A digital health care platform appeared to improve glycemia and increased weight loss in diabetes over 24 weeks, according to Lee et al. (p. 959). Notably, the app-based system includes an artificial intelligence (AI) system that apparently recognizes specific foods in photographs and automatically estimates quantities and nutritional values. The 48-week multicenter (Korea-based) open-label randomized controlled trial included adults with type 2 diabetes with HbA1c in the range of 7.0–8.5% and BMI ≥23 kg/m2. Three groups then received routine diabetes care or used the platform with or without supervision from medical staff. The group that received supervision also intermittently used continuous glucose monitoring. According to the authors, the decreases in HbA1c over 24 weeks were significantly larger in the group using the platform unsupervised (–0.32 ± 0.58%) and supervised (–0.49 ± 0.57%) compared with the group receiving standard care (–0.06 ± 0.61%). The groups that used the platform also experienced significantly greater weight loss than those who received standard care. They note that while the reductions in HbA1c were relatively moderate, this was likely a function of having a low baseline. They also suggest that some level of patient education and feedback will increase the effect size. In terms of strengths, the authors point to the automatic integration of data from various devices and the AI-based food recognition approach. In this case, they add that this addresses the cognitive burden of dietary records, which is one of the most laborious parts of diabetes self-management (and wider nutrition research). Further research will be needed to address some limitations, most notably in areas of generalizability and validity. “Our research suggests the feasibility of improving glycemia and weight loss solely using an integrated digital health care platform without modifying medication,” said author Jae Hyeon Kim. “Especially for individuals who are capable and motivated, introduction of a convenient, user-friendly digital integrated health care platform such as ours could be an efficacious tool for improving type 2 diabetes self-management.”

Lee et al. An integrated digital health care platform for diabetes management with AI-based dietary management: 48-week results from a randomized controlled trial. Diabetes Care 2023;46:959–966

The use of sulfonylureas as an add-on therapy in diabetes is unlikely to have raised cardiovascular or all-cause mortality risks, according to Wang et al. (p. 967). Specifically, this therapy was used as an add-on to metformin and compared to dipeptidyl peptidase 4 (DPP-4) inhibitors and thiazolidinediones rather than more modern therapies. “Our study provides the most robust real-world evidence for the cardiovascular safety of sulfonylureas being prescribed in addition to metformin,” said author Ewan R. Pearson. Based on the findings, the authors suggest that sulfonylureas should remain part of diabetes treatment portfolios, given their efficacy, benefits, and cost-effectiveness. There have been persistent doubts about the cardiovascular safety of the drug class, although this tends to be based on early studies that had suboptimal designs compared to modern standards. The findings come from a cohort study that included individuals with type 2 diabetes in Scotland who did not manage to reduce HbA1c sufficiently with metformin and initiated sulfonylureas, DPP-4 inhibitors, and thiazolidinediones as an add-on therapy. A composite of major adverse cardiovascular events (MACE) was used as a primary outcome, while individual end points and all-cause mortality were secondary outcomes. Notably, the authors also included an instrumental variable approach to control potential confounding. “We have also developed robust methodology for estimating causal treatment effects using electronic health care records,” Pearson noted. In all cases following adjustment, the authors found the three drug classes did not differ in terms of either the primary or secondary outcome, meaning that prescribing sulfonylureas is likely not significantly associated with increased risks for MACE or all-cause mortality. They note that their approach has produced results nearly identical to those of major cardiovascular outcome trials (CAROLINA and TOSCA.IT) that involved comparisons of the three drug classes as add-on therapies to metformin. “Our findings support the recent international guidelines, which recommend sulfonylureas should remain part of the global diabetes treatment portfolio,” Pearson added. “Physicians can feel confident to prescribe these drugs when needed.”

