Introduction

The global incidence of malignant tumors is on the rise, with a notable increase in gastrointestinal (GI) cancers such as esophageal, gastric, liver, pancreatic and colorectal cancers1. These types of cancer are major contributors to cancer incidence and mortality worldwide, significantly impacting human health. Gastric cancer stands out as one of the primary GI malignancies, ranking as the second leading cause of cancer-related deaths, as reported by the World Health Organization2. Meanwhile, colorectal cancer ranks as the fourth most prevalent cancer globally and the third leading cause of cancer-related mortality3. Studies indicate that GI tumors are influenced by both environmental factors—including smoking, diet, and obesity—and individual factors such as genetic predisposition, age, and gender4. The World Cancer Research Fund/American Institute for Cancer Research suggests that up to 50% of digestive tract cancers could be prevented through lifestyle and behavioral modifications5. Research findings affirm that consuming red and processed meats increases the risk of colorectal cancer6, whereas intake of dairy products and dietary fibers may decrease the risk of GI cancers7,8. Poor dietary habits, characterized by low consumption of fruits and vegetables and high consumption of red meats and fats, are linked to increased oxidative stress and apoptosis, promoting tumor growth and spread9,10.

In recent years, dietary research has shifted from examining individual foods or nutrients to adopting a more comprehensive approach focused on dietary patterns. Dietary pattern analysis explores the combined effects of multiple foods and nutrients, reflecting the complexity of real-world diets11. Compared to studies focused on isolated foods or nutrients, dietary pattern analysis offers a more accurate prediction of disease risk12. Studies have found that both the Healthy Eating Index (HEI-2020) and alternative Mediterranean Diet (aMED) are associated with reduced GI cancers risk. For example, previous studies have demonstrated that high adherence to the Mediterranean diet is associated with a reduced risk of colorectal and gastric cancers13. Similarly, studies on the HEI-2020 have shown associations between high HEI scores and reduced cancer risks, including some GI cancers, due to adherence to dietary guidelines that emphasize fruits, vegetables, and whole grains while limiting processed meats and added sugars14,15. However, comparisons between HEI-2020 and aMED scores remain limited.

HEI-2020, developed by the U.S. Department of Health and Human Services and the USDA, assesses diet quality based on adherence to the Dietary Guidelines for Americans16,17. In contrast, the Mediterranean Diet (MED) scores focus on high intake of vegetables, fruits, whole grains, and olive oil, with minimal red and processed meats, which are linked to lower cancer risk due to antioxidants like polyphenols18,19,20. The aMED adapts this pattern for Western diets, allowing a broader application beyond the Mediterranean context21.

Additionally, biomarkers such as uric acid and albumin may play roles in cancer risk.

Uric acid acts as an antioxidant at low levels but promotes oxidative stress and inflammation at high levels, potentially contributing to tumor development22,23,24. Albumin reflects both nutritional status and systemic inflammation, with low levels associated with higher cancer risk25,26. Healthy dietary patterns, indicated by high HEI-2020 and aMED scores, generally correlate with higher albumin and lower inflammatory markers, potentially aiding cancer prevention27,28.

While the MED is renowned for its health benefits and its role in preventing various chronic diseases, dietary habits and lifestyles differ markedly across populations, underscoring the need for further research into other dietary indices. The HEI-2020, an advanced and comprehensive dietary assessment tool, has yet to be extensively explored in cancer prevention contexts. This study addresses this gap by evaluating the associations between HEI-2020 and the aMED with GI cancer risk, offering new insights into the impacts of these dietary patterns on cancer prevention. Using data from the 2005–2018 National Health and Nutrition Examination Survey (NHANES), the study also examines the potential mediating effects of serum albumin and uric acid. Understanding how these dietary patterns and biomarkers influence GI cancers development can offer valuable insights for cancer prevention and dietary guidance.

