Search

Exercise increases nocturnal hypoglycemia risk which may impact sleep in T1D. Adults with T1D (n=25; 19 CSII; age 33±14yrs; HbA1c 7.6±1.3%; T1D duration 15yrs [IQR: 10, 22]) were divided into 3 exercise groups: aerobic, interval, or resistance, performed twice/week using instructional videos for 4 weeks. Participants used CGM, and wore an activity/sleep monitor (Garmin Vivosmart 3). Sleep duration was lower on nights following interval exercise compared to nights following sedentary days (p=0.04, Figure 1a). Participants experienced hypoglycemia earlier in the night following interval or aerobic exercise, compared to sedentary or resistance exercise (Figure 1b). Exercise types differentially impact sleep loss and nocturnal hypoglycemia. Disclosure: P.G. Jacobs: Consultant; Self; SFC Fluidics. Research Support; Self; Dexcom, Inc. Stock/Shareholder; Self; Pacific Diabetes Technologies. Other Relationship; Self; AgaMatrix. Z. Li: None. P. Calhoun: Stock/Shareholder; Self; Dexcom, Inc. R.L. Gal: None. R. Beck: None. J.R. Castle: Advisory Panel; Self; Novo Nordisk Inc. Consultant; Self; ADOCIA, Dexcom, Inc., Zealand Pharma A/S. Other Relationship; Self; Pacific Diabetes Technologies, Roche Diabetes Care. M.A. Clements: Consultant; Self; Glooko, Inc. Other Relationship; Self; Glooko, Inc. E. Dassau: Consultant; Self; Eli Lilly and Company. Research Support; Self; Dexcom, Inc., DreaMed Diabetes, Tandem Diabetes Care, Xeris Pharmaceuticals, Inc. Speaker's Bureau; Self; Roche Diabetes Care. Other Relationship; Self; Dexcom, Inc., Insulet Corporation, Roche Diabetes Care. F.J. Doyle: Research Support; Self; DreaMed Diabetes, Tandem Diabetes Care, Xeris Pharmaceuticals, Inc. Stock/Shareholder; Self; Mode AGC. Other Relationship; Self; Dexcom, Inc., Insulet Corporation, Roche Diabetes Care. M.B. Gillingham: None. C.K. Martin: Advisory Panel; Self; EHE Health, NaturallySlim. Consultant; Self; Florida Hospital, Gila Therapeutics, OpenFit, Zafgen, Inc. Research Support; Self; Egg Board, Helmsley Trust, IDEA Public Schools, National Institutes of Health, NIHR, Patient-Centered Outcomes Research Institute, USDA, Weight Watchers International, Inc. Other Relationship; Self; Academy of Nutrition and Dietetics, Louisiana State Univ./Pennington Biomedical Research Center. M.R. Rickels: Consultant; Self; Semma Therapeutics, Inc. Research Support; Self; Xeris Pharmaceuticals, Inc. S.R. Patton: None. M. Riddell: Advisory Panel; Self; Zucara Therapeutics Inc. Consultant; Self; Lilly Diabetes. Research Support; Self; Dexcom, Inc., Insulet Corporation. Speaker's Bureau; Self; Medtronic, Novo Nordisk A/S. Funding: The Leona M. and Harry B. Helmsley Charitable Trust (T1-DEXI Main Study) [ABSTRACT FROM AUTHOR]

Introduction: Glycemic control during exercise is challenging for people with type 1 diabetes. Better quantification of insulin dynamics at different exercise intensities and across insulin loading conditions is necessary for accurate insulin dosing during exercise. Methods: Twenty seven participants with type 1 diabetes were evaluated during 3 separate 10-hour fasting aerobic exercise studies with low, medium, and high levels of insulin infusion (100%, 150%, and 300% of basal rate) across two exercise arms: moderate (40-45% of VO2 max), and intense (60-65% of VO2 max) exercise. Glucose levels were clamped during a 3-hour run-in period, followed by 45 minutes of exercise on a treadmill. 