Association between serum 25(OH)D, oxidative balance score, and mortality among individuals with metabolic syndrome: a cohort study Authors: Xuena Zhang, Xiaobing Feng, Jie Liu, Xiaohui Zhang, Qian Zhou, Yu Xu, Wenjuan Wang, Wanjun Luo Published: 2025 ABSTRACT AbstractObjectiveTo investigate the separate, joint, as well as interactive associations of serum 25-hydroxyvitamin D [25(OH)D] concentrations and oxidative balance score (OBS) with mortality among individuals with metabolic syndrome (MetS).MethodsThe analysis included 12,078 participants with MetS from the National Health and Nutrition Examination Survey (NHANES) 2001–2018. Mortality was ascertained by linkage to National Death Index records through 31 December 2019.ResultsDuring 99,690 person-years of follow-up, 2357 deaths were documented, including 654 cardiovascular disease (CVD) deaths and 518 cancer deaths. A pronounced “L-shaped” nonlinear relationship was observed between 25(OH)D and mortality; the hazard ratio (HR) [95% confidence interval (CI)] for 25(OH)D ≥ 75.0 vs. < 50.0 nmol/L (reference) were 0.71 (0.61, 0.82), 0.65 (0.48, 0.86), and 0.77 (0.57, 1.05) for all-cause, CVD, and cancer mortality, respectively. A reverse linear relationship was demonstrated between OBS and mortality; the HRs (95% CI) for high OBS vs. low OBS (reference) were 0.76 (0.68, 0.86), 0.70 (0.58, 0.85), and 0.75 (0.60, 0.94) for all-cause, CVD, and cancer mortality, respectively. In the joint analyses, the combination of 25(OH)D ≥ 75.0 nmol/L and high OBS was associated with the lowest risk of all-cause (HR 0.57, 0.46–0.70) and CVD mortality (HR 0.48, 0.34–0.69). In contrast, participants with 25(OH)D levels of 50.0-74.9 nmol/L and high OBS presented the lowest risk of cancer mortality (HR 0.52, 0.34–0.81). A significant synergistic additive interaction between OBS and sufficient 25(OH)D levels on CVD mortality was observed [relative excess risk due to interaction (RERI) = 0.29, 95% CI: 0.02–0.57], with 47% of the total protective effect attributable to their interaction.ConclusionsAdequate 25(OH)D and higher OBS are significantly associated with lower risk of mortality and exhibit enhanced protective effects on CVD mortality risk.Graphical abstract Supplementary InformationThe online version contains supplementary material available at 10.1186/s13098-025-02058-z.Keywords: 25-hydroxy vitamin d, Oxidative balance score, Joint effects, Mortality, Metabolic syndrome Introduction Metabolic syndrome (MetS), a constellation of metabolic risk factors (e.g., insulin resistance, atherogenic dyslipidemia, central obesity, and hypertension), has become a global public health problem with over one billion people affected globally in 2018 [1, 2]. Nearly one-third of adults were estimated to have MetS in the US [3–5]. MetS is associated not only with risk of developing chronic metabolic diseases (such as CVD and diabetes) but also with elevated long-term specific mortality [6, 7]. It is well established that besides genetic factors (such as variants in genes like PNPLA3), environmental factors like lifestyle and dietary habits play a role in this disorder [8, 9]. Therefore, identifying modifiable factors is crucial for reducing these risks and improving overall health outcomes with MetS. Previous studies have shown that insufficient vitamin D is associated with a higher risk of MetS [10, 11]. A prospective study involving 59 non-diabetic, overweight, and obese Saudi adults demonstrated that regular sun exposure and increased consumption of vitamin D-rich foods contributed to a reduction in the prevalence of high triglycerides and hypertension [12]. Meanwhile, a longitudinal study of 6,682 adults found that improving vitamin D status was associated with a significantly lower risk of MetS [13]. Another study showed that vitamin D insufficiency and MetS predict elevated all-cause and CVD mortality [14]. Recent data further suggest that higher vitamin D status was associated with lower all-cause and cancer mortality in individuals with MetS aged ≥ 40 years [15]. As individuals with MetS aged 40 years and older, this restricts the ability to generalize the study’s findings to the general population. Moreover, dietary and lifestyle pro-oxidant/antioxidant exposure has been measured by the oxidative balance score (OBS) [16]. Several studies have determined that higher OBS was associated with lower MetS risk [15, 17]. A cohort study with diabetes and pre-diabetes revealed a protective association between higher OBS and lower all-cause and CVD mortality [16]. Similarly, an analysis of 3,621 participants with MetS showed that elevated OBS was inversely associated with all-cause mortality [18]. However, these findings should be interpreted with caution due to several limitations, including relatively small sample sizes and a limited number of deaths (n = 181). Furthermore, no studies have investigated the relationship between OBS and cause-specific mortality (e.g., CVD and cancer) among individuals with MetS. Importantly, the potential synergistic effects of vitamin D status and OBS on the risk of mortality among individuals with MetS remain unexplored. Utilizing data from the National Health and Nutrition Examination Survey (NHANES), this study investigates the independent, joint, and interactive associations of serum 25(OH)D concentrations and OBS with all-cause, CVD, and cancer mortality among individuals with MetS, which aims to provide optimal antioxidant dietary and lifestyles guidance and target serum 25(OH)D levels for MetS. Methods Study population NHANES is a nationally representative, multistage, probability-based survey conducted by the National Center for Health Statistics (NCHS), a division of the Centers for Disease Control and Prevention (CDC), to assess the nutritional and health status of the U.S. population. The study protocol was approved by the NCHS Institutional Review Board, with all participants providing written informed consent. Detailed methodology and baseline characteristics have been documented elsewhere [17]. MetS was defined according to established guidelines [19, 20] as meeting at least three of the following criteria: (1) The fasting blood glucose (FBG) > 100 mg/dL(5.6mmol/L), or drug treatment for diabetes; (2) High-density lipoprotein cholesterol (HDL-C) levels below 40/50 mg/dL (1.0/1.3 mmol/L) in males/females, or drug treatment for reduced HDL-C; (3) Plasma triglycerides (TG) > 150 mg/dL (1.7 mmol/L), or drug treatment for raised TG; (4) Waist circumference > 102/88 cm (males/females); (5) Systolic blood pressure (BP) of 130 mmHg or higher, and/or