Multicenter, Randomized Trial of a Bionic Pancreas in Type 1 Diabetes.

BACKGROUND: Currently available semiautomated insulin-delivery systems require individualized insulin regimens for the initialization of therapy and meal doses based on carbohydrate counting for routine operation. In contrast, the bionic pancreas is initialized only on the basis of body weight, makes all dose decisions and delivers insulin autonomously, and uses meal announcements without carbohydrate counting. METHODS: In this 13-week, multicenter, randomized trial, we randomly assigned in a 2:1 ratio persons at least 6 years of age with type 1 diabetes either to receive bionic pancreas treatment with insulin aspart or insulin lispro or to receive standard care (defined as any insulin-delivery method with unblinded, real-time continuous glucose monitoring). The primary outcome was the glycated hemoglobin level at 13 weeks. The key secondary outcome was the percentage of time that the glucose level as assessed by continuous glucose monitoring was below 54 mg per deciliter; the prespecified noninferiority limit for this outcome was 1 percentage point. Safety was also assessed. RESULTS: A total of 219 participants 6 to 79 years of age were assigned to the bionic-pancreas group, and 107 to the standard-care group. The glycated hemoglobin level decreased from 7.9% to 7.3% in the bionic-pancreas group and did not change (was at 7.7% at both time points) in the standard-care group (mean adjusted difference at 13 weeks, -0.5 percentage points; 95% confidence interval [CI], -0.6 to -0.3; P<0.001). The percentage of time that the glucose level as assessed by continuous glucose monitoring was below 54 mg per deciliter did not differ significantly between the two groups (13-week adjusted difference, 0.0 percentage points; 95% CI, -0.1 to 0.04; P<0.001 for noninferiority). The rate of severe hypoglycemia was 17.7 events per 100 participant-years in the bionic-pancreas group and 10.8 events per 100 participant-years in the standard-care group (P = 0.39). No episodes of diabetic ketoacidosis occurred in either group. CONCLUSIONS: In this 13-week, randomized trial involving adults and children with type 1 diabetes, use of a bionic pancreas was associated with a greater reduction than standard care in the glycated hemoglobin level. (Funded by the National Institute of Diabetes and Digestive and Kidney Diseases and others; ClinicalTrials.gov number, NCT04200313.).

Strength training is more effective than aerobic exercise for improving glycaemic control and body composition in people with normal-weight type 2 diabetes: a randomised controlled trial.

AIMS/HYPOTHESIS: Type 2 diabetes in people in the healthy weight BMI category (<25 kg/m2), herein defined as 'normal-weight type 2 diabetes', is associated with sarcopenia (low muscle mass). Given this unique body composition, the optimal exercise regimen for this population is unknown. METHODS: We conducted a parallel-group RCT in individuals with type 2 diabetes (age 18-80 years, HbA1c 47.5-118.56 mmol/mol [6.5-13.0%]) and BMI <25 kg/m2). Participants were recruited in outpatient clinics or through advertisements and randomly assigned to a 9 month exercise programme of strength training alone (ST), aerobic training alone (AER) or both interventions combined (COMB). We used stratified block randomisation with a randomly selected block size. Researchers and caregivers were blinded to participants' treatment group; however, participants themselves were not. Exercise interventions were conducted at community-based fitness centres. The primary outcome was absolute change in HbA1c level within and across the three groups at 3, 6 and 9 months. Secondary outcomes included changes in body composition at 9 months. Per adherence to recommended exercise protocol (PP) analysis included participants who completed at least 50% of the sessions. RESULTS: Among 186 individuals (ST, n=63; AER, n=58; COMB, n=65) analysed, the median (IQR) age was 59 (53-66) years, 60% were men and 83% were Asian. The mean (SD) HbA1c level at baseline was 59.6 (13.1) mmol/mol (7.6% [1.2%]). In intention-to-treat analysis, the ST group showed a significant decrease in HbA1c levels (mean [95% CI] -0.44 percentage points [-0.78, -0.12], p=0.002), while no significant change was observed in either the COMB group (-0.35 percentage points, p=0.13) or the AER group (-0.24 percentage points, p=0.10). The ST group had a greater improvement in HbA1c levels than the AER group (p=0.01). Appendicular lean mass relative to fat mass increased only in the ST group (p=0.0008), which was an independent predictor of HbA1c change (beta coefficient -7.16, p=0.01). Similar results were observed in PP analysis. Only one adverse event, in the COMB group, was considered to be possibly associated with the exercise intervention. CONCLUSIONS/INTERPRETATION: In normal-weight type 2 diabetes, strength training was superior to aerobic training alone, while no significant difference was observed between strength training and combination training for HbA1c reduction. Increased lean mass relative to decreased fat mass was an independent predictor of reduction in HbA1c level. TRIAL REGISTRATION: ClinicalTrials.gov NCT02448498. FUNDING: This study was funded by the National Institutes of Health (NIH; R01DK081371).