6,6-2H2-dideuterated glucose was infused to match endogenous glucose production during the entire trial. Glucose tracer data were fitted to a two-compartment ODE model, solved via Bayesian inference. We intend to quantify insulin and non-insulin mediated disposal of glucose by observing the rate of disposal (Rd) across different rates of insulin infusion in each exercise arm. Results: Area under the curve (AUC) of Rd during 45 minutes of moderate exercise increased by 0.084 mmol/L for every percent increase in baseline insulin infusion rate (95% CI=0.064-0.104, p<0.001). AUC during intense exercise increased by 0.109 mmol/L (95% CI= 0.079-0.139, p<0.001). Mixed-effects model analysis showed no difference between exercise arms. Conclusion: Results indicate Rd is substantial during exercise but that the difference in glucose uptake during aerobic exercise at 40-45% VO2 max may be comparable with 60-65% VO2 max. Disclosure: T.P. Nguyen: None. P.G. Jacobs: Stock/Shareholder; Self; Pacific Diabetes Technologies. J.R. Castle: Advisory Panel; Self; Novo Nordisk Inc., Zealand Pharma A/S. Consultant; Self; Dexcom, Inc. Research Support; Self; Dexcom, Inc., Xeris Pharmaceuticals, Inc. L.M. Wilson: None. D. Branigan: None. V. Gabo: None. B. Senf: None. F.H. Guillot: None. J. El Youssef: None. Funding: National Institutes of Health [ABSTRACT FROM AUTHOR]

Dual-hormone systems show promise to reduce hypoglycemia, but require a stable liquid glucagon formulation. XeriSol™ glucagon is a shelf-stable glucagon product for this use. Nineteen subjects with T1D are completing a single-center three-day outpatient study comparing control modes of the Oregon Artificial Pancreas system: 1) dual hormone (DH) closed-loop with Novolog™ insulin and XeriSol™ glucagon, 2) insulin-only single hormone (SH) closed-loop, 3) insulin-only predictive low glucose suspend system (PLGS). In clinic aerobic exercise (45 minutes at 60% VO2max) and home exercise was completed with each arm. SH and DH used automated exercise detection to adaptively dose insulin/glucagon for predicted hypoglycemia and tailored mealtime dosing based on past meal responses with an adaptive algorithm. An interim analysis after 7 subjects was completed for safety. The primary outcome measures are % time 70-180 mg/dL and % time <70 mg/dL for the 3-day study and the 4-hour period from start of exercise until the next meal. From start of exercise until the next meal, DH showed a clinically meaningful reduction of time in hypoglycemia compared with PLGS and SH. For the entire study, DH showed a clinically meaningful lower time <70mg/dL. No serious events or unexpected side effects occurred. Disclosure: L.M. Wilson: None. P.G. Jacobs: Stock/Shareholder; Self; Pacific Diabetes Technologies. N. Resalat: None. R. Reddy: None. J. El Youssef: None. D. Branigan: None. J.A. Leitschuh: Other Relationship; Self; AgaMatrix. B. Senf: None. V. Gabo: None. J.R. Castle: Advisory Panel; Self; Novo Nordisk Inc., Zealand Pharma A/S. Consultant; Self; Dexcom, Inc. Research Support; Self; Dexcom, Inc., Xeris Pharmaceuticals, Inc. Funding: JDRF [ABSTRACT FROM AUTHOR]