diastolic BP at or more than 85 mmHg, or lower hypertension therapy. After excluding participants with undiagnosed MetS (n = 37615), missing 25(OH)D data (n = 959), ineligible follow-up data (n = 8), and pregnancy/lactation (n = 88), 12,078 individuals were included in the final analysis. Supplemental Fig. 1 presents the flowchart of the study. Associations between serum 25(OH)D concentrations and OBS with all-cause, CVD, and cancer mortality among individuals with MetS in NHANES 2001–2018. (A) all-cause mortality risk with different levels of serum 25(OH)D concentrations; (B) all-cause mortality risk with different levels of OBS; (C) CVD mortality risk with different levels of serum 25(OH)D concentrations; (D) CVD mortality risk with the different level of OBS; (E) cancer mortality risk with different levels of serum 25(OH)D concentrations; (F) cancer mortality risk with the different level of OBS. The results were adjusted for age (continuous), sex (male, or female), race/ethnicity (non-Hispanic white, non-Hispanic black, Mexican American, or other), education level (less than high school, high school or equivalent, college or above), family income-poverty ratio (< 1.0, 1.0–3.0, or > 3.0), family history of diabetes (yes or no), HbA1c (< 6.5% or ≥ 6.5%), insulin use (yes or no), VD assessment month (November 1 through April 30, or May 1 through October 31), self-reported hypertension, hypercholesterolemia, CVD, and diabetes (yes or no), dietary supplement use (yes or no). Main findings: Fig. 1 shows significant nonlinear inverse relationships between serum 25(OH)D and all-cause, CVD, and cancer mortality (all Poverall <0.05, Pnonlinearity <0.05), with inflection points at 75.8, 84.0, and 74.5 nmol/L, respectively. In contrast, linear inverse relationships were demonstrated between OBS and all-cause, CVD, and cancer mortality (all Poverall <0.05; Pnonlinearity>0.05) Measurement of serum 25(OH)D Serum 25(OH)D was measured using DiaSorin RIA (2001–2006) and a standardized liquid chromatography-tandem mass spectrometry (LC-MS/MS, 2007–2018). RIA values were converted to LC-MS/MS-equivalent per CDC protocols [21]. Serum 25(OH)D status was categorized into three groups: deficient (< 50.0 nmol/L), insufficient (50.0–74.9 nmol/L), and sufficient (≥ 75.0 nmol/L) according to the Guidelines of the Endocrine Society Clinical Practice [22]. Assessment of oxidative balance score The OBS incorporated 20 components (16 dietary nutrients from the first 24-hour dietary recalls and 4 lifestyle factors), consisting of 15 antioxidants and 5 pro-oxidants. The Oxidative Balance Score (OBS) is a comprehensive indicator established and validated in previous studies for assessing an individual’s overall oxidative/antioxidant status [16, 23, 24]. Most components have been previously used in OBS calculations [25], while six components including riboflavin [26], niacin [27], vitamin B6 [28], vitamin B12 [29], magnesium [30], and copper [31] were newly incorporated based on emerging evidence linking them to oxidative stress. Smoking status was evaluated using serum cotinine levels to capture both active tobacco use and exposure to environmental tobacco smoke. The 15 antioxidants included: dietary fiber, carotenoids (Retinol Equivalents, RE), riboflavin, niacin, vitamin B6, total folate, vitamin B12, vitamin C, vitamin E (Alpha-Tocopherol Equivalent, ATE), calcium, magnesium, zinc, copper, selenium, and Physical activity. The 5 pro-oxidants considered included: Body mass index (BMI), alcohol consumption, total fat, iron, and cotinine. To account for sex differences in dietary intake, body composition, and metabolism, which may influence the distribution of these variables and reduce misclassification bias, all other components were scored by sex according to tertiles [32, 33]. In addition, each component was assigned an equal weight in the scoring system. Antioxidative components were assigned 0, 1, and 2 points for the lowest to highest tertiles, respectively, whereas pro-oxidant components were inversely scored, receiving 2, 1, and 0 points from the lowest to highest tertiles. Tertile cut-offs were selected to balance statistical power and clinical interpretability (usually categorizing levels as low, moderate, and high). Detailed cut-off values for each component are provided in Supplementary Table 1. Specific scoring for lifestyle factors was as follows: (1) alcohol consumption was categorized into non-drinker (2 point), low-to-moderate drinker (< 2 drinks/d in males and < 1 drinks/d in females, 1 point), and heavy drinker (≥ 2 drinks/d in males and ≥ 1 drinks/d in females, 0 point) [34]; (2) Physical activity was categorized by weekly MET-minutes (leisure-time moderate/vigorous activity × frequency; metabolic equivalent, MET); and (3) BMI was determined by dividing weight in kilograms by the square of height in meters. The total OBS was calculated by summing the scores of all components, with a theoretical range of 0–39 points. Higher scores reflect greater antioxidant factors. Using the overall sample median of 20 points, the OBS was categorized into low (0–19 points) and high (20–39 points) groups. Outcome ascertainment Mortality (all-cause, CVD, and cancer mortality) was obtained from the National Death Index through 31 December, 2019 [35]. Disease-specific death was identified using the International Classification of Diseases-10th Revision (ICD-10) codes. CVD mortality (ICD-10: I00–I09, I11, I13, and I20–I51) and cancer mortality (ICD-10: C00–C97) were identified. Covariate assessment The information on age, sex, race, education level, family income, disease status, month of blood collection, and medication use was gathered from standardized questionnaires. Waist circumference, BP, and BMI were obtained from physical examinations. The family income-poverty ratio level was defined as the total family income divided by the poverty threshold [36]. Supplement use was defined as intake of any vitamins, minerals, or other dietary supplements within the past month. Serum 25(OH)D concentrations assessment month was grouped into two categories: November 1 to April 30 and May 1 to October 31, corresponding to the six-month examination period. Additionally, hemoglobin Alc (HbAlc), FBG, TG, and HDL-C were determined among participants who provided blood samples. Stringent procedures were followed for blood collection and analysis, with details provided elsewhere [34]. Statistical analysis All analyses incorporated sample weights, clustering, and stratification to account for the complex sampling design of NHANES. We compared