A 1:1 sex-matched case-control study was conducted in Fujian Province, China, involving participants who met predefined inclusion and exclusion criteria. All participants were aged 18–75 years and were residents of Fujian, defined as having lived in the study area for at least 6 months within the 12 months preceding the survey.

The case group consisted of newly diagnosed patients with GC confirmed by histopathological or cytological examination between February and December 2024 at Fujian Medical University Union Hospital. The inclusion criteria for cases were: (1) aged 18–75 years; (2) resident of Fujian Province (residing in the study area for ≥6 months within the past year); (3) able to communicate effectively; and (4) provided written informed consent and voluntarily participated in the study. Exclusion criteria for cases included: (1) history of any cancer; (2) pregnancy or lactation; and (3) extreme daily energy intake (females >3,600 kcal or < 500 kcal; males >4,200 kcal or < 600 kcal).

The control group consisted of healthy residents recruited during the same period from nine prefecture-level cities in Fujian Province, matched by sex. Inclusion criteria for controls were: (1) aged 18–75 years; (2) resident of Fujian Province (residing in the study area for ≥6 months within the past year); (3) able to communicate effectively; and (4) provided written informed consent and voluntarily participated in the study. Exclusion criteria for controls were: (1) any history or current diagnosis of malignant tumors; (2) presence of major diseases (e.g., stroke or psychiatric disorders); and (3) extreme daily energy intake (females >3,600 kcal or < 500 kcal; males >4,200 kcal or < 600 kcal).

The sample size is calculated using the following formula: where M is the total number of pairs to be investigated; m is the number of pairs with inconsistent case-control exposure status, P₀ is the exposure rate of high DII in the control population of about 33%, the expected dietary index-induced exposure to the risk of developing gastric cancer (OR (RR)) is 1.77, Z_1−α/2 is the standardized normal deviation corresponding to the α level, and Z_β is the standardized normal deviation corresponding to the 1-β level, with the stipulation that α = 0.05, The two-sided test with a degree of certainty 1-β of 0.9, checking the table gives Z_1−α/2 = 1.96 and Z_β = 1.28, M = 265 was calculated, and at least 265 pairs of study participants, totalling 530, were needed for this study.

This study referred to the dietary component inflammatory potential scoring method proposed by Shivappa et al. (15). The CHINA-DII calculation process followed the same procedure as the original DII developed by Shivappa et al. The calculation involved five steps, summarized as follows: (1) Dietary intake data of study participants were obtained through dietary questionnaire surveys. (2) For each dietary component, a Z-score representing individual exposure was calculated using the following formula: Z = (individual intake of a dietary component – the mean intake of that component from the Chinese adult dietary intake database)/standard deviation of intake from the Chinese database. (3) To reduce the influence of right-skewed distributions, the calculated Z-scores were then centralized and converted into percentile scores (q) ranging from −1 to +1, with 0 as the midpoint. (4) The dietary inflammatory index score for each dietary component was calculated as follows: CHINA-DII (individual component) = q × i, where “i” represents the literature-derived inflammatory effect score of the dietary component and “q” represents the centralized percentile value. (5) The total CHINA-DII score was obtained by summing the individual component-specific scores as follows: CHINA-DII = i₁ × q₁ + i₂ × q₂ + …+ i_n × q_n.

Continuous variables with a normal distribution were described as means and standard deviations (SD), while those not normally distributed were presented as medians and interquartile ranges (IQR: P25, P75). Categorical variables were expressed as frequencies and percentages (N, %). Between-group comparisons were performed using the chi-square test for categorical variables and the t-test or analysis of variance (ANOVA) for continuous variables, as appropriate.