Background: Using continuous glucose monitoring (CGM), we investigated potential sex-related differences in glycemia during exercise in persons living with type 1 diabetes (T1D) or without diabetes. Methods: Participants used CGM (Dexcom G6), exercise/meal diaries, or an activity monitor with a meal capture smart phone app, to help identify home based exercise and meal events over a 10 day to 4 wk period. Results: The analyses included 40 T1D (age: 36±15 yrs; BMI: 26.3±3.1 kg/m2; HbA1c: 7.5±1.5%; 35% female; mean±SD) and 120 persons without diabetes (age: 30±20 years; BMI: 24.4±3.2 kg/m2; HbA1c: 5.1±0.3; 64% female) and 283 and 619 exercise events, respectively. The frequency distribution of all CGM values were comparable between males and females overall, but values had a wider distribution among T1D as compared to non-T1D (Fig). Hyperglycemia (> 180 mg/dL) represented ~0.1% of the exercise time in participants without T1D vs. 31% and 24% of the time in the female and male participants without diabetes, respectively. Hypoglycemic values (<70 mg/dL) represented 7.1% and 3.0% of the time in the non-T1D females and males, respectively (P<0.001), vs. 4.4% and 5.0% in the T1D females and males (P=0.67), respectively. Conclusions: Based on CGM, mild hypoglycemic exposure during exercise is surprisingly common in individuals with and without diabetes and may be particularly prevalent in females without diabetes. Disclosure: M. Riddell: Advisory Panel; Self; Zealand Pharma A/S. Consultant; Self; Lilly Diabetes. Research Support; Self; Dexcom, Inc., Insulet Corporation. Speaker's Bureau; Self; Novo Nordisk Inc., Sanofi. Stock/Shareholder; Self; Zucara Therapeutics Inc. Z. Li: None. N. DSouza: None. C. Yeung: None. D. Kesibi: None. S. R. Patton: None. R. Beck: Consultant; Self; Bigfoot Biomedical, Inc., Diasome Pharmaceuticals, Inc., Insulet Corporation, Lilly Diabetes, vTv Therapeutics. Research Support; Self; Beta Bionics, Inc., Dexcom, Inc., Medtronic, Novo Nordisk, Tandem Diabetes Care. P. G. Jacobs: Board Member; Self; Pacific Diabetes Technologies. Research Support; Self; Dexcom, Inc. M. A. Clements: Consultant; Self; Eli Lilly and Company. Employee; Self; Glooko, Inc. Research Support; Self; Abbott Diabetes, Dexcom, Inc. R. L. Gal: None. F. J. Doyle: Advisory Panel; Self; Mode AGC. Other Relationship; Self; Dexcom, Inc., Insulet Corporation, Roche Diabetes Care. C. K. Martin: Advisory Panel; Self; EHE Health. Board Member; Self; NaturallySlim. Research Support; Self; American Society for Nutrition, Leona M. and Harry B. Helmsley Charitable Trust, Lilly, National Institutes of Health, Patient-Centered Outcomes Research Institute, U. S. Department of Agriculture, WW. Other Relationship; Self; ABGIL, Academy of Nutrition and Dietetics. P. Calhoun: None. J. Sherr: Advisory Panel; Self; Cecelia Health, Insulet Corporation, Medtronic. Consultant; Self; Insulet Corporation, Lexicon Pharmaceuticals, Inc., Lilly Diabetes, Medtronic. Research Support; Self; Dexcom, Inc., Insulet Corporation, Medtronic. Speaker's Bureau; Self; Lilly Diabetes. J. R. Castle: Advisory Panel; Self; Novo Nordisk, Zealand Pharma A/S. Consultant; Self; ADOCIA. Research Support; Self; Dexcom, Inc. Stock/Shareholder; Spouse/Partner; Insulet Corporation. Stock/Shareholder; Self; Pacific Diabetes Technologies. M. R. Rickels: Advisory Panel; Self; Semma Therapeutics, Sernova, Corp. Research Support; Self; Xeris Pharmaceuticals, Inc. [ABSTRACT FROM AUTHOR]