baseline characteristics across vitamin D groups using variance for continuous variables and Rao–Scott chi-square test for categorical variables. Pearson time was determined from the date of the interview to the date of death or the end of follow-up (31 December, 2019), whichever occurred first. Cox proportional hazard models were utilized to compute hazard ratios (HRs) and 95% confidence intervals (CIs) for the association of serum 25(OH)D and OBS with all-cause, CVD, and cancer mortality. Joint associations were assessed by cross-classifying vitamin D status and OBS levels. Three models were fitted. Model 1 was served with no adjustments. In model 2, we adjusted for age (continuous), sex (male or female), and race/ethnicity (non-Hispanic white, non-Hispanic black, Mexican American, or other). In model 3, we additionally adjusted for education level (less than high school, high school or equivalent, college or above), family income-poverty ratio (< 1.0, 1.0–3.0, or > 3.0), family history of diabetes (yes or no), HbA1c (< 6.5% or ≥ 6.5%), insulin use (yes or no), vitamin D assessment month (November 1 through April 30, or May 1 through October 31), presence of hypertension, hypercholesterolemia, CVD, or diabetes (yes or no), and dietary supplement use (yes or no). We further evaluated the multiplicative and additive interactions between 25(OH)D and OBS on the risk of mortality. Additive interaction was assessed using relative excess risk due to interaction (RERI) and attributable proportion (AP) due to interaction. The absence of interaction was indicated if the 95% CIs for RERI and AP included 0 [37]. Missing values of covariates (≤ 12.2%) were imputed by chained equations with 5 imputations through SAS PROC MI, followed by PROC MIANALYZE. To assess the dose-response relationship between serum 25(OH)D and OBS with mortality, we employed restricted cubic spline (RCS) regression with three knots (10th, 50th, and 90th), with the multivariate adjustment mentioned above. Nonlinearity was assessed via the likelihood ratio test comparing 2 models: one with only the linear term and the other with the linear and the cubic spline terms. To evaluate the robustness of our findings, we conducted a series of sensitivity analyses. First, cox proportional hazard models were also utilized to compute the robustness of the association between the OBS and all-cause, CVD, and cancer mortality (removing one component at a time from the total score). Second, we respectively investigated the joint association of serum 25(OH)D with both dietary OBS (DOBS) and lifestyle OBS (LOBS) on mortality. Third, participants with a history of CVD and cancer were further excluded. Fourth, to explore whether there was a sex and age difference, subgroup analyses were conducted by age (< 60 years, ≥ 60 years) and sex (males, females). Fifth, to account for potential reverse causality, we excluded participants who died within 2 years of follow-up. Sixth, we further adjusted for C-reactive protein (CRP) to evaluate the potential mediation (only available in the NHANES 2001–2010, n = 6298). Seventh, a sensitivity analysis was conducted without the unknown cause of deaths. Eighth, we performed a sensitivity analysis by adjusting for only one variable (education level) in Model 3 to assess potential collinearity among the variables. In addition, we conducted a sensitivity analysis with the OBS scoring system is based on uniform gender cut-off points of tertiles. Finally, stratified analyses were also conducted by race/ethnicity (non-Hispanic white: n = 5538; or others: n = 6540, including non-Hispanic black, Mexican American, and other races) to explore the potential implications. The analyses were performed using SAS Survey Procedure software (version 9.4; SAS Institute, Cary, NC) and R software (version 4.3.2), with a two-sided P < 0.05 deemed statistically significant. Results The baseline characteristics of the adults are shown in Table 1. Among 12,078 adults with MetS (mean age, 54.6 years; 54.93% female; 45.85% non-Hispanic white), the weighted mean concentration of serum 25(OH)D was 64.8 (63.8, 65.9) nmol/L; 36.5% deemed deficient vitamin D (< 50 nmol/L); and 73.6% considered insufficient vitamin D (< 75 nmol/L). Participants with sufficient vitamin D (≥ 75 nmol/L) were generally older, more likely to be non-Hispanic white, tended to have higher education levels and family income-poverty ratios, and were less likely to have a history of cancer at baseline. Table 1Baseline characteristics of participants with metabolic syndrome according to serum 25(OH)D concentrations in NHANES 2001–2018Serum 25(OH)D concentrations (nmol/L) ∗Total< 50.050.0–74.9.0.9≥ 75.0Number of participants12,078440944843185Age (mean ± SE), years54.6 ± 0.2251.7 ± 0.3253.2 ± 0.3259.2 ± 0.34Female6635 (54.93)2607 (59.13)2187 (48.77)1841 (57.80)Race/ethnicity Non-Hispanic white5538 (45.85)1261 (28.60)2281 (50.87)1996 (62.67) Non-Hispanic black2451 (20.29)1471 (33.36)609 (13.58)371 (11.65) Mexican American2173 (17.99)1030 (23.36)840 (18.73)303 (9.51) Other1916 (15.86)647 (14.67)754 (16.82)515 (16.17)Education levels Less than high school3715 (31.21)1481 (34.39)1457 (32.88)777 (24.54) High school or equivalent2957 (24.84)1030 (23.91)1064 (24.01)863 (27.26) College or above5232 (43.95)1796 (41.70)1910 (43.11)1526 (48.20)Family income-poverty ratio ≤ 1.02448 (22.14)1018 (25.38)916 (22.18)514 (17.63) 1.0–3.05032 (45.52)1900 (43.37)1834 (44.42)1298 (44.53) > 3.03575 (32.34)1093 (27.25)1379 (33.40)1103 (37.84)Insulin use No10,914 (90.37)3946 (89.50)4103 (91.50)2865 (89.98) Yes1163 (9.63)463 (10.50)381 (8.50)319 (10.02)Family history of diabetes No6226 (53.57)2217 (52.67)2412 (55.74)1597 (51.75) Yes5396 (46.43)1992 (47.33)1915 (44.26)1489 (48.25)Self-reported diseases Diabetes5023 (41.59)1935 (43.89)1719 (38.34)1369 (42.98) Hypertension7770 (64.33)2716 (61.60)2754 (61.42)2300 (72.21) Hypercholesterolemia5804 (48.05)1842 (41.78)2183 (48.68)1779 (55.86) CVD2347 (19.43)833 (18.89)808 (18.02)706 (22.17) Cancer1528 (12.65)392 (8.89)543 (12.11)593 (18.62) Dietary supplement use6352 (52.59)1505 (34.13)2408 (53.70)2439 (76.58)Month of vitamin D assessment November through April5725 (47.40)2477 (56.18)1999 (44.58)1249 (39.22) May through October6353 (52.60)1932 (43.82)2485 (55.42)1936 (60.78)HbAlc (%) < 6.5%8692 (72.11)2994 (68.05)3354 (75.00)2344 (73.69) ≥ 6.5%3361 (27.89)1406 (31.95)1118 (25.00)837 (26.31) OBS (mean ± SE)20.0 ± 0.1118.4 ± 0.1620.5 ± 0.1820.9 ± 0.19Estimates accounted for complex survey designs. Data are numbers (percentages) unless otherwise indicated. SE: Standard