Abstract
Background
The impact of diet on public health has always been a hot topic. This study first identified the association and sex differences between the modified Chinese version of the Mediterranean-DASH Intervention for Neurodegenerative Delay (cMIND) diet and all-cause mortality in Chinese older adults.
Methods
The data were obtained from the China Longitudinal Healthy Longevity Survey. The Cox regression model analyzed the association between the cMIND diet and all-cause mortality, conducted trend tests, and performed extensive subgroup and interaction analyses. Restricted cubic splines (RCS) were used to test the dose-response relationship. The random forest model ranked the importance of the components of the cMIND diet. Propensity score matching and two sensitivity analyses confirmed the robustness of the results.
Results
Compared to participants in Q1 of the cMIND diet, those in Q4 had a 23.3% lower risk of death (HR = 0.767, 95% CI: 0.722–0.815), with males benefiting more than females (female: HR = 0.810, 95% CI: 0.748, 0.878 vs. male: HR = 0.707, 95% CI: 0.643–0.778). The RCS results indicated a significant dose-response relationship between the above associations. The interaction analyses identified significant moderating effects of ethnic group, marital status, years of schooling, exercise, BMI, diabetes, and residence. The random forest analyses’ results showed that the amount of staple food is the most important dietary factor influencing all-cause mortality.
Conclusion
The identification of associations and the manifestation of sex differences in this study provided evidence in support of the optimization of dietary structure and precision dietary interventions for Chinese older adults.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12889-026-26311-w.
Keywords: Older adults, MIND, All-cause mortality, Sex differences, CLHLS
Introduction
As a critical factor in maintaining organismal health, diet continues to receive significant attention in the field of public health [1]. One of the main risk factors for mortality worldwide has been found to be unhealthy eating habits, and studies have shown that dietary imbalances account for 20% of fatalities [2]. Previous nutritional epidemiological studies have predominantly focused on the associations between individual nutrients (such as proteins and vitamins) or specific food categories (such as whole grains) and all-cause mortality or disease mortality [3–5]. However, dietary pattern research, by systematically integrating multidimensional parameters such as macronutrient ratios, micronutrient density, food processing methods, and cooking techniques, has proved more effective than single-nutrient studies in revealing the synergistic/antagonistic effects of nutrients on the mechanisms of disease development [6]. Numerous dietary patterns, including the Southern European Atlantic diet, Healthy Eating Patterns, and Plant-based and planetary-health diets, have been shown to exhibit dose-response relationships with lower all-cause mortality based on data gathered from extensive prospective cohort studies [7–9]. Notably, because of its distinct biological benefits in preventing neurodegenerative diseases, the MIND (Mediterranean-DASH Intervention for Neurodegenerative Delay) dietary pattern has lately gained attention in nutritional intervention research [10].
The MIND diet was developed by Morris MC et al. and is designed to incorporate the core elements of the Mediterranean diet and the Dietary Approaches to Stop Hypertension (DASH) diet [11]. This dietary pattern emphasizes the intake of plant-based foods such as green leafy vegetables, fruits, and nuts, uses olive oil as the main source of fat, and restricts the intake of animal-based foods and foods high in saturated fat [12]. Existing studies have shown that the assessment tool for the MIND diet can effectively distinguish the dietary quality of different individuals and has good score consistency in different diagnostic groups. The MIND diet can decrease the risk of physical dysfunction, dementia, cardiovascular diseases, and other problems, according to strong evidence from several cohort studies [13–15]. Although large-scale cohort studies in Western populations have established a dose-dependent inverse association of MIND dietary scores with all-cause mortality and disease mortality [10, 14, 16], there remains a paucity of prospective research evidence in Asian populations, particularly among Chinese older adults. Given the significant differences in dietary habits, lifestyles, and genetic factors between Eastern and Western populations, the existing MIND dietary pattern is not completely suitable for the culturally diverse dietary practices in China [17]. In recent years, researchers have adapted the MIND diet to better fit the dietary characteristics of the Chinese population by adjusting the dietary structure, increasing the intake of whole grains, and reducing the proportion of refined flour and rice, leading to the development of a Chinese version of the MIND diet measurement tool (cMIND) [18, 19]. Current studies suggest that adherence to the cMIND diet may significantly reduce the risk of cognitive impairment, depression, and hypertension among Chinese older adults [18–20].