While automated insulin delivery (AID) systems are now commercially available, over 40 percent of people with type 1 diabetes manage their insulin with multiple daily injection therapy (MDI). AID systems can improve time-in-range in adults, however it is not clear if optimal MDI therapy can achieve the same glycemic targets. We have previously shown that a K-nearest-neighbors decision support system (KNN-DSS) for MDI can deliver safe recommendations and achieves high agreement with endocrinologists. Here, we report glycemic outcomes as we compare KNN-DSS with two artificial pancreas systems: fading-memory-proportional-derivative (FMPD), and model-predictive control (MPC). We utilized a platform accepted by U.S. Food and Drug Administration for the evaluation of artificial pancreas technologies, the UVA-Padova simulator. In silico, 100 adults with varying insulin sensitivities, weight, and insulin dosing behaviors underwent a 3-arm study comparing KNN-DSS, FMPD, and MPC algorithms. Each UVA-Padova subject received a unique study meal pattern that was repeated across study arms. The two primary outcome measures were % time-in-range and % time-in-hypoglycemia. Secondary outcome measures included LGBI and HGBI risk index. The KNN-DSS was able to achieve 79.6 ± 15.4 %time-in-range after 12 weeks compared with 83.6 ± 8.3% for the MPC and 84.2 ± 9.24% for the FMPD. The KNN-DSS achieved comparable time below 70 mg/dL (0.79%, [0 1.6]) with the MPC-AP (0.62%, [0 1.8]) and FMPD (0%, [0 0.72]). The KNN-DSS also achieved comparable LGBI (2.55) with the MPC-AP (2.68). This simulation assumes perfect adherence to insulin dosing regimens, and while % time-in-range may be higher than expected, the results indicate that decision support in MDI may yield comparable glycemic outcomes with automated insulin delivery systems if people are compliant. Results of this study will be used to guide the design of an upcoming human clinical trial. Disclosure: N.S. Tyler: None. W. Hilts: None. C.M. Mosquera-Lopez: None. L.M. Wilson: None. R. Dodier: None. J. El Youssef: None. J.R. Castle: Advisory Panel; Self; Novo Nordisk Inc. Consultant; Self; ADOCIA, Dexcom, Inc., Zealand Pharma A/S. Other Relationship; Self; Pacific Diabetes Technologies, Roche Diabetes Care. P.G. Jacobs: Consultant; Self; SFC Fluidics. Research Support; Self; Dexcom, Inc. Stock/Shareholder; Self; Pacific Diabetes Technologies. Other Relationship; Self; AgaMatrix. Funding: The Leona M. and Harry B. Helmsley Charitable Trust (2018PG-T1D001) [ABSTRACT FROM AUTHOR]

People with type 1 diabetes (T1D) have difficulty controlling glucose during exercise. Different types, intensities, and durations of exercise impact glucose differently. There is currently no algorithm that can accurately predict glucose changes during aerobic, resistance, interval, and free-living exercise. A support vector regression (SVR) model was designed to predict the maximum change in glucose during exercise. The model was trained and validated on 30-minute exercise sessions collected twice weekly over 4 weeks from 33 people with T1D (age 33±13 years, BMI 26.3±2.9 kg/m2, 18±12 years since diagnosis). Participants completed study assigned aerobic (n = 40), resistance (n= 33), or interval (n=20) exercise sessions in addition to their typical exercise regimen (n = 254). Participants wore a Dexcom (G5 or G6) or Medtronic CGM and used their own insulin pen (n = 10) or pump to administer insulin. The exercise information and food consumed was reported using a custom app. Heart rate, accelerometry, and sleep metrics were acquired using a Garmin vivosmart watch. Training was done on 90% of the observations while 10% of data were used for testing using 10-fold cross validation. The mean absolute error (MAE) of the model-predicted change in glucose during exercise was 23.59 mg/dL. The SVR algorithm may be useful within a decision support tool to help people with T1D better manage their glucose levels during exercise. Disclosure: G. Young: None. Z. Li: None. P. Calhoun: Stock/Shareholder; Self; Dexcom, Inc. R.L. Gal: None. R. Beck: None. J.R. Castle: Advisory Panel; Self; Novo Nordisk Inc. Consultant; Self; ADOCIA, Dexcom, Inc., Zealand Pharma A/S. Other Relationship; Self; Pacific Diabetes Technologies, Roche Diabetes Care. M.A. Clements: Consultant; Self; Glooko, Inc. Other Relationship; Self; Glooko, Inc. E. Dassau: Consultant; Self; Eli Lilly and Company. Research Support; Self; Dexcom, Inc., DreaMed Diabetes, Tandem Diabetes Care, Xeris Pharmaceuticals, Inc. Speaker's Bureau; Self; Roche Diabetes Care. Other Relationship; Self; Dexcom, Inc., Insulet Corporation, Roche Diabetes Care. F.J. Doyle: Research Support; Self; DreaMed Diabetes, Tandem Diabetes Care, Xeris Pharmaceuticals, Inc. Stock/Shareholder; Self; Mode AGC. Other Relationship; Self; Dexcom, Inc., Insulet Corporation, Roche Diabetes Care. M.B. Gillingham: None. C.K. Martin: Advisory Panel; Self; EHE Health, NaturallySlim. Consultant; Self; Florida Hospital, Gila Therapeutics, OpenFit, Zafgen, Inc. Research Support; Self; Egg Board, Helmsley Trust, IDEA Public Schools, National Institutes of Health, NIHR, Patient-Centered Outcomes Research Institute, USDA, Weight Watchers International, Inc. Other Relationship; Self; Academy of Nutrition and Dietetics, Louisiana State Univ./Pennington Biomedical Research Center. M.R. Rickels: Consultant; Self; Semma Therapeutics, Inc. Research Support; Self; Xeris Pharmaceuticals, Inc. S.R. Patton: None. M. Riddell: Advisory Panel; Self; Zucara Therapeutics Inc. Consultant; Self; Lilly Diabetes. Research Support; Self; Dexcom, Inc., Insulet Corporation. Speaker's Bureau; Self; Medtronic, Novo Nordisk A/S. P.G. Jacobs: Consultant; Self; SFC Fluidics. Research Support; Self; Dexcom, Inc. Stock/Shareholder; Self; Pacific Diabetes Technologies. Other Relationship; Self; AgaMatrix. Funding: Exercise in Diabetes Initiative (T1-DEXI Main Study) [ABSTRACT FROM AUTHOR]

Structured exercise can help patients with type 1 diabetes (T1D) achieve physical activity targets, however the effects of exercise type on percent time in range (TIR) and glucose variability are unclear. The Type 1 Diabetes Exercise Initiative Pilot Study provided the opportunity to examine glycemic control on structured exercise days compared to sedentary days. Adults with T1D ([mean ± SD] age= 32 ± 14 years; HbA1C= 7.8 ± 1.6%), using a continuous glucose monitor, were assigned to complete twice weekly aerobic, interval or resistance exercise sessions, using study-developed instructional videos for a period of 4 weeks. TIR was higher (Figure 1) and mean glucose concentration was lower overall on exercise days compared with sedentary days (160 ± 52 vs. 176 ± 49 mg/dL, P<0.001). Glucose variability was similar between exercise and sedentary days (coefficient of variation: 36 ± 12% vs. 37% ± 12%). In summary, participants had lower 24-hour mean glucose levels and a greater percent time in range on active days compared with inactive days. Disclosure: M. Riddell: Advisory Panel; Self; Xeris Pharmaceuticals, Inc. Research Support; Self; Dexcom, Inc. Speaker's Bureau; Self; Insulet Corporation, Medtronic MiniMed, Inc. Stock/Shareholder; Self; Zucara Therapeutics Inc. Z. Li: None. R. Beck: Other Relationship; Self; Abbott Laboratories, Ascensia Diabetes Care, Bigfoot Biomedical, Dexcom, Inc., Insulet Corporation, Lilly Diabetes, Roche Diabetes Care, Tandem Diabetes Care. R.L. Gal: None. P.G. Jacobs: Stock/Shareholder; Self; Pacific Diabetes Technologies. J.R. Castle: Advisory