Error; CVD, cardiovascular disease; OBS, oxidative balance score∗Vitamin D status was categorized into three groups: deficient (< 50.0 nmol/L), insufficient (50.0–74.9 nmol/L), and sufficient (≥ 75.0 nmol/L) according to the Endocrine Society Clinical Practice Guidelines Baseline characteristics of participants with metabolic syndrome according to serum 25(OH)D concentrations in NHANES 2001–2018 Estimates accounted for complex survey designs. Data are numbers (percentages) unless otherwise indicated. SE: Standard Error; CVD, cardiovascular disease; OBS, oxidative balance score ∗Vitamin D status was categorized into three groups: deficient (< 50.0 nmol/L), insufficient (50.0–74.9 nmol/L), and sufficient (≥ 75.0 nmol/L) according to the Endocrine Society Clinical Practice Guidelines Relationship between serum 25(OH)D concentrations, OBS, and mortality Over 99,690 person-years of follow-up (2357 totassociation of serum 25(OH)D concentrations and OBS with all-cause, CVD, and cancer ml deaths; 654 CVD deaths, 518 cancer deaths), multivariable-adjusted analyses revealed significant inverse associations between serum 25(OH)D levels and mortality. Compared to deficient vitamin D group (< 50.0 nmol/L), the multivariable-adjusted HRs and 95% CIs for insufficient (50.0–74.9 nmol/L) and sufficient (≥ 75 nmol/L) vitamin D group were 0.73 (0.63, 0.84) and 0.71 (0.61, 0.82) for all-cause mortality, respectively; 0.71 (0.55, 0.92) and 0.65 (0.48, 0.86) for CVD mortality, respectively; 0.70 (0.51, 0.96) and 0.77 (0.57, 1.05) for cancer mortality, respectively. Meanwhile, participants with high OBS (≥ 20) had lower risks of all-cause (HR 0.76, 0.68–0.86), CVD (HR 0.70, 0.58–0.85), and cancer mortality (HR 0.75, 0.60–0.94) than those with low OBS (< 20) (Table 2). Similarly, the HR (95%CIs) in the high DOBS and LOBS were 0.79 (0.70, 0.88), 0.67 (0.59, 0.76) for the risk of all-cause mortality, respectively (Table 2). In sensitivity analyses, after removing one component at a time from the total OBS, the HRs did not significantly affect after full adjustment (Supplementary Table 2). Table 2Association of serum 25(OH)D concentrations and OBS with all-cause, CVD, and cancer mortality among individuals with MetSMortality outcomeNumber of deathsModel 1∗Model 2†Model 3‡HR (95% CI) P HR (95% CI) P HR (95% CI) P All-cause mortality Serum 25(OH)D concentrations Deficient933/4409ReferenceReferenceReference Insufficient864/44840.83 (0.73,0.94)0.0040.66 (0.58,0.76)< 0.0010.73 (0.63,0.84)< 0.001 Sufficient560/31851.02 (0.86,1.20)0.830.63 (0.54,0.73)< 0.0010.71 (0.61,0.82)< 0.001OBS Low OBS1325/5962ReferenceReferenceReference High OBS1032/61160.66 (0.59,0.73)< 0.0010.67 (0.60,0.75)< 0.0010.76 (0.68,0.86)< 0.001DOBSLow DOBS1331/5917ReferenceReferenceReferenceHigh DOBS1026/61610.65(0.58,0.73)< 0.0010.71(0.64,0.79)< 0.0010.79(0.70,0.88)< 0.001LOBS Low LOBS1001/5065ReferenceReferenceReference High LOBS1356/70130.83(0.75,0.93)0.0010.56(0.50,0.62)< 0.0010.67(0.59,0.76)< 0.001 CVD mortality Serum 25(OH)D concentrations Deficient249/4409ReferenceReferenceReference Insufficient250/44840.82 (0.66,1.01)0.060.63 (0.49,0.81)< 0.0010.71 (0.55,0.92)0.009 Sufficient155/31850.94 (0.71,1.26)0.690.56 (0.42,0.75)< 0.0010.65 (0.48,0.86)0.003OBS Low OBS387/5962ReferenceReferenceReference High OBS267/61160.57 (0.47,0.69)< 0.0010.60 (0.50,0.72)< 0.0010.70 (0.58,0.85)< 0.001DOBS Low DOBS386/5917ReferenceReferenceReference High DOBS268/61610.57 (0.47,0.69)< 0.0010.63 (0.53,0.76)< 0.00010.72 (0.59,0.88)0.001LOBS Low LOBS282/5065ReferenceReferenceReference High LOBS372/70130.83 (0.67,1.03)0.090.55 (0.46,0.68)< 0.0010.70 (0.56,0.86)0.001 Cancer mortality Serum 25(OH)D concentrations Deficient209/4409ReferenceReferenceReference Insufficient177/44840.87 (0.65,1.16)0.340.67 (0.50,0.90)0.0080.70 (0.51,0.96)0.03 Sufficient132/31851.15 (0.86,1.53)0.340.71 (0.53,0.95)0.020.77 (0.57,1.05)0.10OBS Low OBS285/5962ReferenceReferenceReference High OBS233/61160.70 (0.56,0.87)0.0010.71 (0.57,0.88)0.0020.75 (0.60,0.94)0.01DOBS Low DOBS279/5917ReferenceReferenceReference High DOBS239/61610.69(0.55,0.86)0.0010.73(0.58,0.92)0.0070.78(0.62,0.97)0.03LOBS Low LOBS221/5065ReferenceReferenceReference High LOBS297/70130.87(0.72,1.05)0.140.63(0.51,0.77)< 0.0010.66(0.53,0.81)< 0.001CI, confidence interval; HR, hazard ratio; MetS, metabolic syndrome; OBS, oxidative balance score; DOBS, dietary oxidative balance score; LOBS, lifestyle oxidative balance score; CVD, cardiovascular diseaseSerum 25(OH)D concentrations: deficient (< 50.0 nmol/L), insufficient (50.0–74.9 nmol/L), and sufficient (≥ 75.0 nmol/L). OBS: low OBS level (0–19), and high OBS (20–39), based on the median score of OBS (median = 20). Dietary and Lifestyle OBS were divided into low/high groups based on the median score, respectively∗Model 1 was served with no adjustments†Model 2: adjusted for age (continuous), sex (male, or female), and race/ethnicity (non-Hispanic white, non-Hispanic black, Mexican American, or other)‡Model 3: further adjusted (from Model 2) for education level (less than high school, high school or equivalent, college or above), family income-poverty ratio (< 1.0, 1.0–3.0, or > 3.0), family history of diabetes (yes or no), HbA1c (< 6.5% or ≥ 6.5%), insulin use (yes or no), VD assessment month (November 1 through April 30, or May 1 through October 31), self-reported hypertension, hypercholesterolemia, CVD, and diabetes (yes or no), dietary supplement use (yes or no)Main findings: Compared to deficient vitamin D group, sufficient vitamin D group was associated with a 27% lower risk of all-cause mortality (HR = 0.73, 95%CI: 0.63–0.84) and 35% lower risk of CVD mortality (HR = 0.65, 95%CI: 0.48–0.86); Meanwhile, compared to low OBS, high OBS was associated with a 24% lower risk of all-cause mortality (HR = 0.76, 95%CI: 0.68–0.86), a 30% lower risk of CVD mortality (HR = 0.70, 95%CI: 0.59–0.83), and 25% lower risk of cancer mortality (HR = 0.75, 95%CI: 0.60–0.94) in Model 3 Association of serum 25(OH)D concentrations and OBS with all-cause, CVD, and cancer mortality among individuals with MetS CI, confidence interval; HR, hazard ratio; MetS, metabolic syndrome; OBS, oxidative balance score; DOBS, dietary oxidative balance score; LOBS, lifestyle oxidative balance score; CVD, cardiovascular disease Serum 25(OH)D concentrations: deficient (< 50.0 nmol/L), insufficient (50.0–74.9 nmol/L), and sufficient (≥ 75.0 nmol/L). OBS: low OBS level (0–19), and high OBS (20–39), based on the median score of OBS (median = 20). Dietary and Lifestyle OBS were divided into low/high groups based on the median score, respectively ∗Model 1 was served with no adjustments †Model 2: adjusted for age (continuous), sex (male, or female), and race/ethnicity (non-Hispanic white, non-Hispanic black, Mexican American, or other) ‡Model 3: further adjusted (from Model 2) for education level (less than high school, high school or equivalent, college or above), family income-poverty ratio (< 1.0, 1.0–3.0, or > 3.0), family history of diabetes (yes or no), HbA1c (< 6.5% or ≥ 6.5%), insulin use (yes or no), VD assessment month (November 1 through April 30, or May 1 through October 31), self-reported hypertension, hypercholesterolemia, CVD, and diabetes (yes or no), dietary supplement use (yes or no) Main findings: Compared to deficient vitamin D group, sufficient vitamin D group was associated with a 27% lower risk of all-cause mortality (HR = 0.73, 95%CI: 0.63–0.84) and 35% lower risk of CVD mortality (HR = 0.65, 95%CI: 0.48–0.86); Meanwhile, compared to low OBS, high OBS was associated with a 24% lower risk of all-cause mortality (HR = 0.76, 95%CI: 0.68–0.86), a 30% lower risk of CVD mortality (HR = 0.70, 95%CI: 0.59–0.83), and 25% lower risk of cancer mortality (HR = 0.75, 95%CI: 0.60–0.94) in Model 3 Figure 1 shows significant nonlinear inverse relationships between serum 25(OH)D and all-cause (Poverall <0.001, Pnonlinearity <0.001), CVD (Poverall <0.001, Pnonlinearity =0.01), and cancer mortality (Poverall <0.001, Pnonlinearity =0.002), with inflection points at 75.8, 84.0, and 74.5 nmol/L, respectively. The risk of mortality decreased sharply until these thresholds before stabilizing. In contrast, linear inverse relationships were demonstrated between OBS and all-cause, CVD, and cancer mortality (all Poverall <0.001; Pnonlinearity>0.05). Stratified analyses Stratified analyses revealed consistent associations between 25(OH)D and all-cause mortality in each stratum of OBS and 25(OH)D. Notably, high OBS was associated with a lower risk of CVD mortality in the stratum of the deficient vitamin D (VD) group (HR 0.52, 0.37–0.75) and cancer mortality in the stratum of the insufficient VD group (HR 0.57, 0.38–0.86). Whereas, 25(OH)D showed significant associations with CVD mortality in the low OBS group and with cancer mortality in the high OBS group (Fig. 2). Stratified analysis of serum 25(OH)D concentrations and OBS on risk of all-cause, CVD, and cancer mortality among individuals with MetS CI, confidence interval; HR, hazard ratio; MetS, metabolic syndrome; OBS, oxidative balance score; CVD, cardiovascular disease; 25(OH)D, 25-hydroxy vitamin D; VD, vitamin D Serum 25(OH)D concentrations: deficient (< 50.0 nmol/L), insufficient (50.0–74.9 nmol/L), and sufficient (≥ 75.0 nmol/L). OBS: low OBS level (0–19), and high OBS (20–39), based on the median score of OBS (median = 20). The results were adjusted for age (continuous), sex (male, or female), race/ethnicity (non-Hispanic white, non-Hispanic black, Mexican American, or other), education level (less than high school, high school or equivalent, college or above), family income-poverty ratio (< 1.0, 1.0–3.0, or > 3.0), family history of diabetes (yes or no), HbA1c (< 6.5% or ≥ 6.5%), insulin use (yes or no), VD assessment month (November 1 through April 30, or May 1 through October 31), self-reported hypertension, hypercholesterolemia, CVD, and diabetes (yes or no), dietary supplement use (yes or no). Main findings: Fig. 2 shows revealed consistent associations between 25(OH)D and all-cause mortality in each stratum of OBS and 25(OH)D. High OBS was associated with a lower risk of CVD mortality in stratum of deficient VD group (HR 0.52, 0.37–0.75) and cancer mortality in stratum of insufficient VD group (HR 0.57, 0.38–0.86). 25(OH)D showed significant associations with CVD mortality in the low OBS group and with cancer mortality in the high OBS group Interactive and joint association of serum 25(OH)D concentrations and OBS with mortality In the joint analyses, combinations of sufficient VD (≥ 75 nmol/L) and high OBS were associated with the lowest all-cause (HR 0.57, 95% CI: 0.46, 0.70) and CVD mortality (HR 0.48, 95% CI: 0.34, 0.69) compared with the deficient VD (< 50.0 nmol/L) and low OBS group in the fully adjusted model. Meanwhile, insufficient VD with high OBS was associated with the lowest cancer mortality risk (HR 0.52, 95% CI: 0.34, 0.81; Fig. 3, Supplementary Table 3). No significant multiplicative interaction between OBS and sufficient VD level was found on the risk for all-cause (HR: 0.99, 95%CI: 0.80, 1.22), CVD (HR: 1.41, 95%CI: 0.93, 2.14), and cancer mortality (HR: 1.01, 95%CI: 0.65, 1.57), respectively. However, significant additive interaction between OBS and sufficient VD level on CVD mortality (RERI = 0.29, 0.02–0.57; AP = 0.47, 0.03–0.91) was found (shown in Table 3), which suggested that there would be 45% of the total CVD mortality risk is attributable to the interaction. In other words, the joint effect of sufficient VD level and high OBS on CVD mortality risk was lower than the sum of their individual effects. Joint associations of serum 25(OH)D concentrations and OBS on risk of all-cause, CVD, and cancer mortality among individuals with MetS CI, confidence interval; HR, hazard ratio; MetS, metabolic syndrome; OBS, oxidative balance score; CVD, cardiovascular disease; 25(OH)D, 25-hydroxy vitamin D Serum 25(OH)D concentrations: deficient (< 50.0 nmol/L), insufficient (50.0–74.9 nmol/L), and sufficient (≥ 75.0 nmol/L). OBS: low OBS level (0–19), and high OBS (20–39), based on the median score of OBS (median = 20). The results were adjusted for age (continuous), sex (male, or female), race/ethnicity (non-Hispanic white, non-Hispanic black, Mexican American, or other), education level (less than high school, high school or equivalent, college or above), family income-poverty ratio (< 1.0, 1.0–3.0, or > 3.0), family history of diabetes (yes or no), HbA1c (< 6.5% or ≥ 6.5%), insulin use (yes or no), VD assessment month (November 1 through April 30, or May 1 through October 31), self-reported hypertension, hypercholesterolemia, CVD, and diabetes (yes or no), dietary supplement use (yes or no). Main findings: Fig. 3 shows combinations of sufficient VD (≥ 75 nmol/L) and high OBS were associated with the lowest all-cause (HR 0.57, 95% CI: 0.46, 0.70) and CVD mortality (HR 0.48, 95% CI: 0.34, 0.69) compared with the deficient VD (< 50.0 nmol/L) and low OBS group in the fully adjusted model. Meanwhile, insufficient VD with high OBS was associated with lowest cancer mortality risk (HR 0.52, 95% CI: 0.34, 0.81) Table 3Interactive effects of serum 25(OH)D concentrations and OBS with