Diabetes is a major public health challenge worldwide. From 1990 to 2021, global diabetes prevalence increased threefold, with projections indicating over 1.3 billion affected individuals by 2025 if uncontrolled (1). According to diagnostic criteria established by the American Diabetes Association (ADA), diabetes mellitus is classified as a persistent metabolic disorder that is diagnostically confirmed through sustained hyperglycemia (defined as fasting plasma glucose ≥ 7.0 mmol/L) or elevated glycosylated hemoglobin (HbA1c) levels ≥ 6.5%, pathophysiologically rooted in a disorder of insulin production or cellular responsiveness (2). Prediabetes, a transitional phase between normal glucose regulation and diabetes, affects approximately 720 million people globally (3, 4). Cardiovascular disease, which is strongly linked to diabetes, contributes to over 50% of mortality in this population (5). Epidemiological investigations have consistently established that individuals with diabetes or prediabetes have significantly heightened risks for both all-cause mortality and cardiovascular-specific fatality outcomes (6–8).

Gut microbial composition and functionality exert critical influences on diabetes pathogenesis (9, 10). The gut microbiome can influence systemic inflammation, insulin sensitivity and energy balance through various mechanisms, such as the fermentation of dietary fiber, the production of short-chain fatty acids (SCFA), and the regulation of gut-brain signaling (11–13). A substantial corpus of research has demonstrated that individuals with diabetes or prediabetes have a reduced abundance and functional capacity of bacteria which produce butyrate (14). Dietary factors, particularly fiber intake, strongly regulate butyrate-producing microbial populations, potentially enhancing glycemic control (15). Nutritional patterns are the primary determinants of gut microbiome structure (15). Evidence indicates that dietary-derived microbial metabolites positively modulate insulin secretion, sensitivity, and diabetes incidence (16–19). Dietary modifications alter host microbial communities, cardiometabolic profiles, and immune responses through interconnected pathways (20). Epidemiological studies have confirmed that a nutritionally imbalanced dietary structure characterized by excessive intake of red meat and processed meat products, alcohol abuse, frequent consumption of refined carbohydrates (such as white bread), and sugar-added beverages has a significant dose-dependent association with the prediabetes state and the risk of diagnosed diabetes (21, 22).

Many studies have examined the effects of individual dietary components and their impact on gut microbiota (23, 24). According to Kase et al., the DI-GM was developed based on a large number of literature reviews and indicated that DI-GM is related to indirect biomarkers of intestinal microbiota diversity (25). The DI-GM aims to quantify the effects of nutritional patterns on the structure and metabolic characteristics of the gut microbiome by systematically evaluating the interaction between diet and microbiome. Based on a large number of review articles, 14 types of foods and nutrients have been determined to have beneficial or adverse effects on the gut microbiota. Compared with HEI-2015 and the Mediterranean Diet Score (MDS), the DI-GM focuses on a broader range of gut microbiome attributes (for example, the production of short-chain fatty acids (SCFA), changes in microbiota phyla, and specific bacterial species), providing a more comprehensive assessment of diet-microbiome relationships (25, 26). Malesza et al.‘s research indicates that a high-fat diet is associated with reduced microbial richness and increased intestinal permeability (27). A high-fat diet rich in meat protein may promote different and less diverse populations of sulfur-metabolizing bacteria (28). Recent studies have shown a negative relationship between DI-GM and the risk of developing diabetes (29). In addition, the incidence of other diseases associated with DI-GM, such as stroke (30), chronic kidney disease (31), and depression (32), is negatively correlated. Despite growing evidence on gut microbiota-diet interactions, the longitudinal associations linking the DI-GM to mortality risk in individuals with diabetes or prediabetes have not been fully established. The purpose of our study was to systematically evaluate the prognostic value of DI-GM for all-cause mortality and cardiovascular-specific mortality endpoints in a high-risk population using the NHANES data system.

This study is pioneering in its use of the substantial sample size of the NHANES database to appraise the connection between DI-GM and survival outcomes among people with diabetes or prediabetes. Our analysis revealed an inverse dose–response relationship between DI-GM scores and the risks of all-cause and cardiovascular mortality. This association was found to be highly consistent, suggesting that relatively high DI-GM scores are linked to a lower likelihood of both types of mortality. These results underscore the clinical significance of dietary patterns that enhance microbial diversity in the context of diabetes management and mortality reduction.

Insulin resistance is a critical factor in the development and progression of diabetes and its precursor states (34, 35). Individuals with lower gut microbiome diversity tended to show more significant overall obesity and insulin resistance characteristics (36). An imbalance in intestinal flora may lead to a reduction in the number of beneficial bacteria that produce short-chain fatty acids (SCFAs), especially butyric acid, which is directly related to decreased insulin sensitivity and abnormal blood sugar control (37–39). A plethora of studies have demonstrated that the diversity of the gut microbiome plays a crucial role in maintaining optimal insulin sensitivity and mitigating insulin resistance (40). An imbalance in the gut microbiome can lead to the proliferation of harmful bacteria, such as proteobacteria, which transform choline and carnitine from food into trimethylamine (TMA). Once produced, TMA is further metabolized in the liver to form trimethylamine-N-oxide (TMAO), a compound that substantially elevates the risk of cardiovascular disease by facilitating atherosclerosis, thrombosis, and inflammatory processes (41). In addition, the metabolite of the gut microbiome, TMAO, is positively correlated with all-cause mortality and cardiovascular mortality (42). A positive association has been demonstrated between elevated TMAO levels and the increased probability of developing diabetes (43). Studies have shown a connection between dietary changes and shifts in the gut microbiome (37, 44).