Panel; Self; Novo Nordisk Inc., Zealand Pharma A/S. Consultant; Self; Dexcom, Inc. Research Support; Self; Dexcom, Inc., Xeris Pharmaceuticals, Inc. M.B. Gillingham: None. M.A. Clements: Advisory Panel; Self; Glooko, Inc. Consultant; Self; Eli Lilly and Company. Speaker's Bureau; Self; Medtronic. S.R. Patton: None. E. Dassau: Consultant; Self; Eli Lilly and Company, Insulet Corporation. Research Support; Self; Dexcom, Inc., DreaMed Diabetes, Ltd., Insulet Corporation, Roche Diabetes Care, Tandem Diabetes Care, Xeris Pharmaceuticals, Inc. Speaker's Bureau; Self; Roche Diabetes Care. Other Relationship; Self; ModAGC. F.J. Doyle: Consultant; Self; ModeAGC. Other Relationship; Self; Insulet Corporation. C.K. Martin: Advisory Panel; Self; EHE. Consultant; Self; ACAP Health, Florida Hospital, WW, Zafgen, Inc. Other Relationship; Self; Academy of Nutrition and Dietetics. P. Calhoun: Stock/Shareholder; Self; Dexcom, Inc. M.R. Rickels: None. Funding: The Leona M. and Harry B. Helmsley Charitable Trust [ABSTRACT FROM AUTHOR]

People with type 1 diabetes (T1D) estimate the carbohydrate (CHO) content of meals to enable accurate insulin dosing, yet protein and fat content of meals also influences post-prandial glycemia. As part of an observational study examining the impact of exercise and nutrient consumption on glycemia, we examined accuracy in estimating macronutrient content of free-living meals via a novel phone app. Participant estimates of nutrient content were compared to expert nutrition analyses performed via the Remote Food Photography Method (RFPM). We report results from 30 of 33 randomized with analyzable food photos. Participants were 15-65 years (32±14 years); 27% identified as female. Participants were asked to take photos before and after meals/snacks on up to 16 days over a 28-day period, enter CHO estimates and estimate if meals were low (<13%), typical (13-<18%), or high (≥18%) protein and if meals were low (<26%), medium (26-<32%), or high (≥32%) fat. The phone app plus RFPM captured 92±27% of estimated energy needs. Of 1,292 food photos analyzed, 429 contained < 25 g (small), 641 contained 25-75 g (medium), and 222 contained > 75 g (large) amounts of CHO. Participants estimated CHO in small or medium meals within 10 g of the expert analyses. They were less accurate estimating CHO for larger meals (-56 ± 46 g). Likewise, most correctly categorized low and typical protein (63%, 50%) as well as low and typical fat (62%, 67%) meals. Few correctly categorized meals high in protein (17%) or fat (16%). Participants' estimation accuracy for larger meals did not differ by T1D duration. The phone app successfully collected individuals' food intake and estimated macronutrient intake, but showed that participants consistently underestimated nutrient intake for large meals. Accurate estimation of total macronutrients in meals could be leveraged to improve insulin decision support tools and closed loop systems; development of tools to improve macronutrient estimation skills should be considered. Disclosure: M.B. Gillingham: None. C.K. Martin: Advisory Panel; Self; EHE. Consultant; Self; ACAP Health, Florida Hospital, WW, Zafgen, Inc. Other Relationship; Self; Academy of Nutrition and Dietetics. S.R. Patton: None. Z. Li: None. P.G. Jacobs: Stock/Shareholder; Self; Pacific Diabetes Technologies. M. Riddell: Advisory Panel; Self; Xeris Pharmaceuticals, Inc. Research Support; Self; Dexcom, Inc. Speaker's Bureau; Self; Insulet Corporation, Medtronic MiniMed, Inc. Stock/Shareholder; Self; Zucara