all-cause, CVD, and cancer mortality among individuals with MetSInteractive itemsInteractive effects (95% CI)All-cause mortalityCVD mortalityCancer mortality VD insufficiency and OBS Multiplicative interaction0.98(0.81,1.18)1.16(0.80,1.68)0.90(0.60,1.35) Additive interaction RERI0.01(−0.15,0.17)0.15(−0.12,0.42)−0.04(−0.37,0.29) AP0.01(−0.22,0.25)0.23(−0.19,0.65)−0.06(−0.64,0.51) VD sufficiency and OBS Multiplicative interaction0.99(0.80,1.22)1.41(0.93,2.14)1.01(0.65,1.57) Additive interaction RERI0.02(−0.14,0.19)0.29(0.02,0.57)0.03(−0.33,0.39) AP0.04(−0.23,0.30)0.47(0.03,0.91)0.04(−0.49,0.58)CI, confidence interval; 25(OH)D, 25-hydroxy vitamin D; MetS, metabolic syndrome; OBS, oxidative balance score; CVD, cardiovascular disease; VD, vitamin D; RERI, relative excess risk due to interaction; AP, attributable proportionSerum 25(OH)D concentrations: deficient (< 50.0 nmol/L), insufficient (50.0–74.9 nmol/L), and sufficient (≥ 75.0 nmol/L). OBS: low OBS level (0–19), and high OBS (20–39), based on the median score of OBS (median = 20)The results were adjusted for age (continuous), sex (male, or female), race/ethnicity (non-Hispanic white, non-Hispanic black, Mexican American, or other), education level (less than high school, high school or equivalent, college or above), family income-poverty ratio (< 1.0, 1.0–3.0, or > 3.0), family history of diabetes (yes or no), HbA1c (< 6.5% or ≥ 6.5%), insulin use (yes or no), VD assessment month (November 1 through April 30, or May 1 through October 31), self-reported hypertension, hypercholesterolemia, CVD, and diabetes (yes or no), dietary supplement use (yes or no)Main findings: Significant additive interaction between OBS and sufficient VD level on CVD mortality (RERI = 0.29, 0.02–0.57; AP = 0.47, 0.03–0.91) was found, which suggested that there would be 45% of the total CVD mortality risk is attributable to interactionValues in bold indicate statistical significance atP< 0.05 Interactive effects of serum 25(OH)D concentrations and OBS with all-cause, CVD, and cancer mortality among individuals with MetS CI, confidence interval; 25(OH)D, 25-hydroxy vitamin D; MetS, metabolic syndrome; OBS, oxidative balance score; CVD, cardiovascular disease; VD, vitamin D; RERI, relative excess risk due to interaction; AP, attributable proportion Serum 25(OH)D concentrations: deficient (< 50.0 nmol/L), insufficient (50.0–74.9 nmol/L), and sufficient (≥ 75.0 nmol/L). OBS: low OBS level (0–19), and high OBS (20–39), based on the median score of OBS (median = 20) The results were adjusted for age (continuous), sex (male, or female), race/ethnicity (non-Hispanic white, non-Hispanic black, Mexican American, or other), education level (less than high school, high school or equivalent, college or above), family income-poverty ratio (< 1.0, 1.0–3.0, or > 3.0), family history of diabetes (yes or no), HbA1c (< 6.5% or ≥ 6.5%), insulin use (yes or no), VD assessment month (November 1 through April 30, or May 1 through October 31), self-reported hypertension, hypercholesterolemia, CVD, and diabetes (yes or no), dietary supplement use (yes or no) Main findings: Significant additive interaction between OBS and sufficient VD level on CVD mortality (RERI = 0.29, 0.02–0.57; AP = 0.47, 0.03–0.91) was found, which suggested that there would be 45% of the total CVD mortality risk is attributable to interaction Values in bold indicate statistical significance atP< 0.05 Similar results were found for the joint associations of serum 25(OH)D and DOBS with all-cause, CVD, and cancer mortality (Supplementary Table 4). However, the joint associations were attenuated for LOBS, given that only the combination of high LOBS with either insufficient vitamin D (50.0–74.9 nmol/L) or sufficient vitamin D (≥ 75.0 nmol/L showed lower risk (all P < 0.05) compared to the reference group ((low LOBS with deficient vitamin D < 50.0 nmol/L), while other LOBS-vitamin D combinations were non-significant (Supplementary Table 4). Subgroup analysis of serum 25(OH)D concentrations and the OBS with mortality Subgroup analyses revealed stronger associations of 25(OH)D and OBS with all-cause and CVD mortality in younger (age < 60 years), males, and Non-Hispanic white, but not in those with other race for CVD mortality (HR 0.67; 95% 0.31, 1.41), while cancer mortality was more pronounced in females (Supplementary Table 5, Supplementary Table 11). The joint associations of serum 25(OH)D and OBS with mortality did not change significantly in individuals without a baseline history of CVD or cancer (Supplementary Table 6). Excluding participants who died within 2 years of follow-up did not notably alter the results (Supplementary Table 7). When further adjusting for CRP, the results did not materially change (n = 6298, Supplementary Table 8). The inverse association was observed after excluding cases with an unknown cause of death (Supplementary Table 9). The association between 25(OH)D, OBS, and mortality was attenuated when using an OBS scoring system based on uniform gender cut-off points of tertiles (Supplementary Table 10). It is plausible that gender-specific grouping criteria may more accurately capture the relationships between 25(OH)D, OBS, and mortality [32, 33]. Adjustment for education level alone was performed to assess potential collinearity, and the associations remained robust and largely unchanged (Supplementary Table 12). Discussion This large prospective study revealed two distinct mortality risk patterns: a pronounced “L-shaped” nonlinear relationship between 25(OH)D and mortality, and a linear inverse relationship between OBS and mortality. Individuals with both sufficient vitamin D (≥ 75.0 nmol/L) and high OBS exhibited the lowest risks of all-cause and CVD mortality, whereas those with insufficient vitamin D (50.0–74.9.0.9 nmol/L) combined with a high OBS showed the strongest protective effect against cancer mortality. These associations were particularly more pronounced for all-cause/CVD mortality in younger adults (< 60 years) and males, and for cancer mortality in females. Notably, a synergistic additive interaction was observed between OBS and sufficient 25(OH)D (≥ 75.0 nmol/L) for CVD mortality, accounting for 48% of the total protective effect. Our findings underscored the significance of considering the combined effect of serum 25(OH)D concentrations and OBS for the follow-up management of individuals with MetS. Numerous studies have demonstrated that vitamin D deficiency is associated with a higher risk of all-cause and cause-specific mortality in the general population [38, 39], as well as in individuals