Intestinal diols and lactones in urine are indirect biomarkers of intestinal microbiota diversity. DI-GM was positively correlated with the levels of intestinal diols and lactones in urine, indicating enhanced diversity of the intestinal microbiota. A higher DI-GM score indicates enhanced intestinal microbiota diversity (25). Recently, Wu et al. conducted a cross-sectional study of 21,640 people in the United States and found that higher DI-GM scores were significantly associated with a reduced risk of diabetes (29). However, there is still a scarcity of longitudinal studies that have investigated the issue of mortality in patients diagnosed with diabetes mellitus or prediabetes. According to our results, a pronounced inverse association was observed between DI-GM levels and mortality outcomes, with both all-cause and cardiovascular death risks demonstrating marked reductions (p < 0.05). This suggests that DI-GM is associated with a reduced risk of death in patients with diabetes or prediabetes. The results in RCS demonstrated a negative linear correlation between DI-GM and BGMS for all-cause and cardiovascular mortality (non-linear p > 0.05). Our research results also indicate that UGMS is negatively correlated with all-cause mortality and cardiovascular mortality in patients with diabetes or prediabetes, but the difference was no statistically significant (p > 0.05). Tindall and Shreiner et al. demonstrated that UGMS focuses on components such as red meat and refined grains, which generate uremic toxins such as TMAO through intestinal flora metabolism, activate the NLRP3 inflammatory body, and induce endothelial cell apoptosis (45, 46). In addition, a diet high in fat and sugar but low in fiber, fruits and vegetables is associated with chronic inflammation (47). In our study, there was no significant association between UGMS and mortality. This might be due to the shorter course of diabetes in the high-exposure group of UGMS in the sample, which led to the cumulative effect of chronic inflammation not being fully manifested. Dietary optimization has demonstrated efficacy in fostering the proliferation of beneficial gut microbiota, thereby reducing the risk of death, consistent with the findings of previous studies. The absence of statistical significance in the coarse model may indicate the influence of confounding factors, particularly age. Interestingly, we performed a stratified analysis and discovered that DI-GM exerted a more pronounced effect on the mortality risk among middle-aged and elderly individuals, which might be attributable to age-related decline in metabolic reserves. Zhang et al. discovered that an augmentation in the score of DI-GM was correlated with the risk of developing sarcopenia, a condition that was prevalent among middle-aged and elderly populations and is connected with inadequate metabolic reserves (48). A substantial body of research has demonstrated that the gut microbiota plays a pivotal role in the aging process. Higher DI-GM scores have been found to be associated with a reduced risk of accelerated aging (49, 50). Our results showed a significant interaction between smoking and race in the subgroup analysis. Research indicates that smoking significantly increases the risk of all-cause and cardiovascular mortality in individuals with diabetes. However, smoking cessation is associated with a lower risk and a strong positive correlation with other cardiovascular health outcomes (including heart failure and peripheral artery disease) (51–53). Therefore, it may mask the protective factors of DI-GM, leading to an increased risk of death. The prevalence of diabetes also varies by race, with the highest prevalence among Hispanics/Latinos (54, 55).

DI-GM has many potential applications in clinical practice. By quantifying the impact of diet on the gut microbiota, DI-GM can identify high-risk groups for diabetes, cardiovascular diseases and cancer (56–58). Based on the beneficial and unfavorable components of DI-GM, dietary plans are customized for patients, such as guiding prediabetic patients to adjust the proportion of dietary fiber to reduce the risk of progression (32, 35, 57, 59).

This investigation presents several notable strengths. First, as a novel dietary evaluation metric, DI-GM has been examined for its associations with mortality in populations with diabetes or prediabetes for the first time, to our knowledge. Second, the nationally representative NHANES cohort employed a multistage probability cluster sampling design, ensuring methodological rigor and population generalizability. Third, comprehensive covariate adjustments minimized confounding biases, with sensitivity analyses confirming result robustness. Several limitations warrant consideration. First, inherent methodological constraints exist in DI-GM’s current formulation, particularly regarding the inclusion criteria for emerging dietary components requiring empirical validation. Second, our analysis lacked granular data on food preparation methods, potentially influencing microbiome interactions. Third, NHANES dietary recalls omitted tea subtype specifications, possibly affecting DI-GM scoring accuracy. Future investigations should prioritize validating disputed dietary-microbiome interactions, and elucidating microbiota-mediated dietary influences on metabolic outcomes.

The research discovered that the DI-GM score exhibits an inverse relationship with both all-cause mortality and cardiovascular mortality among individuals with diabetes or pre-diabetes. A greater DI-GM score has been found to correlate with enhanced diversity of the gut microbiota, which is associated with a lower risk of mortality. These results highlight the importance of consuming foods that are beneficial to the gut microbiota in the prevention and control of diabetes and its early stages.