Methods

Study population

This study analyzed data from the NHANES spanning 2005 to 2018. NHANES is a comprehensive cross-sectional survey that evaluates the health and nutritional status of both adults and children in the U.S. every two years. The survey uses a multistage stratified sampling method and includes modules on demographics, diet, physical exams, lab tests, and questionnaires. All methods were performed in accordance with the relevant guidelines and regulations. Information on the NHANES database can be accessed on the official website (https://www.cdc.gov/nchs/nhanes/index.htm).

From the data covering seven cycles from 2005 to 2018, a total of 70,190 participants were initially considered. Exclusions were made for 19,268 participants due to incomplete data for calculating the HEI-2020, 1,346 for missing aMED scores, and 23,256 for absent covariate data. Ultimately, 26,320 participants were selected for the study, consisting of 12,722 males and 13,598 females. Additional details are depicted in Fig. 1.

Fig. 1
figure 1

Flowchart for the Study Population: : From NHANES 2005–2018.

Full size image

Assessment of HEI-2020 and aMED scores

Dietary intake data for this study were obtained from the NHANES database using a 24-hour dietary recall questionnaire administered twice—first through face-to-face interviews and then via telephone within ten days. This approach ensured comprehensive data on the types and quantities of food consumed, along with the times and locations of consumption. The assessments were conducted by trained researchers at Mobile Examination Centers.

The HEI-2020 utilizes data from these dietary recalls combined with the USDA’s food pattern equivalents database. Using a scoring algorithm provided by the National Cancer Institute, the HEI-2020 evaluates diet quality based on thirteen components divided into two categories: adequacy and moderation. Adequacy components, which include vegetables, fruits, whole grains, protein foods, dairy, and healthy fats, reward higher intakes with higher scores. Moderation components, targeting intake of added sugars, sodium, refined grains, and saturated fats, assign higher scores for lower intakes17.

Adherence to the Mediterranean diet was quantified through the aMED scores. This process involved two steps: first, linking the 24-hour dietary recall data to the USDA’s Food Patterns Equivalents Database to standardize food component measurements29. Second, calculating the aMED scores based on intake levels of key food components like fruits, vegetables, red and processed meats, whole grains, alcohol, nuts, legumes, fish, and the ratio of monounsaturated to saturated fatty acids30. For this scoring, intakes of red and processed meats below the median, and moderate alcohol consumption, are each awarded one point; intakes of the other foods above the median also receive one point, with higher aMED scores indicating greater adherence to the MED.

Definition of GI cancers

In this study, GI cancers were identified based on responses to the survey question, “What kind of cancer?” Only responses explicitly indicating the presence of GI tumors were considered valid. The study focused on six types of GI cancers from the NHANES database: esophageal cancer, gastric cancer, liver cancer, pancreatic cancer, colon and rectal cancer. Detailed information on cancer types and tHEI-2020r distribution can be found in Supplementary Table 1. For more detailed information, please refer to the NHANES 2005–2006 MCQ section on the official website (https://wwwn.cdc.gov/Nchs/Nhanes/2005-2006/MCQ_H.htm#MCQ230a).

Assessment of serum albumin and uric acid

In the NHANES study, participants’ fasting serum samples were stored at -30 degrees Celsius and sent to the National Center for Environmental Health. Trained researchers measured serum albumin concentration (g/dL) using the bromocresol purple dye-binding method and determined serum uric acid levels (mg/dL) using the uricase oxidation method. Both measurement processes underwent strict calibration and quality control to ensure accuracy and consistency. More specific details can be seen in https://wwwn.cdc.gov/Nchs/Nhanes/.