Previous research has also confirmed that biological sex influences responses to diet-based interventions [21]. Physiological and metabolic differences between males and females may lead to varying nutritional needs, suggesting that a one-size-fits-all approach might underestimate the health benefits of the cMIND dietary pattern by overlooking sex-specific metabolic requirements. To date, no studies have investigated the sex differences in the association of the cMIND diet with all-cause mortality, a gap that may hinder the implementation of precision nutrition strategies. Additionally, individual differences in the diet-health relationship are also evident in other demographic characteristics such as age and residence [22, 23]. Existing literature has revealed the complex interplay between unhealthy lifestyles and diet in relation to mortality [24, 25]. As an illustration, a ten-year prospective cohort study revealed that individuals who followed the Mediterranean diet, moderate drinking, exercise, and non-smoking had a 65% reduced risk of all-cause mortality compared to people who did not follow any of these healthy lifestyle habits [25]. Concurrently, health status is a direct factor contributing to mortality in most older adults, with numerous studies reporting on the relationship between diet and mortality of patients who have specific diseases such as cardiovascular disease and cancer [26, 27]. Therefore, basic demographic characteristics, lifestyle, and health status may manifest in various forms and become controllable factors influencing the effects of diet. However, how these factors influence the association between the cMIND diet and the mortality of older adults has not been fully explored.
Investigating the association between the cMIND diet and all-cause mortality in the Chinese population and exploring sex differences in dietary responses are crucial for precisely formulating dietary recommendations to optimize metabolic health and improve survival rates among older adults. Therefore, our research is designed to systematically evaluate the association of the cMIND diet with all-cause mortality and its sex differences in Chinese older adults, based on data from the Chinese Longitudinal Healthy Longevity Survey (CLHLS). Additionally, we will explore how basic demographic characteristics, lifestyle, and health status variables influence this association, to provide a scientific basis for future precision nutritional interventions targeting older adults.
Methods
Study design and population
The CLHLS employs a multi-stage, stratified, cluster random sampling method, designed to conduct a nationally representative survey of Chinese adults aged 65 and above. The survey covers 23 provinces, municipalities, and autonomous regions across China, representing approximately 85% of the elderly population. The information collected by the project includes basic characteristics, socioeconomic status, behavioral habits, dietary status, and physical health conditions of older adults. Detailed information about the CLHLS has been reported in previous studies [28]. The CLHLS received ethical approval from the Biomedical Ethics Committee of Peking University, China (IRB00001052–13074). Each participant provided informed consent before data collection. The study adheres to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.
This study utilized follow-up data from the CLHLS database 2008–2018. In this follow-up analysis, 16,954 participants who were aged 65 and above were involved. Ultimately, 15,699 participants were involved in this analysis after older adults who did not have dietary and covariate information during follow-up and at baseline were excluded. Figure 1 depicts the detailed selection process.
Fig. 1.
Flowchart of sample selection
Measurement of the cMIND diet
Huang X et al. developed the cMIND diet scale for the Chinese population based on the MIND diet scale and the CLHLS food frequency questionnaire, and evaluated its reliability in the population [19]. Specifically, the cMIND diet scale focuses on the types and intake of staple foods based on the MIND diet pattern to reflect their core position in the traditional Chinese diet. The intake of tea and garlic, which are important in the Chinese diet and have local health evidence, was added. “Olive oil” was replaced by “healthy cooking oil”, which is more in line with the actual situation of the Chinese diet. At the same time, “berries” with low consumption frequency were replaced by “fresh fruits” with a wider range. Given that red wine is not commonly consumed among older adults in China, this item was not included. The final cMIND diet included 12 items: staple food, amount of staple food per day, grease, fresh vegetables, fresh fruits, fish, mushroom or algae, food made from beans, nut, garlic, tea, and sugar [20]. The scores of the three items of staple food type, staple food amount, and cooking oil were 0 or 1, and the scores of the remaining nine items were 0, 0.5, or 1. The total score is 12 points, and the higher the score, the higher the compliance with the cMIND diet [18]. Recent studies have also shown that the cMIND measurement tool has good applicability in older adults in China. In this study, the cMIND diet scores were divided into four groups. Detailed scoring criteria are presented in Supplementary Table S1.