Participants were categorized into low and high CHINA-DII groups based on the median CHINA-DII score in the control group. Univariate and multivriable logistic regression models were applied to assess the association between CHINA-DII categories and gastric cancer (GC) risk, with odds ratios (ORs) and corresponding 95% confidence intervals (CIs) calculated. Two models were applied to assess the ORs between CHINA-DII and risk of GC: Model 1, unadjusted; Model 2 was adjusted for age group, marital status, smoking, and perceived daily stress level. In addition, CHINA-DII was also analyzed as a continuous variable to evaluate the risk change per one–standard deviation (SD) increase in CHINA-DII score.

To further explore whether the association between CHINA-DII and GC risk varied across subgroups, stratified analyses were conducted based on demographic variables significantly associated with GC risk in univariate analysis. We tested interactions by adding CHINA-DII × subgroup terms to logistic models, with P-interaction determined via likelihood ratio tests. All statistical tests were two-sided, and a P-value < 0.05 was considered statistically significant. Data analysis was performed using IBM SPSS Statistics version 26.0.

This study was conducted in accordance with the principles of the Declaration of Helsinki and was approved by the Ethics Committee of Fujian Medical University (FJMU No. 2020 [53]). Before participation, the purpose and content of the study were fully explained to the patients, and informed consent was obtained. Participants were free to withdraw from the study at any time if they experienced any discomfort, and refusal to participate had no impact on their medical care. All personal information of the participants was kept strictly confidential at all times.

Table 1 presents the baseline demographics of the study population. A total of 672 participants were included in the analysis, comprising 336 gastric cancer cases and 336 matched controls. The mean age of the participants was 55.31 ± 10.83 years, with cases being significantly older than controls (56.76 ± 10.34 vs. 53.86 ± 11.13 years, P < 0.001). Males accounted for 56.5% and females accounted for 44.5%. Most participants were married (90.9%), and over half had an education level of secondary school or below.

There were no significant differences between the case and control groups in terms of sex, education level, occupation, household income, or alcohol consumption (P > 0.05 for all). However, cases were more likely to report moderate or high levels of perceived daily stress (32.4 vs. 48.8%, P < 0.001), to be current smokers (39.0 vs. 30.4%, P = 0.019), and to be married (93.8 vs. 88.1%, P = 0.011). Additionally, the CHINA-DII score was significantly lower in the control group than in the case group (−2.18 ± 0.61 vs. −2.04 ± 0.62, P = 0.003).

As shown in Table 2, there were no statistically significant differences between the case and control groups in total energy, protein, carbohydrate, fat, cholesterol, dietary fiber, and most micronutrient intakes (P > 0.05). However, the intake of vitamin C and vitamin D was significantly lower in the case group than in the control group. Specifically, the mean intake of vitamin C was 98.77 ± 67.89 mg in cases compared to 119.26 ± 81.58 mg in controls (P < 0.001), and the mean intake of vitamin D was 2.10 ± 1.23 μg in cases vs. 2.40 ± 1.14 μg in controls (P < 0.001). No significant differences were observed in the intake of β-carotene, vitamin E, or other vitamins and minerals between the two groups.

As shown in Table 3, in the univariate analysis, higher intakes of dietary fiber, vitamin B6, vitamin C, and vitamin D were inversely associated with GC risk (P < 0.05). However, multivariable logistic regression analysis revealed that only higher intakes of vitamin C and vitamin D were significantly associated with a reduced risk of GC. Individuals in the high-intake group had a 31% lower risk of GC for vitamin C (OR = 0.69, 95% CI: 0.50–0.95, P = 0.023) and a 33% lower risk for vitamin D (OR = 0.67, 95% CI: 0.48–0.92, P = 0.014), compared to those in the low-intake group.

As shown in Table 4, higher CHINA-DII scores were significantly associated with increased GC risk. In the unadjusted model (Model 1), participants in the high CHINA-DII group had a 1.50-fold higher risk of GC compared to those in the low CHINA-DII group (OR = 1.50, 95% CI: 1.11–2.04, P = 0.009). This association remained statistically significant after adjusting for age group, marital status, smoking, and perceived daily stress level in Model 2 (OR = 1.45, 95% CI: 1.05–1.99, P = 0.023). When CHINA-DII was analyzed as a continuous variable, each one SD increase in score was associated with a 1.26-fold higher risk of GC (OR = 1.26, 95% CI: 1.07–1.48, P = 0.006).

