Reply to “Comment on: Editor’s Note on Clin‐ <scp>STAR</scp> Corner: Practice‐Changing Advances in Diabetes Care”

Dear Editor I thank Drs. Kahkoska and Pratley for their comments on the Editor's Note I wrote about continuous glucose monitors (CGMs). I know and have tremendous respect for Dr. Pratley's work on diabetes, especially having had the opportunity to visit him at the AdventHealth Diabetes Institute in Orlando where he serves as a medical director. I agree that diabetes is a common and highly morbid condition in older adults and I understand that type 1 diabetes is increasing in this population, among whom discontinuing insulin is not an appropriate option. I also recognize that CGMs can be helpful in younger, healthier, and cognitively intact older diabetics. However, I maintain that their feasibility and impact on outcomes in older type 2 diabetics, especially among those with cognitive impairment, remain to be proven. Drs. Kahkoska and Pratley state in their letter: (1) “Importantly, the suggestion that many older adults are unable to use CGMs due to age-related impairments is not supported by available evidence”; and (2) “categorical statements about the abilities of older adults are not only inaccurate, they are harmful.” They cite three publications to support these statements. One is a study coauthored by Dr. Pratley involving 203 patients in 20 clinics with a mean age of 68 [1]. Of the 84 participants in the CGM group, 16 had cognitive impairment based on the NIH Toolbox Cognition Battery. Cognition was a secondary outcome, but no subgroup analyses of outcomes in the 16 participants in the CGM group with baseline cognitive impairment was reported. In another study cited, 1954 patients with coexisting diabetes and MCI/ADRD were identified, among whom only 50 (2.6%) were CGM users [2]. In adjusted regression analyses these 50 participants had 80% lower mortality over the 4-year study period than non-users but also had 130% higher mortality when hospitalized for hypoglycemia. The details of cognitive function were not included, and despite adjustment for many medical conditions it is possible that healthier participants may have been using CGMS, thus biasing the results in favor of CGM use. In a third study cited, in which cognition was measured in 49 eligible people with cognitive impairment, only 17 consented and 5 withdrew. Thus only 12 participated [3]. The device was reported as acceptable to these 12 participants, and carers reported greater ease in monitoring the participant's glucose concentrations. However, completeness of data capture varied considerably due to the need for users to conduct ≥ 3 scans per day. Having served in the editorial leadership of JAGS for the past 33 years, and now in my fifth year as Editor-in-Chief, I do not know any experienced Geriatrics researchers who would identify these studies as rigorous evidence that CGMs have been shown to improve outcomes in older people with cognitive impairment. Further, I do not believe in categorical statements and do not intend my comment as such, and I certainly do not want to cause harm. I am a firm believer in person-centered and evidence-based care, and my response to anyone who asks me what to do about an individual patient is always “It depends.” In this situation, I would rather err on the side of assuming that current versions of CGMs have not been shown to improve outcomes in most cognitively impaired older diabetics. To do so could, in fact, increase the already tremendous burden of iatrogenic hypoglycemia that results in hundreds to thousands of emergency department visits, hospitalizations, and related complications and costs associated with inappropriately aggressive treatment of diabetes in vulnerable older adults in the US annually [4]. The author is solely responsible for the content of this editorial. The author has nothing to report. The author declares no conflicts of interest. This publication is linked to a related comment by Kahkoska and Pratley. To view this article, visit https://doi.org/10.1111/jgs.70077.