Measurement of mortality
Official death certificates, family interviews, or municipal residential committees were used to determine the vital status and death date of deceased participants. The time interval between the baseline interview date and the death date or the final survey for survivors was used to determine the follow-up period [29]. Other studies have confirmed that mortality data collection is reliable [30].
Covariate
Covariates included data on age, sex, ethnic group, marital status, residence, years of schooling, living arrangements, drinking, smoking, exercise, hypertension, diabetes, and Body Mass Index (BMI). The specific variable categorization and assignment information are presented in Supplementary Table S2.
Statistical analysis
The samples were used for descriptive statistical analysis after data cleaning. The Kolmogorov-Smirnov test was employed to determine the normality of the continuous variables. Continuous variables that followed a normal distribution were expressed as mean ± standard deviation (M ± SD), while categorical variables were described using frequencies and percentages (n, %). For continuous and categorical variables, t-tests and χ2 tests were used to compare baseline characteristics across different sex groups and cMIND diet score groups.
Spearman correlation analysis was conducted to determine the relationships between individual cMIND diet components. Kaplan-Meier curve analysis and log-rank tests were used to assess sex differences in overall survival. The association between cMIND and all-cause mortality was examined using Cox proportional hazards regression models. We initially constructed three statistical models. Model 1 was a crude model. Model 2 controlled for age, sex, ethnic group, marital status, and years of schooling. Based on Model 2, Model 3 further adjusted for residence, living arrangements, drinking, smoking, exercise, BMI, hypertension, and diabetes. In order to estimate linear trends, the median of each group was considered as a continuous variable, with the lowest quartile (Q1) of cMIND as the reference group.
Moreover, Cox proportional hazards regression models were used to evaluate the associations between a single cMIND diet item and all-cause mortality, and adjusted for the same covariates listed in Model 3. Hazard ratios (HRs) and 95% confidence intervals (95% CIs) were reported for all Cox proportional hazards regression models.
Using the same covariates as in Model 3, we conducted restricted cubic spline analysis at the percentiles (5th, 35th, 65th, and 95th) of the cMIND diet distribution for a better understanding of the dose-response relationship between the cMIND diet and all-cause mortality in Chinese older adults.
Additionally, based on Model 3, subgroup analyses were performed, including variables such as age, sex, ethnic group, marital status, years of schooling, residence, living arrangements, drinking, smoking, exercise, BMI, hypertension, and diabetes. The interaction effects of subgroup variables were calculated. To evaluate the sex-specific importance of the cMIND diet and its individual 12 components on all-cause mortality, we split the entire sample at random into a 7:3 training and test set. Grid optimization was used to determine the model’s ideal parameters, and 10-fold cross-validation was used to confirm the accuracy of the model. The importance of dietary components was ranked based on the mean decrease in the Gini index.
We conducted supplementary analyses to better explain the sex differences in high cMIND benefit values. First, propensity score matching (PSM) was applied. A 1:1 nearest-neighbor matching algorithm was used to match participants in the lowest quartile (Q1) and the highest quartile (Q4) of cMIND scores, with a caliper value of 0.1, to eliminate bias and control for potential confounding variables. Confounding factors included all covariates considered in this study. An absolute standardized mean difference (ASMD) of less than 0.1 indicated a negligible imbalance between the two groups. After matching, we reanalyzed the sex differences in the association of cMIND scores with all-cause mortality using the same methods as before.