Implementation of Psychosocial Screening into Diabetes Clinics: Experience from the Type 1 Diabetes Exchange Quality Improvement Network.

PURPOSE OF REVIEW: Although advances in diabetes technology and pharmacology have significantly and positively impacted diabetes management and health outcomes for some, diabetes care remains burdensome and can be challenging to balance with other life priorities. The purpose of this article is to review the rationale for assessment of psychosocial domains in diabetes care settings and strategies for the implementation of psychosocial screening into routine practice. Survey data from the Type 1 Diabetes Exchange Quality Improvement Network is highlighted. RECENT FINDINGS: Implementation of psychosocial screening requires identifying the population; selecting validated tools to assess target domains; determining frequency of screening and mode of survey delivery; and scoring, interpreting, documenting, and facilitating referrals such that these processes are part of clinical workflows. Recognizing the influence of psychosocial factors for people with diabetes (PWD), professional society guidelines for comprehensive diabetes care recommend the integration of psychosocial screening into routine care.

High-parametric evaluation of human invariant natural killer T cells to delineate heterogeneity in allo- and autoimmunity.

Human invariant natural killer T (iNKT) cells are a rare innate-like lymphocyte population that recognizes glycolipids presented on CD1d. Studies in mice have shown that these cells are heterogeneous and are capable of enacting diverse functions, and the composition of iNKT cell subsets can alter disease outcomes. In contrast, far less is known about how heterogeneity in human iNKT cells relates to disease. To address this, we used a high-dimensional, data-driven approach to devise a framework for parsing human iNKT heterogeneity. Our data revealed novel and previously described iNKT cell phenotypes with distinct functions. In particular, we found 2 phenotypes of interest: (1) a population with T helper 1 function that was increased with iNKT activation characterized by HLA-II+CD161- expression, and (2) a population with enhanced cytotoxic function characterized by CD4-CD94+ expression. These populations correlate with acute graft-versus-host disease after allogeneic hematopoietic stem cell transplantation and with new onset type 1 diabetes, respectively. Our study identifies human iNKT cell phenotypes associated with human disease that could aid in the development of biomarkers or therapeutics targeting iNKT cells.

Facilitators and Barriers to Smart Insulin Pen Use: A Mixed-Method Study of Multidisciplinary Stakeholders From Diabetes Teams in the United States.

This study sought to identify barriers and facilitators to successful smart insulin pen (SIP) use and gauge prescribing practices and integration into clinical practice by assessing provider and care team perspectives at participating endocrinology clinics within the T1D Exchange Quality Improvement Collaborative. The identified provider-related, patient-related, and clinic- and operational-level barriers and facilitators varied based on clinic knowledge, capacity, and resources. High-impact barriers included insurance coverage and prescribing processes; high-impact facilitators included improved diabetes clinic visit quality and use of SIPs as an alternative to insulin pump therapy. Findings indicated the need for provider and care team education and training on proper SIP features, use, and prescribing.

Baseline Quality Improvement Capacity of 33 Endocrinology Centers Participating in the T1D Exchange Quality Improvement Collaborative.

This article describes the evolution of the Type 1 Diabetes Exchange Quality Improvement Collaborative (T1DX-QI) and provides insight into the development and growth of a successful type 1 diabetes quality improvement (QI) program. Since its inception 8 years ago, the collaborative has expanded to include centers across the United States with varying levels of QI experience, while simultaneously achieving many tangible improvements in type 1 diabetes care. These successes underscore the importance of learning health systems, data-sharing, benchmarking, and peer collaboration as drivers for continuous QI. Future efforts will include recruiting additional small- to medium-sized centers focused on adult care and underserved communities to further the goal of improving care and outcomes for all people living with type 1 diabetes.

Impact of diabetes mellitus on clinical outcomes after heart transplantation.