with prediabetes or diabetes [36, 40], metabolic dysfunction-associated fatty liver disease [41], CVD [42], and hypertension [43]. However, evidence regarding the potential health effect of 25(OH)D in preventing premature death among individuals with MetS remains scarce. Most recently, Wu et al. [15] found an “L-shaped” nonlinear dose-response relationship between 25(OH)D and cancer/all-cause mortality among participants aged ≥ 40 years with MetS, but its generalizability remains limited to all age groups. In another study involving 1801 participants from the Ludwigshafen Risk and Cardiovascular Health, Thomas et al. [14] reported that sufficient VD was significantly associated with a 72% and 64% reduction in the risk ofCVD and all-cause mortality, respectively. Notably, our study provides further evidence that adequate VD is significantly associated with a lower risk of all-cause, CVD, and cancer mortality among adults (age > 18 years) with MetS, irrespective of the type of mortality or age. Meanwhile, our study showed an L-shaped nonlinear dose-response relationship between 25(OH)D and mortality, suggesting that this relationship plateaued at approximately 74–84nmol/L. This finding is consistent with the Endocrine Society Clinical Practice Guidelines, which have defined VD sufficiency as ≥ 75 nmol/L [22]. However, since no further reduction in mortality was observed beyond approximately 84 nmol/L, excessive supplementation may be unnecessary among individuals with MetS. Observational analyses indicated threshold relationships between serum 25(OH)D levels and mortality [44]. For all-cause, CVD, and cancer mortality, inverse associations were more pronounced at lower 25(OH)D concentrations and attenuated at higher 25(OH)D levels, approximately above 80 nmol/L, which is consistent with our findings. However, long-term vitamin D supplementation (3200–4000 IU/day), corresponding to serum 25(OH)D levels of about 100–130 nmol/L, significantly increased the risks of hypercalcemia [Relative Risk(RR) = 2.21], falls (RR = 1.25), and hospitalization (RR = 1.16), with risks escalating further above 100 nmol/L [45]. Future studies are warranted to investigate whether maintaining 25(OH)D levels within this range represents an optimal level for mitigating premature mortality risks among individuals with MetS. Although several earlier studies have reported an inverse relationship between OBS and the risk of CVD [46], as well as CVD/all-cause mortality in the general population [47], diabetes and prediabetes [16], evidence specifically among individuals with MetS remains scarce. A recent study involving 3621 adults with MetS (181 deaths) did find that higher OBS was significantly associated with a lower risk of all-cause mortality [18]; however, it could not examine cause-specific mortality due to insufficient numbers of cause-specific deaths. Our large prospective study (n = 12078) provided robust evidence that higher OBS was associated with lower risks of CVD, cancer, and all-cause mortality, with reverse linear dose-response relationships confirmed after multivariate adjustment. Collectively, these findings underscore the potential value of increasing OBS levels through antioxidant lifestyle and dietary choices as a strategy to prevent premature deaths among individuals with MetS. It seems that insulin resistance, chronic inflammation, and neurohormonal activation are essential players in the progression of MetS [48]. Insulin resistance is characterized by the activation of the renin-angiotensin-aldosterone system (RAAS), which induces oxidative stress [49]. In turn, oxidative stress promotes excessive lipid peroxidation and the generation of reactive oxygen species (ROS), which impair β-cell differentiation and function [50], thereby triggering a cycle of oxidative stress [47, 51]. Moreover, ROS act as potent inflammatory triggers, stimulating the production of pro-inflammatory molecules such as interleukin-6 (IL-6), interleukin-1β (IL-1β), and tumor necrosis factor-α (TNFα) [52, 53]. Our study demonstrates that higher OBS and adequate vitamin D levels are associated with a lower risk of mortality among individuals with MetS. The OBS, calculated from 20 dietary and lifestyle components related to antioxidants and pro-oxidants, reflects the overall balance between pro-oxidant exposures and antioxidant defenses. The joint protective effects of these two factors likely involve coordinated modulation of insulin resistance, chronic inflammation, and neurohormonal activation. Endothelial dysfunction is a hallmark of MetS and a critical precursor to CVD mortality [54]. Insulin resistance and RAAS activation reduce endothelial nitric oxide (NO) bioavailability by inhibiting endothelial NO synthase (eNOS) activity or enhancing NO scavenging by ROS [55]. Vitamin D may counteract these effects through several targeted mechanisms. First, binding of vitamin D to the nuclear vitamin D receptor (VDR) upregulates eNOS expression and activity [56], while attenuating ROS -induced eNOS uncoupling [57]. Second, vitamin D suppresses RAAS activation, thereby mitigating oxidative stress and reducing endothelial cell apoptosis and monocyte adhesion [58, 59]. Third, vitamin D deficiency impairs endogenous antioxidant defenses by downregulating the nuclear factor erythroid 2-related factor 2 (Nrf2) pathway and glutathione synthesis [60–64]. In contrast, a high OBS reflects sufficient dietary antioxidant intake and limited pro-oxidant exposure [33]. Adequate vitamin D and high OBS create a “dual defense” system. Vitamin D enhances intrinsic cellular antioxidant capacity, while OBS provides exogenous antioxidant support. This synergy reduces oxidative damage to endothelial cells, preserves vascular barrier function, and thereby lowers the risk of atherosclerotic plaque instability and thrombotic events [65, 66]. Chronic inflammation is another hallmark of MetS. Beyond suppressing the expression of IL-1β, IL-6, and TNFα, vitamin D inhibits nuclear factor-κB (NF-κB) activation (a key transcription factor governing pro-inflammatory gene expression) [67, 68]. In individuals with MetS, NF-κB is hyperactivated by ROS and insulin resistance [68]. Vitamin D upregulates inhibitor of κB (IκB) proteins, preventing NF-κB nuclear translocation and thereby reducing levels of pro-inflammatory cytokines and adhesion molecules that mediate leukocyte adhesion to vascular endothelium [67, 68]. High OBS complements this effect by quenching ROS that activate NF-κB [66]. Inadequate levels of both vitamin D and OBS lead to sustained NF-κB activation and