Therapeutics Inc. M.R. Rickels: None. J.R. Castle: Advisory Panel; Self; Novo Nordisk Inc., Zealand Pharma A/S. Consultant; Self; Dexcom, Inc. Research Support; Self; Dexcom, Inc., Xeris Pharmaceuticals, Inc. M.A. Clements: Advisory Panel; Self; Glooko, Inc. Consultant; Self; Eli Lilly and Company. Speaker's Bureau; Self; Medtronic. E. Dassau: Consultant; Self; Eli Lilly and Company, Insulet Corporation. Research Support; Self; Dexcom, Inc., DreaMed Diabetes, Ltd., Insulet Corporation, Roche Diabetes Care, Tandem Diabetes Care, Xeris Pharmaceuticals, Inc. Speaker's Bureau; Self; Roche Diabetes Care. Other Relationship; Self; ModAGC. F.J. Doyle: Consultant; Self; ModeAGC. Other Relationship; Self; Insulet Corporation. P. Calhoun: Stock/Shareholder; Self; Dexcom, Inc. R.L. Gal: None. R. Beck: Other Relationship; Self; Abbott Laboratories, Ascensia Diabetes Care, Bigfoot Biomedical, Dexcom, Inc., Insulet Corporation, Lilly Diabetes, Roche Diabetes Care, Tandem Diabetes Care. Funding: The Leona M. and Harry B. Helmsley Charitable Trust [ABSTRACT FROM AUTHOR]

Most people with type 1 diabetes use multiple daily injections (MDI) therapy, yet there are limited decision support tools for this population. We designed and evaluated a K-nearest-neighbors decision support system (KNN-DSS) that utilizes continuous glucose data and Bluetooth-enabled insulin dose capture devices to detect problematic glycemic patterns and recommend insulin dosage adjustments to maximize time-in-range (70-180 mg/dL) and minimize hypoglycemia (< 70 mg/dL). The KNN-DSS employs distance and class weighting, a carb ratio aggressiveness algorithm, and an expert-opinion quality-control algorithm to filter erroneous recommendations. The KNN-DSS was evaluated both in silico and in comparison with physician recommendations. In silico virtual patients (n=30) of varying weight, total daily insulin requirement, insulin sensitivity, and compliance with care were evaluated in a 1-year study whereby they received weekly recommendations. The primary outcome measures were the % time-in-range and % time-in-hypoglycemia. The KNN-DSS was further evaluated for accuracy compared with endocrinologist recommendations collected during a 4-week, at-home clinical trial of CGM-augmented MDI therapy (N = 12, 7 female, 30.8 ± 5 years, 79.6 ± 18.9 kg), and compared using Dice similarity coefficient. In silico, the KNN-DSS improved time-in-range from 66.8% at baseline to 81.0% (21% relative improvement, p <0.005), and reduced time-in-hypoglycemia from 3.14% at baseline to 1.37% (56% relative improvement, p < 0.005) after one year. Compared to endocrinologists, the KNN-DSS recommendations to real-world patients were within acceptable agreement 75% of the time, 6% contraindicated, and 19% whereby KNN-DSS gave a recommendation when the physician gave no recommendation. KNN-DSS improves glycemic outcomes in silico and exhibits acceptable agreement with physician recommendations. Future clinical studies are planned. Disclosure: N.S. Tyler: None. C.M. Mosquera-Lopez: None. L.M. Wilson: None. N. Resalat: None. R. Dodier: None. D. Branigan: None. V. Gabo: None. J. El Youssef: None. J.R. Castle: Advisory Panel; Self; Novo Nordisk Inc., Zealand Pharma A/S. Consultant; Self; Dexcom, Inc. Research Support; Self; Dexcom, Inc., Xeris Pharmaceuticals, Inc. P.G. Jacobs: Stock/Shareholder; Self; Pacific Diabetes Technologies. Funding: The Leona M. and Harry B. Helmsley Charitable Trust (2018PG-T1D001) [ABSTRACT FROM AUTHOR]