PURPOSE: Diabetes mellitus (DM) is common among recipients of heart transplantation (HTx) but its impact on clinical outcomes is unclear. We evaluated the associations between pretransplant DM and posttransplant DM (PTDM) and outcomes among adults receiving HTx at a single center. METHODS: We performed a retrospective study (range 01/2008 - 07/2018), n = 244. The primary outcome was survival; secondary outcomes included acute rejection, cardiac allograft vasculopathy, infection requiring hospitalization, macrovascular events, and dialysis initiation post-transplant. Comparisons were performed using Kaplan-Meier and multivariable Cox regression analyses. RESULTS: Pretransplant DM was present in 75 (30.7%) patients and was associated with a higher risk for infection requiring hospitalization (p < 0.05), but not with survival or other outcomes. Among the 144 patients without pretransplant DM surviving to 1 year, 29 (20.1%) were diagnosed with PTDM at the 1-year follow-up. After multivariable adjustment, PTDM diagnosis at 1-year remained associated with worse subsequent survival (hazard ratio 2.72, 95% confidence interval 1.03-7.16). Predictors of PTDM at 1-year included cytomegalovirus seropositivity and higher prednisone dose (> 5 mg/day) at 1-year follow-up. CONCLUSIONS: Compared to HTx recipients without baseline DM, those with baseline DM have a higher risk for infections requiring hospitalization, and those who develop DM after HTx have worse survival.

RESULTS OF A 24-WEEK TRIAL OF TECHNOSPHERE INSULIN VERSUS INSULIN ASPART IN TYPE 2 DIABETES.

OBJECTIVE: To compare glycemic efficacy of Technosphere insulin (TI) versus that of insulin aspart (IA), each added to basal insulin, in type 2 diabetes. METHODS: This randomized, 24-week trial included subjects aged from 18 to 80 years who were treated with subcutaneous insulin for 3 months and had glycated hemoglobin (HbA1C) levels of 7.0% to 11.5%. After receiving stabilized insulin glargine doses during a 4-week lead in, the subjects were randomized to TI or IA. The primary end point was an HbA1C change from baseline, with the differences analyzed by equivalence analyses. RESULTS: In the overall cohort (N = 309; males, 23.3%), mean (SD) age was 58.5 (8.4) years, body mass index was 30.8 (4.7) kg/m2, weight was 82.2 (13.6) kg, and duration of diabetes was 12.2 (7.1) years. An intention-to-treat cohort had 150 subjects randomized to TI (mean [SD] HbA1C: 8.9% [1.1%]) and 154 randomized to IA (mean [SD] HbA1C: 9.0% [1.3%]). At 24 weeks, mean (SD) HbA1C value declined to 7.9% (1.3%) and 7.7% (1.1%) in the TI and IA cohorts, respectively. A treatment difference of 0.26% was not statistically significant, but the predefined equivalency margin was not met. Subjects receiving TI lost 0.78 kg compared to baseline; subjects receiving IA gained 0.23 kg (P =.0007). The incidence of mild/moderate hypoglycemia was lower for the TI cohort, though not statistically significant. CONCLUSION: Both TI and IA resulted in significant and clinically meaningful HbA1C reductions. TI also resulted in significant and clinically meaningful weight reductions. These data support the use of inhaled insulin as a treatment option for individuals with type 2 diabetes.

Inequities in Health Outcomes in Children and Adults With Type 1 Diabetes: Data From the T1D Exchange Quality Improvement Collaborative.

Health care inequities among racial and ethnic groups remain prevalent. For people with type 1 diabetes who require increased medical access and care, disparities are seen in access to care and health outcomes. This article reports on a study by the T1D Exchange Quality Improvement Collaborative evaluating differences in A1C, diabetic ketoacidosis (DKA), severe hypoglycemia, and technology use among racial and ethnic groups. In a diverse cohort of nearly 20,000 children and adults with type 1 diabetes, A1C was found to differ significantly among racial and ethnic groups. Non-Hispanic Blacks had higher rates of DKA and severe hypoglycemia and the lowest rate of technology use. These results underscore the crucial need to study and overcome the barriers that lead to inequities in the care and outcomes of people with type 1 diabetes.

THE GUIDED TRANSFER OF CARE IMPROVES ADULT CLINIC SHOW RATE.

