We included 5,791 older adults in the Atherosclerosis Risk in Communities Study who attended visit 5 (2011–2013; ages 66–90 years). We compared prediabetes (HbA_1c 5.7% to <6.5%), newly diagnosed diabetes (HbA_1c ≥6.5%, prior diagnosis <1 year, or taking antihyperglycemic medications <1 year), short-duration diabetes (duration ≥1 year but <10 years [median]), and long-standing diabetes (duration ≥10 years). Outcomes were all-cause and cardiovascular mortality (median follow-up of 5.6 years).

Participants were 58% female, and 24% had prevalent cardiovascular disease. All-cause mortality rates, per 1,000 person-years, were 21.2 (95% CI 18.7, 24.1) among those without diabetes, 23.7 (95% CI 20.8, 27.1) for those with prediabetes, 33.8 (95% CI 25.2, 45.5) among those with recently diagnosed diabetes, 29.6 (95% CI 25.0, 35.1) for those with diabetes of short duration, and 48.6 (95% CI 42.4, 55.7) for those with long-standing diabetes. Cardiovascular mortality rates, per 1,000 person-years, were 5.8 (95% CI 4.6, 7.4) among those without diabetes, 6.6 (95% CI 5.2, 8.5) for those with prediabetes, 11.5 (95% CI 7.0, 19.1) among those with recently diagnosed diabetes, 8.2 (95% CI 5.9, 11.3) for those with diabetes of short duration, and 17.3 (95% CI 13.8, 21.7) for those with long-standing diabetes. After adjustment for other cardiovascular risk factors, prediabetes and newly diagnosed diabetes were not significantly associated with a higher risk of all-cause mortality (hazard ratio [HR] 1.03 [95% CI 0.85, 1.23] and HR 1.31 [95% CI 0.94, 1.82], respectively) or cardiovascular mortality (HR 1.00 [95% CI 0.70, 1.43] and HR 1.35 [95% CI 0.74, 2.49], respectively). Excess mortality risk was primarily concentrated among those with long-standing diabetes (all-cause: HR 1.71 [95% CI 1.40, 2.10]; cardiovascular: HR 1.72 [95% CI 1.18, 2.51]).

In older adults, long-standing diabetes has a substantial and independent effect on short-term mortality. Older individuals with prediabetes remained at low mortality risk over a median 5.6 years of follow-up.

At baseline, 178 lipids were measured by mass spectrometry in 3,668 participants without diabetes from the Malmö Diet and Cancer Study. The population was randomly split into discovery (n = 1,868, including 257 incident cases) and replication (n = 1,800, including 249 incident cases) sets. We used orthogonal projections to latent structures discriminant analyses, extracted a predictive component for T2DM incidence (lipid-PC_DM), and assessed its association with T2DM incidence using Cox regression and lifestyle factors using general linear models.

A T2DM-predictive lipid-PC_DM derived from the discovery set was independently associated with T2DM incidence in the replication set, with hazard ratio (HR) among subjects in the fifth versus first quintile of lipid-PC_DM of 3.7 (95% CI 2.2–6.5). In comparison, the HR of T2DM among obese versus normal weight subjects was 1.8 (95% CI 1.2–2.6). Clinical lipids did not improve T2DM risk prediction, but adding the lipid-PC_DM to all conventional T2DM risk factors increased the area under the receiver operating characteristics curve by 3%. The lipid-PC_DM was also associated with a dietary risk score for T2DM incidence and lower level of physical activity.

A lifestyle-related lipidomic profile strongly predicts T2DM development beyond current risk factors. Further studies are warranted to test if lifestyle interventions modifying this lipidomic profile can prevent T2DM.

A validated survey of 585 eligible family or general internal medicine practices seeing ≥30 adult patients with diabetes in or near Minnesota during 2017 evaluated the presence of 62 SysCMs. From 419 (72%) practices completing the survey, NQF#0729 was determined in 396 (95%) from electronic health records, including 215,842 patients with type 1 or type 2 diabetes.

Three SysCMs were associated with higher rates of meeting performance targets across all practices: 1) a systematic process for shared decision making with patients (P = 0.001), 2) checklists of tests or interventions needed for prevention or monitoring of diabetes (P = 0.002), and 3) physician reminders of guideline-based age-appropriate risk assessments due at the patient visit (P = 0.002). When all three were in place, an additional 10.8% of the population achieved recommended performance measures. In subgroup analysis, 15 additional SysCMs were associated with better care in particular types of practices.

Diabetes care outcomes are better in primary care settings that use a patient-centered approach to systematically engage patients in decision making, remind physicians of age-appropriate risk assessments, and provide checklists for recommended diabetes interventions. Practice size and location are important considerations when redesigning delivery systems to improve performance.