Covariates

The study used several covariates such as confounding factors to accurately assess the relationship between diet quality and GI cancers. These included gender, defined as male or female, and age, which was treated as a continuous variable. Other variables included body mass index (BMI) categorized as < 20, 20–25, 25–30, and ≥ 3031; education level categorized as less than high school, high school, and college or above32; race/ethnicity including Mexican American, Non-Hispanic White, Non-Hispanic Black, and Other; alcohol consumption; smoking status; poverty income ratio (PIR) categorized as < 1, 1–3, and ≥ 333; physical activity levels classified as Moderate, Vigorous, and Inactive34; hypertension, and diabetes. PIR refers to the ratio of monthly family income to a poverty threshold specific to family size. Smoking is defined as smoking at the time of the survey or having smoked more than 100 cigarettes in one’s lifetime35. Alcohol consumption is defined as drinking at the time of the survey or having consumed more than 12 alcoholic beverages in one’s lifetime36. Hypertension was identified by a physician-diagnosed condition, elevated average blood pressure (systolic ≥ 130 and/or diastolic ≥ 85 mmHg), or use of antihypertensive medication37. Diabetes was defined by self-reported physician-diagnosed diabetes, use of insulin or hypoglycemic drugs, fasting blood glucose > 126 mg/dL, glycated hemoglobin ≥ 6.5%, or an oral glucose tolerance test result ≥ 200 mg/dL38.

Statistical analysis

The baseline characteristics of participants grouped by HEI-2020 and aMED scores were described using means (standard deviation [SD]) for continuous variables and counts (percentage) for categorical variables. The Wilcoxon rank-sum test and t-test were employed to analyze differences in continuous variables between groups, while the chi-square test was used for categorical variables. The association was evaluated across tertiles of adherence to HEI-2020 and aMED scores. Logistic regression models were developed to examine the association between HEI-2020, aMED scores (as continuous and categorical variables), and GI cancers, reporting odds ratios (ORs) and 95% confidence intervals (CIs). Model 1 adjusted for age, gender, and race. Model 2 included further adjustments for educational level, alcohol consumption, smoking status, PIR, physical activity, hypertension, and diabetes. These variables were selected based on established risk factors for GI cancers identified in previous epidemiological studies. Additionally, we developed the directed acyclic graphs (DAG) to systematically identify potential confounders and guide model adjustment. More specific details can be seen in Supplementary Figs. 1–2. Restricted cubic spline (RCS) models evaluated the dose-response relationship between HEI-2020-2020, aMED, and GI cancers. Mediation analysis was conducted to determine whether serum albumin and uric acid mediated the relationship between dietary quality indices (HEI-2020 and aMED) and GI cancers. The total effect (TE) represents the direct relationship between dietary quality indices and GI cancers, uninfluenced by mediators. The indirect effect (IE) refers to the impact of dietary quality indices on GI cancers through serum albumin and uric acid. The direct effect (DE) represents the impact of dietary quality indices on GI cancers after controlling for levels of serum albumin and uric acid. A significant IE indicates a notable mediation effect, calculated as IE/TE*100%. Additionally, subgroup analyses were conducted based on age, gender, race, educational level, BMI, PIR, hypertension, and diabetes, and interactions were assessed using the p-values of product terms between HEI-2020, aMED, and each stratifying factor.

All statistical analyses and graphics were performed using the R Project for Statistical Computing (version 4.2.3) and RStudio. All tests were two-sided, and a p-value < 0.05 was considered statistically significant.

Results

Baseline characteristics

The study analyzed 26,320 participants, with demographic characteristics presented in Tables 1 and 2. Participants’ HEI-2020 were categorized into tertiles: Tertile 1 (0-45.39), Tertile 2 (45.39–56.17), and Tertile 3 (56.17–94.19)16, while their aMED scores were similarly grouped into tertiles: Tertile 1 (0–4), Tertile 2 (4–5), and Tertile 3 (≥ 6). Participants in the highest tertile of adherence to HEI-2020 and aMED were more likely to be female, older, have a BMI ≥ 20, hold a college degree or higher, belong to other races, be drinkers and smokers, have a PIR > 3, have hypertension, and lower uric acid levels compared to those in the lowest tertiles. Additionally, those in the highest HEI-2020 tertile were more likely to engage in moderate physical activity, have diabetes, and higher albumin levels, while those in the highest aMED tertile were more likely to engage in vigorous physical activity.

