In This Issue of Diabetes Care

The relationship between food insecurity and diabetes is examined by Levi et al. (p. 1599), who look at the myriad mechanisms that underlie the association. They also examine the mixed success of interventions used in the U.S. and look at the gaps and opportunities for research, policy, and practice, again from the U.S. perspective. Highlighting the extent of the issue, they note that approximately 13.5 million U.S. households likely experience food insecurity, with disproportionately higher rates in minority communities, putting millions of individuals at greater risk of developing diabetes. They point to a bidirectional relationship in which food insecurity can lead to poor health and how poor health (including diabetes) can lead to food insecurity. Specifically, they divide the mechanisms into three areas: nutritional, behavioral, and mental health. The authors then review a range of interventions that might target food insecurity in the context of diabetes, covering programs that either directly give individuals healthy foods or “prescribe” healthy foods by providing households with money to purchase those foods at supermarkets. In a mixed picture of success, the authors describe how most approaches do help reduce food insecurity and improve dietary quality but are less successful on glycemic control and other diabetes health metrics. In a similar manner, the authors also look at the effects of federal nutrition assistance programs, particularly those of two of the largest programs, the Supplemental Nutrition Assistance Program (SNAP) and the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC). Both programs have shown evidence for their efficacy in reducing food insecurity and improving various health aspects, including diabetes outcomes. “Food security plays a critical role in the development and management of diabetes,” said Ronli Levi. “Programs that provide individuals with access to nutritious food can have a positive impact on diabetes prevention and treatment, ultimately reducing diabetes-related disparities.”

The complex relationship between neighborhood environments and diabetes risk and outcomes is explored by Mujahid et al. (p. 1609). They set out a framework that links the two across a continuum of risks and social/physical factors that very much reflect the historical and current social issues faced by many in the U.S. Starting from the perspective that behavioral and biological factors alone do not fully explain widespread inequities in diabetes risk, they look at various aspects of the built environment that might also affect the risk for diabetes. Specifically, they look at the physical built environment as well as factors that might arise because of the way it was built. These include area poverty and deprivation, access to food and recreational spaces, aesthetics, air pollution, social cohesion, disorder, safety, and crime, among others. In all cases, the authors provide example studies that link the built environment and subsequent risk for diabetes, although evident issues remain with methodologies and translating findings to real-world changes. A major theme to also emerge is the outsized influence of historic and structural racism on current general disease risks and specifically diabetes risk. They highlight numerous studies that show how historic practices (slavery, racism, segregation, and even terrorism) have influenced the design of neighborhood environments and how in myriad ways they have created inequities in health risks, including those for diabetes. The authors go on to make a series of recommendations for future research and the observational and interventional studies that might be needed to address many of the issues they identify. They also look at the research needed to address health inequities that the built environment influences, particularly to address structural racism that still appears rampant in the U.S. Interestingly, they also show how neighborhood-level factors might be incorporated into clinical care to identify local communities with high medical and social needs to develop interventions that improve diabetes outcomes.

A rapid reduction in blood glucose levels (i.e., HbA_1c) is not associated with early progression of mild or moderate nonproliferative diabetic retinopathy, according to Simó et al. (p. 1633). As a result, the authors suggest that clinicians should not be afraid to quickly optimize blood glucose levels in such patients. They caution, however, that the strategy should not be used in patients with vision-threatening diabetic retinopathy, for whom a careful decrease in HbA_1c should be planned. They also add that before any intensive glucose-lowering approach is used, the presence and degree of diabetic retinopathy should be determined. Concerns about early worsening of diabetic retinopathy following rapid reductions in HbA_1c in type 1 diabetes have been around since it was first identified in the Diabetes Control and Complications Trial (DCCT) in the 1990s. However, this potential complication also affects patients with type 2 diabetes. The findings come from a nested case-control study of 1,150 individuals with type 2 diabetes and mild or moderate nonproliferative diabetic retinopathy but with some retinopathy progression and matched control individuals with stable retinopathy. Rapid HbA_1c reduction was the main variable in question, and whether it worsened diabetic retinopathy. The authors found no significant difference in HbA_1c reduction in the case subjects or control individuals. There was no association with worsening of diabetic retinopathy in either unadjusted or adjusted modeling. Confounding variables included diabetes duration, baseline HbA_1c, hypertension, and antidiabetes drug use, among others. Stratification by baseline HbA_1c showed that those with higher levels did not have higher risk for early worsening of diabetic retinopathy. “Our results suggest that the rapid reduction of HbA_1c is not associated with progression of mild or moderate diabetic retinopathy,” said author Rafael Simó. “This is important because the vast majority of patients with type 2 diabetes have no retinopathy or only present with early stages of the disease. Therefore, in these patients the optimization of metabolic control should not be delayed.”

Depression may be a significant causal factor for type 2 diabetes, with some evidence indicating that the relationship is mediated by BMI, according to Maina et al. (p. 1707). Conversely, however, type 2 diabetes is not a significant causal factor for depression. The findings, the authors suggest, highlight the importance of trying to prevent type 2 diabetes at the onset of depression and the need to maintain a healthy weight in the context of both conditions. The findings come from Mendelian randomization analysis and genome-wide association studies (GWAS) of the UK Biobank and of other public databases to comprehensively examine the relationship between type 2 diabetes and depression. Observational studies on the relationship have pointed to it being bidirectional. However, as with all observational studies, concerns remain about unmeasured confounding and the potential for reverse causality, meaning the relationship remains unclear. As the authors point out, Mendelian randomization potentially can circumvent the issues to understand the relationship more fully. They found that Mendelian randomization could demonstrate a significant causal effect of depression on type 2 diabetes but not the other way around. Further analysis then uncovered evidence that the effect from depression to type 2 diabetes was mediated to some extent (∼37%) by BMI. The authors used GWAS to identify potential shared mechanisms, and they found that GWAS of type 2 diabetes and depression separately did not identify any shared loci. Moving to multiphenotype GWAS (to increase power), they found that a model of type 2 diabetes and major depressive disorder (the binary definition of depression) did not identify any significant associations. In contrast, the same analysis applied to type 2 diabetes and depression measured with Patient Health Questionnaire 9 uncovered seven single nucleotide polymorphisms that were shared between type 2 diabetes and depression. With further analysis, the authors then broadly conclude that insulin secretion and inflammation are potential pathophysiological functions underlying the effect of depression on type 2 diabetes development.