Digital Lifestyle Intervention Effectively Reduces Diabetes Risk

A recent U.S. study published in Nutrients evaluated the effectiveness of a digital lifestyle modification program in a large group of prediabetic, diabetic, and healthy adults, focusing on reducing cardiovascular and diabetes risk and improving metabolic markers. Researchers found that the intervention significantly decreased 10-year diabetes risk by nearly 46% in prediabetic individuals and increased diabetes remission rates, highlighting the potential of lifestyle changes, though the study’s non-randomized design and voluntary participation may have introduced bias.

Background

Diabetes mellitus, characterized by elevated fasting glucose or HbA1c, is a major risk factor for several complications, including neuropathy, retinopathy, kidney disease, and atherosclerotic cardiovascular disease (ASCVD). Prediabetes, affecting approximately one-third of middle-aged and older U.S. adults, increases the risk of progressing to diabetes; risk factors include inactivity, family history, and obesity. While behavioral lifestyle interventions promoting 7% weight loss and increased physical activity can reduce diabetes risk by half, traditional in-person programs face adoption challenges. Risk prediction models incorporating factors like glucose, body mass index (BMI), HDL-C, and triglycerides help identify high-risk individuals. The authors previously developed a 10-year diabetes risk model using glycated serum albumin, fasting glucose, adiponectin, and triglycerides, demonstrating high predictive accuracy in the Framingham Offspring Study. Given the importance of lifestyle changes in preventing diabetes and ASCVD, the study sought to test easily implementable interventions targeting high-risk individuals.

Study Design

The study included 133,764 adults categorized as diabetic (7.5%), prediabetic (36.2%), or healthy (56.3%) based on fasting glucose and HbA1c levels. Participants underwent fasting blood tests to measure adiponectin, insulin, glucose, HbA1c, glycated serum protein, hs-CRP, fibrinogen, myeloperoxidase, LpPLA2, LDL-C, small dense LDL-C, and standard lipids. After 6 to 12 months, follow-up blood samples were taken from slightly over 20% of prediabetic and 22% of diabetic participants. Of those with follow-up data, 12.2% of prediabetic and 9.7% of diabetic participants engaged in a voluntary, digital, dietitian-guided lifestyle program focused on dietary and behavioral changes. The program’s impact was assessed using a 10-year diabetes risk model developed by the authors, incorporating fasting glucose, glycated serum albumin, adiponectin, and triglyceride levels. Changes in diabetes risk, metabolic markers, weight loss, and remission rates were analyzed to evaluate the program’s effectiveness compared to those who did not participate.

Results

The diabetic and prediabetic groups were older and heavier, with higher BMI and body weight, and included fewer women than the healthy group. Blood glucose control worsened across groups, with prediabetic and diabetic individuals exhibiting significantly higher HbA1c, fasting glucose, glycated serum protein, fasting insulin, and C-peptide levels. Insulin resistance was markedly elevated, increasing by 75% and 260% in prediabetic and diabetic men, respectively, and by 112% and 306% in women. Insulin production was notably lower only in diabetic subjects, with many showing both insulin resistance and reduced production.

Inflammation markers, particularly hs-CRP, were higher in prediabetic and diabetic individuals, while changes in adiponectin, fibrinogen, myeloperoxidase, and LpPLA2 were more modest. The 10-year risk of diabetes was substantially higher in prediabetics (7% for men, 4.2% for women) compared to healthy individuals (0.6% for men, 0.3% for women).

Lipid profiles showed only modest changes in LDL-C and apolipoproteins. However, prediabetic and diabetic subjects had significantly higher fasting triglycerides and small dense LDL-C, and lower HDL-C levels, indicating a more atherogenic lipid profile. For instance, small dense LDL-C increased by up to 35% in diabetic women, while HDL-C decreased by 23%, and triglycerides rose by 70%.

Lifestyle modification in prediabetic individuals significantly reduced diabetes risk, triglycerides, LDL-C, and insulin resistance, and increased adiponectin levels compared to controls. Predicted diabetes risk decreased by 45.6% in the lifestyle intervention group, compared to 1.6% in the control group. Among diabetic participants, the lifestyle group had a 2.4-fold increase in remission rate (8.2% vs. 3.4%), greater weight loss, and improved glycemic and inflammatory markers.

Conclusions

The study revealed that prediabetic and diabetic individuals exhibit significant metabolic and inflammatory changes compared to healthy individuals. Insulin resistance appears to be a primary driver of early abnormalities, while both insulin resistance and reduced insulin secretion are evident in established diabetes. Elevated hs-CRP suggests a key role for inflammation, with other inflammatory marker changes being less pronounced. Although changes in LDL-C and apolipoproteins were relatively small, higher triglycerides and small dense LDL-C, and lower HDL-C indicate increased cardiovascular risk even before a diabetes diagnosis.

The results highlight that metabolic deterioration begins before the onset of clinical diabetes, emphasizing the importance of early identification and intervention in high-risk individuals. While the digital lifestyle program improved risk markers, the authors note that fully digital interventions may be less effective than blended or face-to-face approaches. Meta-analyses suggest that combined in-person and digital interventions achieve greater rates of return to normal blood sugar levels than digital-only programs. Future research should further examine the roles of inflammation and abnormal lipid levels in the progression of diabetes and cardiovascular disease risk.

Source: Nutrients, April 11, 2025