Table 1 Characteristics of the Study Population by HEI-2020 level.
Full size table
Table 2 Characteristics of the Study Population by aMED score level.
Full size table

Association between HEI-2020, aMED scores and GI cancers

As shown in Table 3, in Model 2, continuous HEI-2020 continued to show a significant negative correlation with the risk of GI cancers (OR 0.84, 95% CI 0.71 to 0.98, p = 0.028). Compared with the lowest tertile of adherence to the HEI-2020, the highest tertile was associated with a 30% lower risk of GI cancers (OR = 0.70, 95% CI: 0.50 to 0.98, p = 0.037).

Table 3 The Association between HEI-2020, aMED scores and gastrointestinal cancers in NHANES 2005–2018.
Full size table

As shown in Supplementary Table 2, in Model 2, in the female subgroup, the continuous HEI-2020 was significantly negatively correlated with the risk of GI cancers (OR 0.99, 95% CI 0.97 to 1.00, p = 0.030). In males, the continuous HEI-2020 was significantly negatively correlated with the risk of GI cancers (OR 0.98, 95% CI 0.97 to 1.00, p = 0.020). Compared with the lowest tertile of adherence to the HEI-2020, the highest tertile was associated with a statistically significant 35% reduction in GI cancers risk (OR = 0.65, 95% CI: 0.42–1.00, p = 0.050).

Participants were also grouped by aMED scores into tertiles: Tertile 1 (aMED scores 0–4), Tertile 2 (aMED scores 4–5), and Tertile 3 (aMED scores ≥ 6). In Model 2, the continuous aMED scores was also significantly negatively correlated with the risk of GI cancers (OR 0.80, 95% CI 0.67 to 0.96, p = 0.014), compared with the lowest tertile of adherence to the aMED scores, the highest tertile was associated with a 37% lower risk of GI cancers (OR 0.63, 95% CI 0.44 to 0.92, p = 0.014).

As shown in Supplementary Table 3, in Model 2, for females, the continuous aMED scores was significantly negatively correlated with the risk of GI cancers (OR 0.83, 95% CI 0.68 to 1.00, p = 0.050). Compared with the lowest tertile of adherence to the aMED scores, the highest tertile was associated with a 32% lower risk of GI cancers (OR 0.68, 95% CI 0.41 to 0.98, p = 0.020). For males group, the continuous aMED scores was also significantly negatively correlated with the risk of GI cancers (OR 0.83, 95% CI 0.69 to 1.00, p = 0.050).

Nonlinearity analysis using RCS

This study utilized an RCS model to analyze the dose-response relationship between dietary quality indices and GI cancers, setting knots at the 5th, 35th, 65th, and 95th percentiles, with the 5th percentile as the reference point. The multivariable model was adjusted using the same covariates as previous analyses. Results indicated a significant overall association between both HEI-2020 and GI cancers and aMED and GI cancers; however, the nonlinearity in these relationships was not significant (HEI-2020: p for overall association = 0.036, p for nonlinearity = 0.723; aMED: p for overall association = 0.033, p for nonlinearity = 0.742). Detailed visual representations are provided in Fig. 2.

Fig. 2
figure 2

The Association between HEI-2020, aMED Score, and Gastrointestinal Cancers Using the Restricted Cubic Spline Regression Model. (A), HEI-2020; (B), aMED Score. Graphs show ORs for end according to HEI-2020, aMED Score adjusted for sex, age, ethnicity, education level, drink, smoke, PIR, Hypertension, Diabetes. Data were fitted by a logistic regression model, and the model was conducted with 4 knots at the 5th, 35th, 65th, 95th percentiles of HEI-2020, aMED Score (reference is the 5th percentile). Solid lines indicate ORs, and shadow shape indicate 95% CIs. HEI, Healthy Eating Index; aMED, Alternative Mediterranean Diet; OR, odds ratio; CI, confidence interval.