inflammation. Emerging evidence suggests that vitamin D also plays a role in regulating immune metabolism. VDR expression on pancreatic β-cells helps regulate insulin secretion through calcium homeostasis [50, 69]. Vitamin D counteracts oxidative stress and IL-1β induced β-cell damage, attenuating apoptosis and maintaining insulin secretion [70]. Several mechanistic questions remain, such as whether vitamin D acts synergistically with OBS-related antioxidants to activate the Nrf2 pathway. Further studies are needed to clarify these findings. Our study has several key strengths. First, its prospective design and use of a nationally representative cohort of U.S. adults with MetS enhance the generalizability of the findings. Second, this is the first study to assess the association between OBS and cause-specific mortality, as well as the joint effect of serum 25(OH)D and OBS on mortality among individuals with MetS. Clinically, these findings suggest that serum 25(OH)D testing might potentially be integrated into routine MetS care: guideline-based supplementation may be considered for deficient patients, alongside brief OBS assessments to identify those who could benefit from dietary intervention. However, several limitations should also be considered. First, as an observational study, we cannot determine causality between serum 25(OH)D, OBS, and mortality. Thus, the observed associations should be interpreted with caution. Well-designed randomized controlled trials are required to verify these associations, particularly to assess whether interventions targeting 25(OH)D and OBS can effectively reduce mortality risk. Second, we could not regularly update serum 25(OH)D or OBS-related dietary and lifestyle data. Although individuals with less than 2 years of follow-up were excluded and the results were similar, residual confounding and reverse causality are possible (e.g., healthier individuals may adopt antioxidant-rich diets or use supplements, while chronic disease may induce behavioral changes). Future studies with repeated exposure assessments are needed to clarify how temporal changes in oxidative balance and vitamin D status affect mortality in MetS. Third, the OBS included 16 dietary and 4 lifestyle components, so it was challenging to acquire more factors. Fourth, studies have indicated that the waist-to-height ratio was a more effective screening tool than waist circumference according to NCEP III criteria [48, 71]. Fifth, race/ethnicity-stratified analyses were constrained by small subgroup sample sizes, reducing statistical power. Therefore, the result should be interpreted with caution. Studies with adequate representation specific racial/ethnic are critically needed to better understand these potential disparities. Sixth, our findings could be influenced by unmeasured confounders, such as genetic susceptibility and chronic psychosocial stress, as well as limitations in the assessment of sunlight exposure. Although we adjusted for major lifestyle and clinical confounders, these unmeasured factors could theoretically affect the observed associations. For instance, sunlight exposure may be independently associated with mortality, potentially leading to overestimation of vitamin D’s protective effects. Future studies should incorporate genetic factors, stress, and sunlight exposure to address these potential confounders and further elucidate the underlying causal mechanisms. Conclusions In a nationally representative sample of U.S. adults with MetS, we found that participants with both adequate serum 25(OH)D and higher OBS were significantly associated with lower mortality. Notably, a synergistic effect in the prevention of premature CVD mortality was indicated among them. These findings suggest a potential association between vitamin D status, antioxidant-rich diet and lifestyle, and lower mortality risk among individuals with MetS. However, caution is warranted when extrapolating these findings beyond non-Hispanic White populations. Furthermore, given the inherent limitations of observational studies in establishing causality, further validation through randomized controlled trials is warranted. Supplementary Information Supplementary Material 1 Acknowledgements None. Abbreviations 25-hydroxy vitamin D Oxidative balance score Metabolic syndrome National health and nutrition examination survey Cardiovascular disease Hazard ratio Confidence intervals Relative excess risk due to interaction National Center for Health Statistics Centers for disease control and prevention Fasting blood glucose High-density lipoprotein cholesterol Triglycerides Blood pressure Retinol equivalents Alpha-tocopherol equivalent Body mass index Metabolic equivalent International classification of diseases-10th revision Hemoglobin Alc Attributable proportion Restricted cubic spline Dietary oxidative balance score Lifestyle oidative balance score C-reactive protein Vitamin D Relative risk Renin-angiotensin-aldosterone system Reactive oxygen species Interleukin-6 Interleukin-1β Tumor necrosis factor-α Nitric oxide endothelial NO synthase Vtamin D receptor Nuclear factor erythroid 2-related factor 2 Nuclear factor-κB Inhibitor of κB Author contributions XZ, WL: designed the research; XZ: conducted analysis and wrote the first draft of the paper; XZ, XF, JL, XZ, QZ, XY, WW, WL: contributed to the interpretation of the results and critical revision of the manuscript for important intellectual content; and all authors: read and approved the final version of the manuscript. Funding The study was supported by grants from the Hubei Province Nature Science Fund (2017CFB709). Data availability The datasets used in this study are publicly available at [https://wwwn.cdc.gov/nchs/nhanes/Default.aspx](https:/wwwn.cdc.gov/nchs/nhanes/Default.aspx). The code book and analytic code will be made available at https://github.com/gcxhahaha123/sas-code-for-mortality-analysis. Declarations Ethics approval and consent to participate All survey protocols were approved by the National Center for Health Statistics Ethics Review Board. All participants provided written informed consent before participation. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Wenjuan Wang and Wanjun Luo contributed equally as co-corresponding authors. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement The datasets used in this study are publicly available at [https://wwwn.cdc.gov/nchs/nhanes/Default.aspx](https:/wwwn.cdc.gov/nchs/nhanes/Default.aspx). The code book and analytic code will be made available at https://github.com/gcxhahaha123/sas-code-for-mortality-analysis.