In This Issue of Diabetes Care Free

Homelessness is an independent risk factor for adverse diabetes outcomes, according to Sharan et al. (p. 1469). Specifically, they found that rates of macrovascular complications, hospitalizations for glycemia issues, and skin/soft tissue infections were higher in individuals who had diabetes and experienced homelessness. In contrast, rates of coronary revascularization procedures were reduced in those with a history of homelessness. Low income levels did not fully account for the overall higher rates found in those who had experienced homelessness. On that basis, the authors conclude that homelessness is likely to independently raise the risk for diabetes complications beyond that related to income disparities or other known risk factors. The findings come from a propensity-matched cohort study that used hospital administrative data from Ontario, Canada. Just over a million individuals with diabetes were eligible for inclusion, of whom the authors identified 6,944 individuals with a history of homelessness. They then found 5,219 individuals who had not experienced homelessness as acceptable matches for those with documented homelessness and a slightly smaller comparison group of low-income individuals. Based on the findings, the authors conclude that targeted outreach practices are essential for the group of individuals experiencing homelessness, noting various interventions likely to result in decreased rates of complications. A notable omission from the study is any consideration of microvascular complications, which they propose should be the focus of further research. “Many of us encounter people with experience of homelessness in clinical practice—oftentimes in acute care settings with acute or chronic complications,” said author David J.T. Campbell. “This study is one of the first to conclusively demonstrate the significant burden of excess adverse outcomes seen in this population. We hope that it can help to increase the awareness of the challenges this population faces in diabetes management within typical practice settings and lead to more investment in targeted programming to improve outcomes.”

Further analysis of the Treatment Options for Type 2 Diabetes in Adolescents and Youth (TODAY) study reveals that youth with type 2 diabetes who are in the traditionally viewed range of tight glycemic control still experience significant deterioration of glycemic control over time. According to the TODAY Study Group (p. 1507), intensification of therapy in youth-onset type 2 diabetes should still be considered, but the findings call into question the use of strategies based on metformin to achieve the goals in the context of youth-onset type 2 diabetes. Various studies (including the TODAY study and the Restoring Insulin Secretion [RISE] Pediatric Medication Study) of therapies centered on metformin have now shown no improvements in insulin sensitivity or β-cell function in youth. As such, the authors suggest focus should turn to more modern agents in a bid to help improve glycemia in youth with type 2 diabetes. The findings come from further analysis of the TODAY study cohort that looked at the effect of an initial 6-month period of glycemic control with metformin on insulin sensitivity and secretion. Participants were then followed for up to 9 years with longitudinal oral glucose tolerance tests. A total of 656 participants were included in five HbA_1c categories, ranging from <5.7% to >8%. The authors found that HbA_1c increased in all five groups during years 2–9, with the steepest increase registered in the group with the lowest initial HbA_1c (i.e., <5.7%). In parallel, there was a decline in C-peptide–derived disposition index, which is a measure of insulin secretion relative to insulin demand. “Overall, we need a better understanding of which youth will go on to develop type 2 diabetes to be able to target them with earlier intervention,” said author Kristen J. Nadeau. “The current work shows us that deterioration of β-cell function is difficult to arrest in youth once dysglycemia is present. In addition, we need more research on the effect of newer diabetes medications and bariatric metabolic surgery on β-cell function in youth who already have type 2 diabetes.”

Individuals with an elevated type 2 diabetes polygenic score but not a diabetes diagnosis are at higher risk of developing hyperglycemia following treatment with glucocorticoids, according to Deutsch et al. (p. 1541). The commonly prescribed drugs are known to have such a side effect in 10–50% of individuals receiving them. According to the authors, the score might help identify individuals at risk and help clinicians adjust dosing, find alternatives, or prompt closer monitoring for hyperglycemia if suitable genetic information becomes more widely available. The findings come from a retrospective analysis of medical records from participants in the Mass General Brigham Biobank who had no diabetes diagnosis and who had received at least one glucocorticoid dose equivalent to ≥10 mg prednisone. The authors then looked at the association between occurrence of hyperglycemia and a polygenic score for type 2 diabetes constructed via meta-analysis of two previously published genome-wide association studies. After they identified 546 suitable individuals who received glucocorticoids, the authors found that 210 developed hyperglycemia while the rest did not. They also found that the polygenic score, which captures the cumulative risk for type 2 diabetes, was significantly associated with glucocorticoid-induced hyperglycemia. Estimated glomerular filtration rate (a measure of kidney function) and glucocorticoid dose were other significant covariates. They found that these two covariates could predict hyperglycemia with an area under the receiver operating curve of 0.65. Incorporating the score boosted that measure (significantly) to 0.68. Based on their findings, the authors suggest that the two conditions share a common pathophysiology but more practically see the score as a method to predict the onset of glucocorticoid-induced hyperglycemia. “Our study brings to light the correlation between genetic risk of type 2 diabetes and hyperglycemia related to glucocorticoid use,” said author Laura N. Brenner. “This study raises the question of how genetic risk scores could be used to predict complications in the future.”

Gestational diabetes mellitus (GDM) during pregnancy appears to influence the risk of future postpartum diabetes and the likelihood of attaining glycemic control following diagnosis, according to McCarthy et al. (p. 1483). Based on the outcomes, the authors suggest that a diagnosis of GDM is a powerful warning sign for early interventions to avoid later development of full-blown diabetes. The findings come from a multiethnic population-based cohort study of postpartum women in New York City that looked to estimate racial and ethnic differences in diabetes risk and later glucose control according to whether they had GDM or not. The study, which included just over 330,000 births, linked hospital discharge and registry data with New York’s A1C registry data. They found that over the study period, the cumulative incidence of diabetes was 11.8% for those with GDM and 0.6% for those without. The influence of GDM status on diabetes risk had an adjusted hazard ratio of 11.5 overall, with only modest differences according to race and ethnicity. The authors also found that GDM influences the length of time it takes to transition to glycemic control once diabetes is diagnosed. The slowest transition was among Black and Hispanic women, and while the authors did not specifically study the reasons for this, they cite missed opportunities to detect GDM, insufficient insurance coverage, and reduced basic care levels as possible contributing socioeconomic factors and health care–related factors that are also the subject of future research. “These findings build on the limited existing evidence of racial and ethnic differences in the influence of GDM on postpartum diabetes risk and control,” author Katharine J. McCarthy said. “Our results highlight the importance of obstetric history taking in facilitating early diabetes detection and control in the transition from obstetric to primary care following pregnancy as well as the need for policies to expand health care coverage following delivery.”