Full size image

Subgroup analysis

Subgroup analyses revealed a consistently negative correlation between HEI-2020 scores and GI cancer risk across various demographics. Notably, as shown in Fig. 3A, stronger negative correlations were observed in subgroups aged over 60 years (OR 0.83, 95% CI 0.69 to 0.99), Non-Hispanic Whites (OR 0.82, 95% CI 0.67 to 1.00), those with a college education or higher (OR 0.76, 95% CI 0.61 to 0.95), PIR between 1 and 3 (OR 0.78, 95% CI 0.62 to 0.98), individuals with hypertension (OR 0.75, 95% CI 0.62 to 0.91), and non-diabetics (OR 0.81, 95% CI 0.67 to 0.98). No interaction effects were detected regarding age, gender, race, education level, BMI, PIR, hypertension, or diabetes.

Fig. 3
figure 3

Subgroup Analyses of the Association between HEI-2020, aMED Score, and Gastrointestinal Cancers. (A), HEI-2020; (B), aMED Score. HEI, Healthy Eating Index; aMED, Alternative Mediterranean Diet;

Full size image

Similarly, as shown in Fig. 3B, a negative correlation between aMED Scores and GI cancer risk was consistent across all subgroups, with more pronounced effects among those with a college education or higher (OR 0.80, 95% CI 0.67 to 0.96), a BMI of 25–30 (OR 0.78, 95% CI 0.64 to 0.96), a PIR of 1–3 (OR 0.81, 95% CI 0.67 to 0.98), hypertensive individuals (OR 0.85, 95% CI 0.73 to 1.00), and non-diabetics (OR 0.82, 95% CI 0.71 to 0.96). Again, no interaction effects were observed.

Mediation analysis

The study conducted a mediation analysis to investigate whether serum uric acid and albumin levels mediate the relationship between dietary quality indices and GI cancer risk. The analysis used HEI-2020 and aMED Scores as independent variables, GI cancers as the dependent variable, and serum uric acid and albumin as mediators.

For HEI-2020, serum uric acid levels exhibited a significant mediating effect, accounting for 1.71% of the association. The IE was − 0.00002 (95% CI: -0.00003 to -0.00000), as depicted in Fig. 4A. For aMED scores, uric acid levels also showed a significant mediating role, representing 3.13% of the association, with an IE of -0.0001 (95% CI: -0.0002 to -0.0000), detailed in Fig. 4C.

Fig. 4
figure 4

Mediation Analyses of the Association between HEI-2020, aMED Score and Gastrointestinal Cancers. (A), serum uric acid, (B), serum albumin partially mediates the relationship between HEI-2020 and GI Cancers; (C), serum uric acid, (D), serum albumin partially mediates the relationship between aMED Score and GI Cancers; HEI, Healthy Eating Index; aMED, Alternative Mediterranean Diet;

Full size image

Regarding HEI-2020, serum albumin levels had a notable mediating effect, explaining 14.29% of the relationship, with an IE of -0.0001 (95% CI: -0.0002 to -0.0000), as shown in Fig. 4B. Similarly, for aMED scores, albumin levels significantly mediated the relationship, accounting for 22.58% of the association, with an IE of -0.0001 (95% CI: -0.0003 to -0.0000), illustrated in Fig. 4D. Detailed representations of these mediating effects are provided in Fig. 4.

Discussion

This study utilized data from NHANES 2005–2018 to examine the relationships between HEI-2020, aMED scores, and GI cancers. The findings indicate that higher HEI-2020 and aMED scores are significantly associated with reduced risks of GI cancers. Mediation analyses suggest that serum albumin and uric acid levels partially mediate these relationships. Subgroup analyses and interaction results reinforce the consistency of these effects across different genders, races, ages, educational levels, BMIs, hypertension, and diabetes, enhancing the reliability of our conclusions.