Objective: Every year, 500,000 youths in the U.S. with chronic disease turn 18 years of age and eventually require transfer to adult subspecialty care. Evidence-based interventions on the organization of transfer of care are limited, although engagement and retention in adult clinic are considered appropriate outcomes. Sustained continuity of care improves patient satisfaction and reduces hospitalization. Methods: We conducted a prospective, nonrandomized cohort study of patients with pediatric endocrine conditions, age 16 to 26 years, enrolled upon referral to the adult endocrine clinic of a physician trained in both adult and pediatric endocrinology (Med+Peds endocrinologist). Patients differed based on whether their referral originated from another pediatric endocrinologist (traditional transfer) or if the Med+Peds endocrinologist previously saw the patient in his pediatric endocrine clinic (guided transfer). Rather than relying on arbitrary age criteria, guided transfer to adult clinic occurred when physician and patient considered it appropriate. The primary outcome was show rate at the first and second adult visits. Results: Of 36 patients, 21 were referred by another pediatric endocrinologist and 15 underwent guided transfer. For traditional transfer, show rate to the first and second visit was 38%, compared to 100% in the guided transfer group (P = .0001). Subgroup analysis of 27 patients with diabetes revealed that both groups had similar initial hemoglobin A1c (P = .38), and the guided transfer group maintained hemoglobin A1c. Conclusion: Most traditional transfers were unsuccessful. Guided transfer was significantly more effective, with every patient successfully transferring, and could be implemented with adult endocrinologists willing to see patients in the pediatric clinic. Abbreviations: DKA = diabetic ketoacidosis; HbA1c = hemoglobin A1c; Med+Peds = Internal Medicine and Pediatrics.

Primary Care Providers in California and Florida Report Low Confidence in Providing Type 1 Diabetes Care.

People with type 1 diabetes may receive a significant portion of their care from primary care providers (PCPs). To understand the involvement of PCPs in delivering type 1 diabetes care, we performed surveys in California and Florida, two of the most populous and diverse states in the United States. PCPs fill insulin prescriptions but report low confidence in providing type 1 diabetes care and difficulty accessing specialty referrals to endocrinologists.

The Design of the Electronic Health Record in Type 1 Diabetes Centers: Implications for Metrics and Data Availability for a Quality Collaborative.

BACKGROUND: Systematic and comprehensive data acquisition from the electronic health record (EHR) is critical to the quality of data used to improve patient care. We described EHR tools, workflows, and data elements that contribute to core quality metrics in the Type 1 Diabetes Exchange Quality Improvement Collaborative (T1DX-QI). METHOD: We conducted interviews with quality improvement (QI) representatives at 13 T1DX-QI centers about their EHR tools, clinic workflows, and data elements. RESULTS: All centers had access to structured data tools, nine had access to patient questionnaires and two had integration with a device platform. There was significant variability in EHR tools, workflows, and data elements, thus the number of available metrics per center ranged from four to 17 at each site. Thirteen centers had information about glycemic outcomes and diabetes technology use. Seven centers had measurements of additional self-management behaviors. Centers captured patient-reported outcomes including social determinants of health (n = 9), depression (n = 11), transition to adult care (n = 7), and diabetes distress (n = 3). Various stakeholders captured data including health care professionals, educators, medical assistants, and QI coordinators. Centers that had a paired staffing model in clinic encounters distributed the burden of data capture across the health care team and was associated with a higher number of available data elements. CONCLUSIONS: The lack of standardization in EHR tools, workflows, and data elements captured resulted in variability in available metrics across centers. Further work is needed to support measurement and subsequent improvement in quality of care for individuals with type 1 diabetes.

CD39 delineates chimeric antigen receptor regulatory T cell subsets with distinct cytotoxic & regulatory functions against human islets.

