patients about the signs, symptoms, and risk factors for hypoglycemia, as well as lifestyle changes to help prevent hypoglycemic events.

"From a healthcare provider perspective, there are a lot of things that could be done, including pharmacological management, deintensifying or switching medications, and potentially continuous glucose monitoring and providing patients with glucagon kits," Dr Karter said.

He noted though that the tool will not identify everyone at risk. "One of the risk factors for hypoglycemia is food insecurity, and there are maybe patients who are poor and don't have enough food in their refrigerator but are trying to continue taking their insulin.

"They may take insulin but skip a meal, and that puts them at high risk for a hypoglycemic event. So, assessing these social determinants of hypoglycemia are important too."

Nevertheless, Dr Karter believes that the development of the tool is a first step in improving the management of hypoglycemia. "It addresses what you would consider the tip of the iceberg, because there's a lot of people who struggle with hypoglycemia who never show up at your emergency department or hospital but still require help."

He hopes that the tool will help increase provider awareness, "and we're hoping that healthcare systems start paying more attention to hypoglycemia."

He added: "We have seen in the past decades a great focus on ways that we can [achieve] good glucose control and that glucocentric focus has maybe come at the expense of patient safety, and so this new focus on hypoglycemia is hopefully balancing that."

Dr Karter concluded: "You can think of it almost as a thermostat. We don't want it too hot, and we don't want it too cold either, so we're hopeful that people will start using these kinds of tools and create interventions to prevent hypoglycemia everywhere."

The research was funded by the US Food and Drug Administration (FDA) under the Safe Use Initiative, a collaborative program aimed at reducing medication-related adverse events.

"This work is an example of how federal agencies can work with private researchers to reduce preventable adverse drug events," John Whyte, MD, MPH, director of professional affairs and stakeholder engagement for the FDA, commented in the press release. "The goal is to identify the patients who are at highest hypoglycemic risk so that healthcare providers can focus their attention on the specific needs of these patients and reduce preventable hypoglycemia harm."

The study was funded by the US Food and Drug Administration. The National Institutes of Health supplied additional funding for the hypoglycemia-related research. Drs Karter is also supported by the National Institute of Diabetes and Digestive and Kidney Disease Centers for Diabetes Translational Research. Disclosures for the coauthors are listed in the paper.  

JAMA Intern Med. Published online August 21, 2017. Article

For more diabetes and endocrinology news, follow us on Twitter and on Facebook .

 

Comments

3090D553-9492-4563-8681-AD288FA52ACE
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.
Post as:

processing....