Continuous Glucose Monitor with Siri Integration Improves Glycemic Control in Legally Blind Patients with Diabetes

Abstract

Continuous glucose monitoring (CGM) is safe and effective in improving glycemic control in insulin-treated patients with diabetes. However, the safety of CGM use in legally blind patients with diabetes is unknown. This is a retrospective study of seven legally blind patients with diabetes on intensive insulin therapy who were using Dexcom G6 with voice-enabled Apple's Siri feature. CGM metrics, hemoglobin A1c, and frequency of severe hypoglycemia were measured for 12 months. There was a significant reduction in A1c after 3, 6, and 12 months of Dexcom G6 with Apple Siri use (P < 0.0001). Time-in-range increased at 12 months (50.9% ± 5.5% at 3 months vs. 56.8% ± 5.5% at 12 months, P = 0.029) without increase in time spent in hypoglycemia. There was a significant reduction in severe hypoglycemia requiring medical assistance for 12 months. Voice-enabled CGM use improved glycemic control and reduced severe hypoglycemia in legally blind patients with diabetes on intensive insulin therapy.

Introduction

Diabetes management can be challenging in visually impaired patients with diabetes on intensive insulin therapy. The use of diabetes technologies is known to improve glycemic control; however, its use may be challenging in visually impaired patients with diabetes and its safety is unknown in this population.¹ As an example, none of the older versions of insulin pumps exhibited good low vision accessibility for a safe use in visually impaired patients with diabetes.^2,3 Low vision accessibility is unknown for current insulin pumps in the United States. Dosing errors and glucose monitoring difficulties can cause serious health consequences in this vulnerable population.⁴ A few glucose meters have the feature of reading aloud of glucose results for visually impaired patients; however, they do not have alerts and glucose trends to optimize glucose control.⁵

Dexcom G6 is a factory-calibrated real-time continuous glucose monitor (CGM) that has Apple Siri integration. Dexcom G6 CGM users can ask Siri to recite their Dexcom glucose readings out loud and display their graph directly on the locked screen. This feature allows patients to learn glucose levels hands free. However, its use in legally blind patients with intensive insulin-treated diabetes has not been previously evaluated. In this observational study, we report the use of the Dexcom G6 Siri feature in seven legally blind patients with diabetes on intensive insulin therapy on glycemic control for 12 months.

Research Design and Methods

In this single-center retrospective study, data on seven legally blind adults with diabetes who were using Dexcom G6 with voice-enabled Apple's Siri feature were retrieved from electronic medical records. Compared with baseline, A1c and severe hypoglycemia requiring a third person or medical assistance for 12 months were analyzed. Time-in-range (TIR, defined as the percentage of time spent 70–180 mg/dL) and time below range (TBR, defined as the percentage of time spent below 70 and 54 mg/dL) were also analyzed after 3, 6, and 12 months of use. Legal blindness is defined as either a visual acuity of 20/200 or less in the better-seeing eye with best conventional correction (meaning with regular glasses or contact lenses) or visual field (the total area an individual can see without moving the eyes from side to side) of 20° or less (also called tunnel vision) in the better-seeing eye. All patients in the study were not able to do basic daily living tasks independently. Data presented as mean ± standard deviation (SD) for continuous variable and paired Student's t-tests were used to assess significant changes at each time point. Linear mixed models with repeated measures were used to examine changes in A1c and CGM parameters (TIR and TBR) over the study period. This study was approved by IRB under exempt category.

