In older adults (≥65 years old) with type 2 diabetes, the prevalence of “diabetes overtreatment,” defined according to glycated hemoglobin (HbA1c) <7% with the use of insulin or sulfonylureas, ranges from 26% to 40% (1–3). Traditionally, diabetes overtreatment is considered as an imbalance in diabetes management when the anticipated harms of glycemic control outweigh the benefits of glucose-lowering therapy. The American Diabetes Association “Standards of Care in Diabetes” refers to diabetes overtreatment as “intensive glycemic management with medication plans including insulin and sulfonylureas in older adults with complex medical conditions” (4). Although the American Diabetes Association recommends a framework for personalized HbA1c goals in accordance with the health status of older adults, most clinical trials have defined overtreatment according to HbA1c threshold ranging from <6.0% to <8.0%, with and without the use of insulin or sulfonylureas. However, in some studies individualized thresholds of HbA1c were used according to the clinical practice guidelines for older adults (4–7). Therefore, historically, definition of diabetes overtreatment has been vague in the scientific literature.
In older adults with diabetes and multimorbidity, the benefits of intensive glycemic management are counterbalanced by increased risks of hypoglycemic events and overall mortality. Evidence from randomized controlled trials showed reduction in microvascular complications with intensive glycemic control but failed to show the benefit of intensive glycemic control in reducing macrovascular complications (8–11). Additionally, studies in older adults with diabetes show a U-shaped relationship between HbA1c and the risk of mortality (12), demonstrating an elevated risk of mortality with HbA1c values <6% and >11% (13). Therefore, a thorough risk-benefit evaluation is needed before initiation of intensive glycemic management in this heterogeneous population.
In this issue of Diabetes Care, Christiaens et al. (14) investigate the definitions of diabetes overtreatment in older adults and correlated them with time spent in a state of hypoglycemia. This work is an ancillary to the HYPOAGE study, which is a multicenter observational study of older adults (≥75 years old) with type 2 diabetes duration ≥1 year who were on insulin treatment for at least 6 consecutive months (15). Participants were followed for 28 days with a double-blinded continuous glucose monitoring (CGM) device. The purpose of this ancillary report is to compare the incidence of hypoglycemia in participants with and without diabetes overtreatment. The main outcome of the study was ≥1% time spent in a state of hypoglycemia (<70 mg/dL), and the authors defined diabetes overtreatment using fixed (HbA1c <7%) and individualized (HbA1c based on health status of older adults) definitions. Of the 134 patients (mean ± SD age 81.6 ± 5.4 years, 41% female, mean HbA1c 7.9%), approximately 19% and 40% were classified as overtreated based on fixed and individualized proxy definitions, respectively. The study confirmed that both fixed and individualized definitions of diabetes overtreatment had low sensitivity (20% and 41%) and accuracy (27% and 44%) in predicting hypoglycemia. Therefore, the study results suggest that the two HbA1c-based definitions of diabetes overtreatment are poor predictors of hypoglycemia in older adults. Since hypoglycemia is more common and life-threatening in older adults, clinical guidelines of CGM-based hypoglycemic targets are more stringent for older adults (<1% of time spent with glucose below target range) compared with the general adult population (<4% of time spent with glucose below target range) (11). In this study, >90% of older adults spend >1% of time in a state of hypoglycemia (<70 mg/dL), irrespective of whether they were in the cohort with diabetes overtreatment or the cohort without diabetes overtreatment.
One strength of this study is the inclusion of a well-characterized cohort of older adults (>75 years old) using insulin, which is a vulnerable population often prone to diabetes overtreatment and hypoglycemia. Additionally, the authors used a novel concept of predicting CGM-measured hypoglycemia by exploring different definitions of diabetes overtreatment. The study highlights the gaps in defining diabetes overtreatment. Clinically, “overtreatment” should be defined according to hypoglycemia-focused outcomes rather than HbA1c-focused outcomes. Munshi et al. (16) showed that hypoglycemia is common in older adults irrespective of increased HbA1c. Similar results were observed in the Diabetes and Aging Study, where severe hypoglycemia was common in older adults across all levels of glycemic control including those with high HbA1c, mean >9% (17). Therefore, measures of diabetes overtreatment based on HbA1c alone are inherently flawed. An alternate method to define diabetes overtreatment clinically could be by the frequency of symptomatic hypoglycemic episodes. However, recurrent hypoglycemia, insulin use, and longer diabetes duration predispose older adults to impaired awareness of hypoglycemia (18), which weakens their intrinsic ability to recognize the onset of hypoglycemic symptoms. Therefore, this study raises important questions. Is there an unmet need for other CGM-focused determinants of diabetes overtreatment like time spent with glucose below target range and glycemic variability (Fig. 1)? Will including CGM-based metrics in diabetes management of older adults with type 2 diabetes achieve the ideal combination of decreasing microvascular complications associated with hyperglycemia without increasing the risk of hypoglycemia? The current study is limited in answering these questions, as this is an observational, ancillary study. Additionally, the use of HbA1c <7% as the cutoff for “diabetes overtreatment” might not be relevant in the current era of advanced technologies and therapeutics where more normal HbA1c (<6.5%) without hypoglycemia is an achievable target.
Potential road map for diagnosis and management of diabetes overtreatment in older adults with type 2 diabetes. Figure was created in BioRender (biorender.com).
Potential road map for diagnosis and management of diabetes overtreatment in older adults with type 2 diabetes. Figure was created in BioRender (biorender.com).
Eventually, the determination of whether an older adult with diabetes is overtreated and requires deintensification of medication depends on multiple factors such as hypoglycemic profile, life expectancy, polypharmacy, geriatric syndromes, missed meals, and multimorbidity. In the past decade, diabetes technology has revolutionized diabetes management in older adults. The Multiple Daily Injections and Continuous Glucose Monitoring in Diabetes (DIAMOND) (19) and Wireless Innovation for Seniors With Diabetes Mellitus (WISDM) studies (20) showed that use of CGM was associated with decrease in HbA1c, glycemic variability, and time spent in a state of hypoglycemia and increase in time spent with glucose in target range in older adults. Similarly, improved glycemic outcomes and decreased hypoglycemia were demonstrated with the use of automated insulin delivery systems in older adults with type 1 diabetes in the OldeR Adult Closed-Loop (ORACL) trial (21) and Automated Insulin Delivery in Elderly With Type 1 Diabetes (AIDE T1D) study (22). So, the question remains: are we at an intersection for diabetes overtreatment in older adults with type 2 diabetes, where the diagnosis and ideal management may be achieved by incorporating the latest technology? (Fig. 1).
See accompanying article, p. 61.
Article Information
Duality of Interest. No potential conflicts of interest relevant to this article were reported.
Handling Editors. The journal editors responsible for overseeing the review of the manuscript were Steven E. Kahn and Vanita R. Aroda.