Subgroup analyses (Table 5) indicated that the positive association between CHINA-DII and GC risk was more pronounced in certain population groups. Specifically, the association was significant among participants aged ≤ 55 years (OR = 2.44, 95% CI: 1.51–3.96, P < 0.001), married individuals (OR = 1.41, 95% CI: 1.01–1.96, P = 0.044), non-smokers (OR = 1.70, 95% CI: 1.14–2.54, P = 0.009), and those with moderate to high levels of perceived daily stress (OR = 2.82, 95% CI: 1.67–4.75, P < 0.001). No significant associations were found among participants aged >55 years, smokers, those with low stress levels, or those who were single, separated, divorced, or widowed (P > 0.05 for all).

GC is a multifactorial disease influenced by genetic, infectious, environmental, and lifestyle-related factors. Among these, dietary factors have attracted substantial attention in primary prevention due to their modifiability. In this study, 23 dietary nutrients were systematically assessed in relation to GC risk, with findings indicating that higher intakes of vitamin C and vitamin D were significantly associated with reduced risk.

From a biological perspective, vitamin C is a potent antioxidant capable of scavenging free radicals and reducing oxidative stress-induced DNA damage (22). It can also inhibit the endogenous formation of carcinogenic N-nitroso compounds, which play a critical role in gastric carcinogenesis (23). These findings are consistent with previous studies. A meta-analysis of 32 prospective studies reported a 19% reduction in GC risk associated with high vitamin C intake (OR = 0.81, 95% CI: 0.68–0.95), with dose-response analysis suggesting that 65 mg/day might offer optimal protection (24). A case-control study in Korea similarly showed that vitamin C intake was significantly lower among GC patients and inversely associated with GC risk (OR = 0.64, 95% CI: 0.46–0.88) (25).

Vitamin D, particularly its active form 1,25-dihydroxyvitamin D, binds to the vitamin D receptor (VDR) and inhibits tumor cell proliferation while promoting apoptosis (26). A meta-analysis of serum 25(OH)D₃ levels revealed a significant inverse association with GC incidence, suggesting that sufficient vitamin D status may be protective (27). However, the relationship between dietary vitamin D intake and GC risk remains inconclusive, as some reviews report no statistically significant associations (28).

It is worth noting that in this study, the primary dietary sources of vitamin D were fish, eggs, and red meat. Given that the study population was based in coastal Fujian Province, where fish intake tends to be higher, this regional dietary pattern may have influenced the observed association. Future large-scale, high-quality, multicenter studies across diverse geographic regions are needed to validate the protective role of vitamin D in GC prevention.

The DII has emerged as a comprehensive indicator of an individual's dietary inflammatory potential and has been implicated in the development of inflammation-related cancers, including GC. Our study found that higher CHINA-DII scores, reflecting more pro-inflammatory diets, were significantly associated with increased GC risk, suggesting that pro-inflammatory dietary patterns may play a critical role in gastric carcinogenesis.

Mechanistically, a high DII score typically reflects a diet rich in pro-inflammatory components such as saturated fats, sugars, and red meats, which can stimulate the production of inflammatory cytokines like TNF-α and IL-6 (29). These cytokines interact with stromal cells, recruit additional inflammatory cells, and maintain a chronic inflammatory microenvironment conducive to tumor proliferation, invasion, angiogenesis, and metastasis. Furthermore, chronic inflammation may reduce the effectiveness of anticancer therapies by altering drug metabolism or vascular permeability and suppressing antitumor immune surveillance, thereby increasing the risk of GC (30).