GI cancers pose a growing global health challenge3. Identifying modifiable risk factors, such as diet quality, is crucial for effective prevention and management strategies. Although the MED is renowned for its health benefits, dietary patterns vary widely across different populations, underscoring the need for further research into other dietary indices. This study contributes new insights into how dietary quality affects GI cancer risks and the roles of serum albumin and uric acid as mediators. These findings highlight the importance of diet quality in cancer prevention and can inform personalized dietary guidance and public health policies.

Our results align with previous studies emphasizing the benefits of high dietary quality in preventing GI cancers39. For instance, a systematic review associated high-quality diets with reduced risks of upper digestive tract cancers40. Furthermore, studies like the Newfoundland study and the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort study have demonstrated that diets rich in anti-inflammatory components correlate with lower cancer incidences10,41. These studies collectively support the notion that enhancing diet quality can significantly mitigate cancer risks. In existing literature, individuals with higher HEI and MED scores generally exhibit a lower risk of GI cancers, particularly when consuming adequate amounts of vegetables, fruits, whole grains, and olive oil13,42,43. These studies collectively support the notion that improving diet quality can substantially reduce cancer risk.

Both HEI-2020 and aMED scores prioritize the intake of fruits and vegetables, known for tHEI-2020r rich antioxidant and anti-inflammatory properties. Antioxidants, such as vitamin C, vitamin E, and β-carotene and polyphenolic compounds, help reduce free radical formation and oxidative DNA damage and suppress inflammatory responses, thus lowering cancer risk44. Anti-inflammatory components can mitigate chronic inflammation, further decreasing tumor risks45. Moreover, these dietary indices advocate for reduced intake of red and processed meats, known for their association with higher cancer risks46, and suggest alternative protein sources like fish, legumes, and nuts47. Reducing intake of nitrates and other carcinogens commonly found in these meats is also emphasized to decrease cancer risks48.

Compared to aMED scores, the relationship between HEI-2020 scores and the risk of digestive tract tumors offers significant protection in certain subgroups; however, the overall impact of aMED scores is less pronounced than that of HEI-2020. When analyzing the two dietary indices both as continuous variables and in tertiles through logistic regression, the protective effect of HEI-2020 is found to be more consistent and pronounced across all levels. This consistency may stem from HEI-2020’s inclusion of a wider array of food categories and nutrients, which enables a more thorough evaluation of dietary quality. The breadth and depth of HEI-2020 facilitate a more reliable reflection of dietary quality changes at various levels, thereby revealing a stronger and more consistent protective effect in the highest tertile. This effect should be credited to the comprehensive nature of HEI-2020, alongside considerations of sample size and statistical power, and its superior capability in assessing dietary quality comprehensively49. In comparison, while aMED also exhibits significant protective effects in the highest tertile, its consistency across different levels does not match that of HEI-2020, leading to a more pronounced protective impact of HEI-2020 against digestive tract tumors50.

Our mediation analysis reveals that serum albumin plays a significant role in mediating the relationship between HEI-2020 scores and GI cancers, accounting for 14.29% of the total effect25. As an acute-phase protein, albumin levels typically decrease during inflammation and oxidative stress, which are conditions often mitigated by a healthy diet rich in protein. This helps maintain higher serum albumin levels, reducing inflammation and oxidative stress. Conversely, uric acid, a byproduct of purine metabolism, is generally linked to higher risks of metabolic syndrome, cardiovascular diseases, and certain cancers51,52. It is known to contribute to oxidative stress by activating enzymes like NADPH oxidase, which promotes the production of reactive oxygen species and stimulates tumor cell proliferation53. These dietary patterns, which recommend high dietary fiber, low fat, and rich antioxidants, help limit uric acid production, thereby aiding in the reduction of inflammation and oxidative stress and potentially inhibiting tumor growth. These findings underscore the importance of comprehensive dietary assessments and their potential implications in understanding the mechanisms behind GI tumor development. Future research should focus on elucidating the complex interactions between healthy dietary patterns and these mediating factors, enhancing our strategies for the prevention and management of GI tumors.