A post hoc analysis was performed of the SWITCH (SWITCH 1: n = 501, type 1 diabetes; SWITCH 2: n = 721, type 2 diabetes) and DEVOTE (n = 7,637, type 2 diabetes) trials utilizing the IHSG low glucose definitions. Patients in all trials were randomized to either insulin degludec or insulin glargine 100 units/mL. In the main analysis, the following definitions were compared: 1) American Diabetes Association (ADA) 2005 (plasma glucose [PG] confirmed ≤3.9 mmol/L with symptoms); and 2) IHSG level 2 (PG confirmed <3.0 mmol/L, independent of symptoms).

In SWITCH 2, the estimated rate ratios of hypoglycemic events indicated increasing differences between treatments with decreasing PG levels until 3.0 mmol/L, following which no additional treatment differences were observed. Similar results were observed for the SWITCH 1 trial. In SWITCH 2, the IHSG level 2 definition produced a rate ratio that was lower than the ADA 2005 definition.

The IHSG level 2 definition was validated in a series of clinical trials, demonstrating its ability to discriminate between basal insulins. This definition is therefore recommended to be uniformly adopted by regulatory bodies and used in future clinical trials.

People with type 1 diabetes (n = 100) investigated at the University Hospital of Montpellier (France) underwent continuous glucose monitoring (CGM) on two consecutive days, providing a total of 200 24-h glycemic profiles. The following parameters were computed: MDG concentration, within-day glycemic variability (coefficient of variation for glucose [%CV]), and risk of hypoglycemia (presented as the percentage of time spent below three glycemic thresholds: 3.9, 3.45, and 3.0 mmol/L).

MDG was significantly higher, and %CV significantly lower (both P < 0.001), when comparing the 24-h glycemic profiles according to whether no time or a certain duration of time was spent below the thresholds. Univariate regression analyses showed that MDG and %CV were the two explanatory variables that entered the model with the outcome variable (time spent below the thresholds). The classification and regression tree procedure indicated that the predominant predictor for hypoglycemia was %CV when the threshold was 3.0 mmol/L. In people with mean glucose ≤7.8 mmol/L, the time spent below 3.0 mmol/L was shortest (P < 0.001) when %CV was below 34%.

In type 1 diabetes, short-term glycemic variability relative to mean glucose (i.e., %CV) explains more hypoglycemia than does mean glucose alone when the glucose threshold is 3.0 mmol/L. Minimizing the risk of hypoglycemia requires a %CV below 34%.

Data for 3,129 U.S. counties, obtained from the Small Area Health Insurance Estimates and Area Health Resources Files, were used to analyze trends in uninsured rates among populations with a household income ≤138% of the federal poverty level. Multivariable analysis adjusted for the percentage of county populations aged 50–64 years, the percentage of women, Distressed Communities Index value, and rurality.

In 2012, 39% of the population in the Diabetes Belt and 34% in non-Belt counties were uninsured (P < 0.001). In 2016 in states where Medicaid was expanded, uninsured rates declined rapidly to 13% in Diabetes Belt counties and to 15% in non-Belt counties. Adjusting for county demographic and economic factors, Medicaid expansion helped reduce uninsured rates by 12.3% in Diabetes Belt counties and by 4.9% in non-Belt counties. In 2016, uninsured rates were 15% higher for both Diabetes Belt and non-Belt counties in the nonexpansion states than in the expansion states.

ACA-driven Medicaid expansion was more significantly associated with reduced uninsured rates in Diabetes Belt than in non-Belt counties. Initial disparities in uninsured rates between Diabetes Belt and non-Belt counties have not existed since 2014 among expansion states. Future studies should examine whether and how Medicaid expansion may have contributed to an increase in the use of health services in order to prevent and treat diabetes in the Diabetes Belt.

We linked data from the National Diabetes Register and the Scandinavian Obesity Surgery Register with four national databases holding information on socioeconomic variables, medications, hospitalizations, and causes of death and matched 5,321 individuals with T2DM who had undergone GBP with 5,321 who had not (age 18–65 years, mean BMI >40 kg/m², mean follow-up >4.5 years). The risks of postoperative outcomes were assessed with Cox regression models.