Human regulatory T cells (Treg) suppress other immune cells. Their dysfunction contributes to the pathophysiology of autoimmune diseases, including type 1 diabetes (T1D). Infusion of Tregs is being clinically evaluated as a novel way to prevent or treat T1D. Genetic modification of Tregs, most notably through the introduction of a chimeric antigen receptor (CAR) targeting Tregs to pancreatic islets, may improve their efficacy. We evaluated CAR targeting of human Tregs to monocytes, a human ß cell line and human islet ß cells in vitro. Targeting of HLA-A2-CAR (A2-CAR) bulk Tregs to HLA-A2+ cells resulted in dichotomous cytotoxic killing of human monocytes and islet ß cells. In exploring subsets and mechanisms that may explain this pattern, we found that CD39 expression segregated CAR Treg cytotoxicity. CAR Tregs from individuals with more CD39low/- Tregs and from individuals with genetic polymorphism associated with lower CD39 expression (rs10748643) had more cytotoxicity. Isolated CD39- CAR Tregs had elevated granzyme B expression and cytotoxicity compared to the CD39+ CAR Treg subset. Genetic overexpression of CD39 in CD39low CAR Tregs reduced their cytotoxicity. Importantly, ß cells upregulated protein surface expression of PD-L1 and PD-L2 in response to A2-CAR Tregs. Blockade of PD-L1/PD-L2 increased ß cell death in A2-CAR Treg co-cultures suggesting that the PD-1/PD-L1 pathway is important in protecting islet ß cells in the setting of CAR immunotherapy. In summary, introduction of CAR can enhance biological differences in subsets of Tregs. CD39+ Tregs represent a safer choice for CAR Treg therapies targeting tissues for tolerance induction.

Disparities in adoption of new diabetic therapies with cardiovascular benefits.

AIMS: Sodium-glucose co-transporter 2 inhibitors (SGLT2i) and glucagon-like peptide-1 agonists (GLP1a) have cardiovascular benefit, but adoption into clinical practice has been lagging. We aim to evaluate use of SGLT2i and GLP1a across socioeconomic strata (SES), medical risk as well as provider type. METHODS: We conducted a retrospective cohort study of the prescription of SGLT2i or GLP1a within 12 months of clinic visit between January 1, 2018 and January 1, 2019 using de-identified claims data. The primary outcome was the composite of a medication fill of either an SGLT2i and/or GLP1a within 180 days of the index visit. RESULTS: Of the total cohort, 125,636 (15.8 %) received either a GLP-1a or SGLT2i.The odds of prescription of either medication was 0.64 [p = 0.006)] in patients with heart failure. Patients who identified as Black, Hispanic or Asian had lower odds of the primary outcome [Black: (AOR 0.81, p < 0.000); Hispanic: (AOR 0.87, p < 0.000); Asian: (AOR 0.83, p < 0.000). The odds was higher for those treated by an endocrinologist versus primary care clinician [AOR 2.12, p < 0.000)]. CONCLUSIONS: Prescriptionof SGLT2i or GLP1a was lower among patients with cardiovascular co-morbidities and those who identified as Black, Hispanic or Asian. Further efforts to minimize these disparities should be pursued.

Prevalence of fear of hypoglycemia in adults with type 1 diabetes using a newly developed screener and clinician's perspective on its implementation.

INTRODUCTION: Fear of hypoglycemia (FoH) affects quality of life, emotional well-being, and diabetes management among people with type 1 diabetes (PwT1D). American Diabetes Association's (ADA) guidelines recommend assessing FoH in clinical practice. However, existing FoH measures are commonly used in research and not in clinical practice. In this study, prevalence of FoH was assessed in PwT1D using a newly developed FoH screener for clinical practice; its association with established measures and outcomes was also determined. In addition, healthcare providers' (HCPs) perspectives on implementing FoH screener into real-world practice were explored. RESEARCH DESIGN AND METHODS: This multiphase observational study used mixed methods in two phases. First, we collected a cross-sectional survey (including the screener) from PwT1D (≥18 years) from T1D Exchange Quality Improvement Collaborative adult clinics. Pearson correlations and regression analyses were performed on diabetes outcome measures using screener scores. Second, we conducted focus groups among HCPs who treat PwT1D and descriptive analysis to summarize results. RESULTS: We included 553 PwT1D. Participants had a mean±SD age of 38.9±14.2 years and 30% reported a high FoH total score. Regression analyses showed that higher A1c and higher number of comorbidities were significantly associated with high FoH (p<0.001). High FoH worry and behavior scores were significantly associated with 8-Item Patient Health Questionnaire and 7-Item Generalized Anxiety Disorder Scale scores. Participants with ≥1 severe hypoglycemia event(s) and impaired awareness of hypoglycemia had higher odds of high FoH. Eleven HCPs participated in focus group interviews; they expressed that the FoH screener is clinically necessary and relevant but poses implementation challenges that must be addressed. CONCLUSIONS: Our results demonstrate FoH is common in PwT1D and affects their psychosocial well-being and diabetes management. In alignment with ADA position statement, HCP focus group results emphasize importance of screening for FoH. Implementing this newly developed FoH screener may help HCPs identify FoH in PwT1D.