The subgroup analysis results indicate that the protective effect of HEI-2020 on gastrointestinal cancer is more pronounced in men, while it does not reach statistical significance in women. This difference may be attributed to physiological factors, including the effects of sex hormones (particularly estrogen) and variations in oxidative stress and antioxidant capacity. First, estrogen plays a significant antioxidant and anti-inflammatory role in women, helping reduce oxidative stress and regulate the production of pro-inflammatory cytokines54. This protective effect may partially substitute for the antioxidant benefits of a high-quality diet, potentially leading to a reduced additional benefit from dietary improvements in women55. Furthermore, studies have shown that women generally exhibit higher activity levels of antioxidant enzymes, such as superoxide dismutase (SOD) and glutathione peroxidase (GPx), compared to men, which further enhances their resistance to oxidative stress56.

This study leverages the HEI-2020, a comprehensive tool reflecting the “2020–2025 Dietary Guidelines for Americans.” It covers a broad spectrum of food categories and nutrients, such as fruits, vegetables, whole grains, proteins, dairy, healthy fats, added sugars, and sodium, enabling a detailed assessment of diet quality57. The use of an RCS model to examine the dose-response relationship between diet quality and GI tumors reveals a significant negative correlation, although nonlinearity in this relationship warrants further study. Additionally, this research initially explores the mediating roles of serum albumin and uric acid, suggesting these biomarkers could be critical pathways through which diet quality impacts the development and progression of GI tumors. The data sourced from the nationally representative NHANES database enhances the robustness and generalizability of our findings, strengthening the potential impact on public health.

This study still has some limitations. First, the cross-sectional design limits the ability to establish causality between dietary quality and GI tumor risk. While there is a significant association between higher diet quality and reduced tumor risk, the direct influence of diet on tumor development remains unclear. Future research should include prospective cohort studies to assess these temporal relationships more accurately. A key methodological consideration pertains to the inherent limitations of NHANES dietary assessment. While the survey employs the validated USDA Automated Multiple-Pass Method, the 24-hour dietary recall methodology is subject to potential measurement error and various forms of bias58,59. These include recall bias, social desirability bias in reporting perceived healthy or unhealthy foods, and challenges in portion size estimation60. Although NHANES implements a second 24-hour recall for a subset of participants to account for within-person variation, this approach may incompletely capture long-term dietary patterns relevant to cancer risk. Nevertheless, validation studies have demonstrated reasonable correlations between NHANES dietary recalls and objective biomarkers for specific nutrients, supporting the utility of these data for population-level dietary assessment61. Moreover, despite adjusting for various factors like age, gender, race, education level, alcohol consumption, smoking status, PIR, physical activity, hypertension, and diabetes in our analyses, other potential confounders such as family history, genetic predispositions, and gut microbiome interactions remain unexplored. Addressing these in future studies through polygenic risk scoring, metabolomics and proteomics could provide a more comprehensive understanding of the intricate links between diet quality and GI tumors.

To strengthen future research in this area, we recommend combining self-reported dietary data with objective measures such as nutritional biomarkers, implementing more frequent dietary assessments, and utilizing emerging technologies for dietary assessment that may reduce recall bias. Furthermore, validating our findings in prospective cohort studies with longer follow-up periods would help establish the temporal relationship between dietary patterns and GI cancer risk more definitively.

Conclusion

This study underscores the significance of diet quality, as assessed by HEI-2020 and aMED scores, in the prevention and management of GI cancers. The HEI-2020, with its comprehensive dietary evaluation, demonstrates potential in digestive tract tumor prevention and warrants further research. Healthcare providers and public health initiatives should enhance awareness around diet quality, as well as develop refined dietary assessment tools to uncover further preventive capabilities. Given diverse dietary habits across cultures, tailoring recommendations to various ethnic groups poses challenges that may impact compliance. Future health education efforts must promote healthy eating practices through customized strategies. Additionally, longer follow-ups and larger cohorts in future studies would strengthen these findings, while exploring the mechanisms of dietary impact on tumor risks could guide tailored dietary guidelines and policies.