Several epidemiological studies support a positive association between DII and GC. A prospective cohort study involving over 100,000 participants found a linear relationship between DII and GC risk [OR per tertile decrease in DII in men: 0.73 (0.53–0.99)] (18). Case-control studies from Korea (OR = 1.47, 95%CI: 1.16–1.85) (31), Iran (OR = 3.39, 95%CI: 1.59–7.22) (32), and Brazil (OR = 2.60, 95%CI: 1.16–5.70) (33) (2018–2022) have consistently demonstrated that higher DII scores were associated with significantly higher risk of gastric ulcer or GC. The EPIC study further reported that each one-SD increase in DII was associated with a 25% increase in GC risk, with those in the highest DII quantile having a 1.66-fold higher risk (OR = 1.66, 95%CI: 1.26–2.20) than those in the lowest (17).

Stratified analyses in this study revealed potential population heterogeneity in the association between CHINA-DII and GC risk. The association was pronounced in individuals aged ≤ 55 years, consistent with findings from a case-control study in Italy (34). This may reflect a heightened susceptibility of the younger gastric mucosa to dietary inflammatory insults or the greater role of diet in early precancerous processes among those without existing structural abnormalities. Animal studies also suggest that younger organisms exhibit stronger inflammatory responses, possibly due to a less mature mucosal barrier (35). Gastric epithelial cells in younger populations have a faster rate of renewal, and this hyperproliferative state may make proliferating cells more susceptible to oxidative damage when exposed to pro-inflammatory diets, leading to an accumulation of DNA repair errors (36). Moreover, younger individuals may be more likely to adopt high-calorie, high-fat, and processed food diets with stronger pro-inflammatory potential (37, 38). For the younger population, early screening for gastric cancer is recommended, combined with dietary assessment for early identification of high-risk individuals.

Among non-smokers, the association between CHINA-DII and GC risk was also stronger, suggesting that in the absence of a dominant carcinogenic exposure such as smoking, pro-inflammatory diets may exert a more independent effect. In contrast, the strong pro-inflammatory and carcinogenic nature of smoking may mask the marginal effects of diet among smokers (39).

The association was also more prominent among married individuals and those with higher levels of perceived daily stress. Married individuals may have more stable and representative long-term dietary habits (40). Differences in gut microbiota diversity were lower in cohabiting individuals (e.g. mates) than in genetically related separated individuals, suggesting that the shared environment drives microbial convergence and that taxa involved in dietary fiber fermentation are more affected by this effect (41). Psychological stress has been shown to enhance inflammatory responses through activation of the hypothalamic-pituitary-adrenal (HPA) axis and related inflammatory pathways (42), potentially amplifying the negative effects of pro-inflammatory diets. Therefore, it is recommended that those with high levels of perceived stress be screened in conjunction with diet and nutritional interventions implemented accordingly.

These findings suggest the need for more targeted and personalized dietary interventions for GC prevention, particularly among younger adults, those experiencing high psychological stress, and non-smokers with unhealthy dietary patterns. From a public health perspective, CHINA-DII-based dietary strategies can provide dietary guidance to workplace wellness programmes and university health services, facilitate the integration of anti-inflammatory dietary education with mental health services in community clinics, and provide targeted guidance to high-risk populations to increase awareness of anti-inflammatory diets.

This study has several limitations. First, as a case-control study, dietary data were retrospectively collected via food frequency questionnaires (FFQs), which may introduce recall bias—especially among patients who might over-report unhealthy dietary behaviors. Second, Helicobacter pylori infection, a critical confounder in GC research, was not assessed and may have influenced risk estimates. Its absence in our analysis means we cannot determine whether the observed associations between dietary inflammation and GC are independent of Helicobacter pylori infection, or if they might be amplified/attenuated in its presence. While this is a common limitation in nutritional epidemiology studies, future research should ideally combine dietary assessments with Helicobacter pylori testing to clarify these relationships. Third, due to incomplete clinical data, GC was not analyzed by histological subtype, limiting the specificity of our findings.