Results

Of seven legally blind adults with intensive insulin-treated diabetes (four women), six had type 1 diabetes and one had Wolfram syndrome. Four of them were using insulin pens and three of them were using an insulin pump. One of them had chronic kidney disease and was on hemodialysis, and two of them had a history of renal transplant with current normal kidney function. The mean ± SD of age of patients was 49.4 ± 15.9 years (range 18–66 years) with baseline A1c mean ± SD of 8.6% ± 1.0% and diabetes duration of 42.8 ± 13.4 years. Six patients were naive to CGM use and one patient had previous experience of CGM but had not used CGM for >12 months. All patients were legally blind >3 years. One patient had three episodes and five patients had one episode of severe hypoglycemia, but no diabetic ketoacidosis in the 12 months before the initiation of Dexcom G6 with Siri. For 12 months of Dexcom G6 with Siri use, there were no episodes of severe hypoglycemia or diabetic ketoacidosis in any of these seven patients. Compared with baseline, there was a significant (all P < 0.0001) reduction in A1c after 3, 6, and 12 months of Dexcom G6 with Apple Siri use (Fig. 1A). TIR was also increased at 12 months compared with baseline (mean ± standard error [SE]; 50.9% ± 5.5% at 3 months vs. 56.8% ± 5.5% at 12 months, P = 0.029) (Fig. 1B) without increase in hypoglycemia (mean ± SE time <70 mg/dL at 3 months [2.3% ± 0.4%], 6 months [1.9% ± 0.4%] and 12 months [2.2% ± 0.4%], all P > 0.05) and severe hypoglycemia (mean ± SE time <54 mg/dL at 3 months [0.6% ± 0.2%], 6 months [0.6% ± 0.2%] and 12 months [0.4% ± 0.2%], all P > 0.05).

FIG. 1.

Change in A1c for 12 months compared with baseline (A) and change in time-in-range for 12 months (B) after Dexcom G6 Apple Siri use in seven legally blind patients receiving intensive insulin treatment. Data presented as mean ± SE. SE, standard error.

Conclusions

Diabetes management is challenging in visually impaired patients with diabetes who require intensive insulin treatment. Insulin administration products such as pens and pumps and glucose monitoring systems are not widely optimized for visually impaired patients with diabetes. Current literature is limited and there are no guidelines how to manage diabetes in patients with severe visual impairment. Dexcom G6 with Siri feature is the first and only real-time CGM that can be used by visually impaired patients to access their glucose levels. Patients inquire glucose reading and direction of change with the voice command of “Siri, what's my blood sugar/glucose?” However, Dexcom G6 Siri feature is passive, meaning by that it will not alert patients by itself. Therefore, patients with diabetes must use Dexcom app on the smartphone with customizable loud alert settings to be alerted for hypoglycemia and hyperglycemia. Once alerted by the Dexcom app, the patient can ask Siri to know glucose level and direction of change to be proactive to improve diabetes control and minimize hypoglycemia. The high and low alerts from Dexcom can be customized and are audibly recognizable and distinguishable, such that they are still meaningful to a blind user. Instant accessibility to glucose levels on CGM may decrease the constant supervision of visually impaired patients on intensive insulin therapy to caregivers and may decrease the response time for hypoglycemia.

To our knowledge, this is the first study reporting improvement in glycemic control and reduction in severe hypoglycemia with the use of voice-enabled CGM (Dexcom G6) in legally blind patients with intensive insulin-treated diabetes. The retrospective study design, single-center, and small sample size with one patient with chronic kidney disease are major limitations. Future studies are needed to understand human factors and quality of life in addition to glycemic efficacy and safety of voice-enabled CGM in legally blind patients with diabetes.

In conclusion, voice-enabled CGM improved glycemic control and decreased severe hypoglycemia in legally blind patients with diabetes. Providers caring for these patients should be aware of the availability of voice-enabled CGM systems.

Authors' Contributions

H.K.A. and V.N.S. designed and conducted the study. J.S.B. conducted statistical analysis. H.K.A. and V.N.S. prepared the article. All authors reviewed the article and edited it for intellectual content and gave final approval for this version to be published. H.K.A. 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.

Author Disclosure Statement

H.K.A. received research grant through University of Colorado from Eli Lilly, Senseonics, REMD, and Mannkind. V.N.S. received research grant through University of Colorado from Sanofi US, Dexcom, Inc., Novo Nordisk, vTv Therapeutics, Mylan GmbH, and National Institute of Health (NIAMS, NIDDK). V.N.S. received honoraria through University of Colorado from Sanofi, Dexcom, and Medscape for work as a speaker/advisory board member. J.S.B. has no conflict of interest.

Funding Information

No funding was received for this article.