Automated Insulin Delivery with Remote Real-Time Continuous Glucose Monitoring for Hospitalized Patients with Diabetes: A Multicenter, Single-Arm, Feasibility Trial.

Introduction: Multiple daily injection insulin therapy frequently fails to meet hospital glycemic goals and is prone to hypoglycemia. Automated insulin delivery (AID) with remote glucose monitoring offers a solution to these shortcomings. Research Design and Methods: In a single-arm multicenter pilot trial, we tested the feasibility, safety, and effectiveness of the Omnipod 5 AID System with real-time continuous glucose monitoring (CGM) for up to 10 days in hospitalized patients with insulin-requiring diabetes on nonintensive care unit medical-surgical units. Primary endpoints included the proportion of time in automated mode and percent time-in-range (TIR 70-180 mg/dL) among participants with >48 h of CGM data. Safety endpoints included incidence of severe hypoglycemia and diabetes-related ketoacidosis (DKA). Additional glycemic endpoints, CGM accuracy, and patient satisfaction were also explored. Results: Twenty-two participants were enrolled; 18 used the system for a total of 96 days (mean 5.3 ± 3.1 days per patient), and 16 had sufficient CGM data required for analysis. Median percent time in automated mode was 95% (interquartile range 92%-98%) for the 18 system users, and the 16 participants with >48 h of CGM data achieved an overall TIR of 68% ± 16%, with 0.17% ± 0.3% time <70 mg/dL and 0.06% ± 0.2% time <54 mg/dL. Sensor mean glucose was 167 ± 21 mg/dL. There were no DKA or severe hypoglycemic events. All participants reported satisfaction with the system at study end. Conclusions: The use of AID with a disposable tubeless patch-pump along with remote real-time CGM is feasible in the hospital setting. These results warrant further investigation in randomized trials.

AI-Human Hybrid Workflow Enhances Teleophthalmology for the Detection of Diabetic Retinopathy.

Objective: Detection of diabetic retinopathy (DR) outside of specialized eye care settings is an important means of access to vision-preserving health maintenance. Remote interpretation of fundus photographs acquired in a primary care or other nonophthalmic setting in a store-and-forward manner is a predominant paradigm of teleophthalmology screening programs. Artificial intelligence (AI)-based image interpretation offers an alternative means of DR detection. IDx-DR (Digital Diagnostics Inc) is a Food and Drug Administration-authorized autonomous testing device for DR. We evaluated the diagnostic performance of IDx-DR compared with human-based teleophthalmology over 2 and a half years. Additionally, we evaluated an AI-human hybrid workflow that combines AI-system evaluation with human expert-based assessment for referable cases. Design: Prospective cohort study and retrospective analysis. Participants: Diabetic patients ≥ 18 years old without a prior DR diagnosis or DR examination in the past year presenting for routine DR screening in a primary care clinic. Methods: Macula-centered and optic nerve-centered fundus photographs were evaluated by an AI algorithm followed by consensus-based overreading by retina specialists at the Stanford Ophthalmic Reading Center. Detection of more-than-mild diabetic retinopathy (MTMDR) was compared with in-person examination by a retina specialist. Main Outcome Measures: Sensitivity, specificity, accuracy, positive predictive value, and gradability achieved by the AI algorithm and retina specialists. Results: The AI algorithm had higher sensitivity (95.5% sensitivity; 95% confidence interval [CI], 86.7%-100%) but lower specificity (60.3% specificity; 95% CI, 47.7%-72.9%) for detection of MTMDR compared with remote image interpretation by retina specialists (69.5% sensitivity; 95% CI, 50.7%-88.3%; 96.9% specificity; 95% CI, 93.5%-100%). Gradability of encounters was also lower for the AI algorithm (62.5%) compared with retina specialists (93.1%). A 2-step AI-human hybrid workflow in which the AI algorithm initially rendered an assessment followed by overread by a retina specialist of MTMDR-positive encounters resulted in a sensitivity of 95.5% (95% CI, 86.7%-100%) and a specificity of 98.2% (95% CI, 94.6%-100%). Similarly, a 2-step overread by retina specialists of AI-ungradable encounters improved gradability from 63.5% to 95.6% of encounters. Conclusions: Implementation of an AI-human hybrid teleophthalmology workflow may both decrease reliance on human specialist effort and improve diagnostic accuracy. Financial Disclosures: Proprietary or commercial disclosure may be found after the references.