During the first years postsurgery, there were small reductions in creatinine and albuminuria and stable estimated glomerular filtration rate (eGFR) in the GBP group. The incidence rates of most outcomes relating to renal function, CV disease, and mortality were lower after GBP, being particularly marked for heart failure (hazard ratio [HR] 0.33 [95% CI 0.24, 0.46]) and CV mortality (HR 0.36 [(95% CI 0.22, 0.58]). The risk of a composite of severe renal disease or halved eGFR was 0.56 (95% CI 0.44, 0.71), whereas nonfatal CV risk was lowered less (HR 0.82 [95% CI 0.70, 0.97]) after GBP. Risks for key outcomes were generally lower after GBP in all eGFR strata, including in individuals with eGFR <30 mL/min/1.73 m².

Our data suggest robust benefits for renal outcomes, heart failure, and CV mortality after GBP in individuals with obesity and T2DM. These results suggest that marked weight loss yields important benefits, particularly on the cardiorenal axis (including slowing progression to end-stage renal disease), whatever the baseline renal function status.

A cross-sectional study of 193 patients with type 2 diabetes and 20 age- and sex-matched controls. Cardiovascular magnetic resonance was used in order to evaluate left ventricular structure and function, and MBF at rest and during adenosine-induced stress. MBF was derived as the mean of the flow within all segments of a midventricular slice.

Patients with type 2 diabetes had higher global MBF at rest (0.81 ± 0.19 mL/min/g) and lower global MBF during stress (2.4 ± 0.9 mL/min/g) than did controls (0.61 ± 0.11 at rest, 3.2 ± 0.8 mL/min/g under stress; both P < 0.01). Patients with macroalbuminuria had lower MBF during stress (1.6 ± 0.5 mL/min/g) than did patients with microalbuminuria (2.1 ± 0.7 mL/min/g; P = 0.04), who in turn had lower MBF during stress than did normoalbuminuric patients (2.7 ± 0.9 mL/min/g; P < 0.01). Patients with severe retinopathy had lower MBF during stress (1.8 ± 0.6 mL/min/g) than did patients with simplex retinopathy (2.3 ± 0.7 mL/min/g; P < 0.05) and those who did not have retinopathy (2.6 ± 1.0 mL/min/g; P < 0.05). Albuminuria and retinopathy were associated with reduced MBF during stress in a multiple regression analysis. Stress-related MBF inversely correlated with myocardial extracellular volume (P < 0.001; R² = 0.37), a measure of diffuse myocardial fibrosis. A trend toward lower basal MBF was observed in patients treated with sodium–glucose cotransporter 2 inhibitors (P = 0.07).

Patients with type 2 diabetes have higher global MBF at rest and lower maximal MBF during vasodilator-induced stress than do controls. Reduced MBF during stress is associated with diabetes complications (albuminuria and retinopathy) and is inversely correlated with diffuse myocardial fibrosis.

This multicenter, double-blind, treat-to-target trial randomized participants to faster aspart (n = 546) or IAsp (n = 545). All available information, regardless of treatment discontinuation or use of ancillary treatment, was used for evaluation of effect.

Noninferiority for the change from baseline in HbA_1c 16 weeks after randomization (primary end point) was confirmed for faster aspart versus IAsp (estimated treatment difference [ETD] –0.04% [95% CI –0.11; 0.03]; –0.39 mmol/mol [–1.15; 0.37]; P < 0.001). Faster aspart was superior to IAsp for change from baseline in 1-h postprandial glucose (PPG) increment using a meal test (ETD –0.40 mmol/L [–0.66; –0.14]; –7.23 mg/dL [–11.92; –2.55]; P = 0.001 for superiority). Change from baseline in self-measured 1-h PPG increment for the mean over all meals favored faster aspart (ETD –0.25 mmol/L [–0.42; –0.09]); –4.58 mg/dL [–7.59; –1.57]; P = 0.003). The overall rate of treatment-emergent severe or blood glucose (BG)–confirmed hypoglycemia was statistically significantly lower for faster aspart versus IAsp (estimated treatment ratio 0.81 [95% CI 0.68; 0.97]).

In combination with insulin degludec, faster aspart provided effective overall glycemic control, superior PPG control, and a lower rate of severe or BG-confirmed hypoglycemia versus IAsp in adults with type 2 diabetes not optimally controlled with a basal-bolus regimen.

This retrospective cohort combined data from the National Veterans Health Administration, Medicare, Medicaid, and the National Death Index. New users of metformin or sulfonylureas were followed from development of reduced kidney function (estimated glomerular filtration rate [eGFR] <60 mL/min/1.73 m² or serum creatinine ≥1.4 mg/dL [female] or 1.5 mg/dL [male]) through hospitalization for lactic acidosis, death, loss to follow-up, or study end. Lactic acidosis hospitalization was defined as a composite of primary discharge diagnosis or laboratory-confirmed lactic acidosis (lactic acid ≥2.5 mmol/L and either arterial blood pH <7.35 or serum bicarbonate ≤19 mmol/L within 24 h of admission). We report the cause-specific hazard of lactic acidosis hospitalization between metformin and sulfonylureas from a propensity score–matched weighted cohort and conduct an additional competing risks analysis to account for treatment change and death.