Wang et al. Cardiovascular safety in type 2 diabetes with sulfonylureas as second-line drugs: a nationwide population-based comparative safety study. Diabetes Care 2023;46:967–977

Data from two population-based studies (Edstorp et al. [p. 1028]) appear to show that a genetic risk score for type 2 diabetes is associated with incidence of latent autoimmune diabetes in adults (LADA). They also found that genetic susceptibility to type 2 diabetes and insulin resistance strengthened the relationship between tobacco use and LADA risk. Based on these findings, they suggest that prevention of tobacco use inhibits the development of LADA, with it being especially important in individuals with genetic predisposition to type 2 diabetes and insulin resistance. The findings come from two Scandinavian studies (Epidemiological Study of Risk Factors for LADA and Type 2 Diabetes [ESTRID] and the Trøndelag Health Study [HUNT]) that overall included 839 individuals with LADA and just under 5,800 individuals with type 2 diabetes. They also included just over 3,000 matched control subjects. Genetic risk scores were then calculated for type 2 diabetes, insulin resistance, and impaired insulin secretion to look at the risks associated with smoking and LADA and any interactions between the various terms. The authors found that the risk for LADA was elevated in individuals with a high score for insulin resistance and who were current and heavy smokers. This was compared to individuals who had low scores for insulin resistance and current tobacco use. When they looked at interactions, they also found that there was an additive interaction between the score for type 2 diabetes and smoking status. They also note that genetic susceptibility to type 2 diabetes or insulin resistance/secretion had limited impact on any association between tobacco use and incidence of type 2 diabetes. “Our study adds to the scarce knowledge of potential risk factors for adult-onset autoimmune diabetes by showing that smokers, particularly those who also have genetic susceptibility to type 2 diabetes and insulin resistance, are at increased risk of LADA,” said author Jessica Edstorp. “This highlights the importance of tobacco use prevention for reducing the diabetes burden and that this is not limited to classical type 2 diabetes.”

Edstorp et al. Incidence of LADA and type 2 diabetes in relation to tobacco use and genetic susceptibility to type 2 diabetes and related traits: findings from a Swedish case-control study and the Norwegian HUNT study. Diabetes Care 2023;46:1028–1036

It has now been just over 3 years since the COVID-19 pandemic was declared by the World Health Organization. In parallel, theories and data have started to emerge that suggest COVID-19 is causing some sort of diabetes in the still-vague realm of long COVID. William T. Cefalu, director of the Division of Diabetes, Endocrinology, and Metabolic Diseases at the National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, has written a commentary (p. 913) on where we presently stand on this issue. He focuses on three reports that appear in this issue of Diabetes Care and raises important questions about what might need to be looked at next. First is Sasidharan Pillai et al. (p. 953), who suggest there was a threefold increase in type 2 diabetes incidence from prepandemic to pandemic years. They conclude it was not the virus per se that caused the increase; rather, it was more about “pandemic-related obesogenic factors.” Next, McKeigue et al. (p. 921) evaluate an apparent increased incidence of type 1 diabetes in Scotland following a severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection. While they report a 1.2-fold increase in incident type 1 diabetes, a parallel cohort analysis found no link. According to Cefalu, the study has notable strengths as well as weaknesses. Last is Holman et al. (p. 938), who make the case that it is not clear yet whether apparent increased incidence of diabetes post-COVID-19 reflects shared risk factors, specific impacts from an actual infection, or stress induced by treatment for an infection. According to Cefalu, the three studies collectively add to the evidence for and against rising incident diabetes cases post-COVID-19 and whether there is a causal relationship. However, there are still numerous unanswered questions, and he raises many of them. He notes that there are still questions around whether COVID-19 infections result in diabetes per se or, perhaps more importantly, whether the imposed social conditions created by the pandemic have unmasked many more cases than otherwise would have been identified.

Cefalu. COVID-19 and rising incidence of diabetes: despite evolving data, an enigma still to be solved. Diabetes Care 2023;46:913–915

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at https://www.diabetesjournals.org/journals/pages/license.