In older adults with diabetes, cognitive impairment can substantially complicate treatment. To reduce hypoglycemia risk in older adults with diabetes and cognitive impairment, guidelines recommend higher glycemic targets and, in severe cases, the abandonment of glycemic goals. The avoidance of agents associated with a high risk of hypoglycemia (e.g., sulfonylureas and insulin) is also recommended (1). However, the extent to which these recommendations are followed has not been well documented. We undertook this study to characterize the prevalence of the use of high-risk glucose-lowering agents in U.S. adults aged ≥65 years with diabetes according to cognitive status.
We analyzed the National Health and Nutrition Examination Survey (NHANES), which uses a multistage probability sampling design to accurately reflect the health status of the U.S. population. We conducted a cross-sectional study of participants aged ≥65 years with self-reported diagnosed diabetes from NHANES 2011–2014, the most recent years with complete cognitive testing (2).
Participants aged ≥65 years underwent standardized cognitive testing. Scores on the Delayed Word Recall, Animal Fluency, and Digit Symbol Substitution tests were converted into z scores, which were then averaged to produce a global cognitive function score. Consistent with prior studies, we defined cognitive impairment as a value in the bottom 25th percentile for global cognitive function (3). Medications were reviewed and classified using a standardized classification system (Multum Lexicon).
We evaluated the prevalence of high-risk glucose-lowering agent use (sulfonylureas, insulin, or glinides) in people with or without cognitive impairment. All analyses were performed using Stata v18.0, and we used survey methods and weighting to generate nationally representative estimates.
We included 611 older adult participants with diabetes (mean age 73 years, mean diabetes duration 14 years, mean HbA1c 7.0%, 50% identifying as female, and 70% identifying as non-Hispanic White). The prevalence of cognitive impairment was 39% (95% CI, 33–45). The percentages of individuals using any high-risk glucose-lowering agent was 64% (95% CI, 54–73) in people with cognitive impairment and 49% (95% CI, 41–56) in those without impairment (Fig. 1). In people with diabetes and cognitive impairment, 31% (95% CI, 22–41) were using insulin.
Prevalence of high-risk glucose-lowering medication use in U.S. adults aged ≥65 years with diabetes according to cognitive status. Data are from NHANES 2011–2014 (2).
Prevalence of high-risk glucose-lowering medication use in U.S. adults aged ≥65 years with diabetes according to cognitive status. Data are from NHANES 2011–2014 (2).
Using the most recent available national data, we found that approximately 2 in 3 older adults with diabetes and cognitive impairment in the U.S. are using high-risk glucose-lowering agents. This includes one-third who are using insulin. Our data indicate that, despite treatment deintensification recommendations for older adults with cognitive impairment, the use of high-risk glucose-lowering agents remains common.
Previous studies have demonstrated that treatment deintensification is uncommon among adults with diabetes, with some groups being at higher risk of potentially inappropriate intensive treatment. Older adults with multiple medical comorbidities are at high risk for inappropriate intensive treatment with glucose-lowering agents (4). Clinical inertia may be a key barrier to deintensification. We extend prior research by establishing the prevalence of high-risk glucose-lowering agent use according to cognitive status in a nationally representative sample of older adults with diabetes.
The American Diabetes Association recommends routine screening for cognitive impairment in all adults with diabetes aged ≥65 years, which can help guide appropriate diabetes pharmacotherapy (1). Our findings in the general U.S. population of older adults with diabetes suggest high rates of unrecognized cognitive impairment.
Study strengths include the nationally representative sample, objective assessment of medication use, and standardized testing for cognitive function by trained personnel. Study limitations include limited sample size and an inability to distinguish between diabetes types. However, considering the study is nationally representative, it is likely that most individuals included have type 2 diabetes.
In summary, the use of high-risk glucose-lowering medications is common among older adults with diabetes and cognitive impairment. Implementing guideline-recommended cognitive screening in older adults with diabetes could help identify patients in whom deintensification is warranted.
Article Information
Acknowledgments. E.S. and J.E.-T. are editors of Diabetes Care but were not involved in any of the decisions regarding review of the manuscript or its acceptance.
Funding. E.S. was supported by National Institutes of Health (NIH)/National Heart, Lung, and Blood Institute (NHLBI) grant K24 HL152440. M.F. was supported by NIH/National Institute of Diabetes and Digestive and Kidney Diseases grant K01 DK138273. J.E.-T. was supported by NIH/NHLBI grant K23 HL153774. E.E. was supported by NIH/NHLBI grant 5T32HL007024.
The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Author Contributions. E.E. researched data, performed the analysis, and drafted the manuscript. E.S., M.F., and J.E.-T. contributed to the analysis and reviewed and edited the manuscript. All authors approved the final version of the manuscript. E.E. had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. E.E. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Handling Editors. The journal editor responsible for overseeing the review of the manuscript was Matthew C. Riddle.