The weighted cohort included 24,542 metformin and 24,662 sulfonylurea users who developed reduced kidney function (median age 70 years, median eGFR 55.8 mL/min/1.73 m²). There were 4.18 (95% CI 3.63, 4.81) vs. 3.69 (3.19, 4.27) lactic acidosis hospitalizations per 1,000 person-years among metformin and sulfonylurea users, respectively (adjusted hazard ratio [aHR] 1.21 [95% CI 0.99, 1.50]). Results were consistent for both primary discharge diagnosis (aHR 1.11 [0.87, 1.44]) and laboratory-confirmed lactic acidosis (1.25 [0.92, 1.70]).

Among veterans with diabetes who developed reduced kidney function, occurrence of lactic acidosis hospitalization was uncommon and not statistically different between patients who continued metformin and those patients who continued sulfonylureas.

Baseline data from the publications of the three trials were obtained and entered into the BRAVO model to predict cardiovascular outcomes. Projected benefits of reducing risk factors of interest (A1C, systolic blood pressure [SBP], LDL, or BMI) on cardiovascular events were evaluated, and simulated outcomes were compared with those observed in each trial.

BRAVO achieved the best prediction accuracy when simulating outcomes of the CANVAS and DECLARE-TIMI 58 trials. For the EMPA-REG OUTCOME trial, a mild bias was observed (~20%) in the prediction of mortality and angina. The effect of risk reduction on outcomes in treatment versus placebo groups predicted by the BRAVO model strongly correlated with the observed effect of risk reduction on the trial outcomes as published. Finally, the BRAVO engine revealed that most of the clinical benefits associated with SGLT2i treatment are through A1C control, although reductions in SBP and BMI explain a proportion of the observed decline in cardiovascular events.

The BRAVO risk engine was effective in predicting the benefits of SGLT2is on cardiovascular health through improvements in commonly measured risk factors, including A1C, SBP, and BMI. Since these benefits are individually small, the use of the complex, dynamic BRAVO model is ideal to explain the cardiovascular outcome trial results.

Four years of data were extracted from a hospital electronic health record system. This included laboratory and point-of-care blood glucose (BG) values to identify biochemical and clinically significant hypoglycemic episodes (BG ≤3.9 and ≤2.9 mmol/L, respectively). We used patient demographics, administered medications, vital signs, laboratory results, and procedures performed during the hospital stays to inform the model. Two iterations of the data set included the doses of insulin administered and the past history of inpatient hypoglycemia. Eighteen different prediction models were compared using the area under the receiver operating characteristic curve (AUROC) through a 10-fold cross validation.

We analyzed data obtained from 17,658 inpatients with diabetes who underwent 32,758 admissions between July 2014 and August 2018. The predictive factors from the logistic regression model included people undergoing procedures, weight, type of diabetes, oxygen saturation level, use of medications (insulin, sulfonylurea, and metformin), and albumin levels. The machine learning model with the best performance was the XGBoost model (AUROC 0.96). This outperformed the logistic regression model, which had an AUROC of 0.75 for the estimation of the risk of clinically significant hypoglycemia.

Advanced machine learning models are superior to logistic regression models in predicting the risk of hypoglycemia in inpatients with diabetes. Trials of such models should be conducted in real time to evaluate their utility to reduce inpatient hypoglycemia.

We performed exploratory analyses to identify potential mediators of the effect of liraglutide on major adverse CV events (MACE; composite of CV death, nonfatal myocardial infarction, or nonfatal stroke) from the following candidates: glycated hemoglobin (HbA_1c), body weight, urinary albumin-to-creatinine ratio (UACR), confirmed hypoglycemia, sulfonylurea use, insulin use, systolic blood pressure, and LDL cholesterol. These candidates were selected as CV risk factors on which liraglutide had an effect in LEADER such that a reduction in CV risk might result. We used two methods based on a Cox proportional hazards model and the new Vansteelandt method designed to use all available information from the mediator and to control for confounding factors.

Analyses using the Cox methods and Vansteelandt method indicated potential mediation by HbA_1c (up to 41% and 83% mediation, respectively) and UACR (up to 29% and 33% mediation, respectively) on the effect of liraglutide on MACE. Mediation effects were small for other candidates.

These analyses identify HbA_1c and, to a lesser extent, UACR as potential mediators of the CV effects of liraglutide. Whether either is a marker of an unmeasured factor or a true mediator remains a key question that invites further investigation.