We also performed two sensitivity analyses to assess the robustness of the results. First, we used multiple imputations by chained equations to handle missing values, ensuring a comprehensive examination of the dataset. Second, recognizing that elderly individuals with hypertension or diabetes often require dietary control and may have specific dietary restrictions that influence their dietary patterns [31], we excluded participants with hypertension and diabetes from the analysis.
All statistical analyses were conducted using SPSS 27.0 and R 4.3.0. A two-tailed P value < 0.05 was considered statistically significant.
Results
Characteristics of the study participants in CLHLS 2008–2018
As shown in Table 1, this study involved a total of 15,699 participants, with a mean age of 87.75 ± 11.08 years, among whom 9,059 (57.70%) were female. Significant statistical differences were found between sexes in different ages, residences, marital status, living arrangements, economic status, years of schooling, smoking history, drinking history, exercise habits, BMI, gastrointestinal diseases, and diabetes (P < 0.05). The results in Table 2 indicated that groups with different cMIND diet scores exhibited significant statistical differences in age, sex, ethnic group, residence, marital status, living arrangements, economic status, years of schooling, smoking history, drinking history, exercise habits, BMI, hypertension, and diabetes (P < 0.05). Supplementary Fig. 1 illustrates the correlations between individual food components within the cMIND diet.
Table 1.
Characteristics of the study participants at baseline by sex
| Variables | Total(n = 15,699) | Male(n = 6,640) | Female(n = 9,059) | Statistic | P |
|---|---|---|---|---|---|
| Age, Mean ± SD | 87.75 ± 11.08 | 84.88 ± 10.31 | 89.86 ± 11.16 | t=-28.90 | < .001 |
| Ethnic group, n(%) | χ²=2.41 | 0.120 | |||
| Other | 926 (5.90) | 369 (5.56) | 557 (6.15) | ||
| Han | 14,773 (94.10) | 6,271 (94.44) | 8,502 (93.85) | ||
| Residence, n(%) | χ²=5.51 | 0.019 | |||
| Rural | 9,448 (60.18) | 3,925 (59.11) | 5,523 (60.97) | ||
| Urban | 6,251 (39.82) | 2,715 (40.89) | 3,536 (39.03) | ||
| Marital status, n(%) | χ²=1,752.36 | < .001 | |||
| Other | 11,117 (70.81) | 3,524 (53.07) | 7,593 (83.82) | ||
| Married | 4,582 (29.19) | 3,116 (46.93) | 1,466 (16.18) | ||
| Living arrangements, n(%) | χ²=11.97 | < .001 | |||
| Other | 2,702 (17.21) | 1,062 (15.99) | 1,640 (18.10) | ||
| In a nursing home | 12,997 (82.79) | 5,578 (84.01) | 7,419 (81.90) | ||
| Economic status, n(%) | χ²=38.60 | < .001 | |||
| Rich | 2,087 (13.29) | 991 (14.92) | 1,096 (12.10) | ||
| Common | 10,795 (68.76) | 4,561 (68.69) | 6,234 (68.81) | ||
| Poor | 2,817 (17.94) | 1,088 (16.39) | 1,729 (19.09) | ||
| Years of schooling, n(%) | χ²=3,443.04 | < .001 | |||
| ≤ 0 year | 9,975 (63.54) | 2,476 (37.29) | 7,499 (82.78) | ||
| 0–6 years | 4,241 (27.01) | 3,011 (45.35) | 1,230 (13.58) | ||
| > 6 years | 1,483 (9.45) | 1,153 (17.36) | 330 (3.64) | ||
| Smoking, n(%) | χ²=1,892.32 | < .001 | |||
| No | 13,055 (83.16) | 4,514 (67.98) | 8,541 (94.28) | ||
| Yes | 2,644 (16.84) | 2,126 (32.02) | 518 (5.72) | ||
| Drinking, n(%) | χ²=1,122.69 | < .001 | |||
| No | 13,067 (83.23) | 4,752 (71.57) | 8,315 (91.79) | ||
| Yes | 2,632 (16.77) | 1,888 (28.43) | 744 (8.21) | ||
| Exercise, n(%) | χ²=355.11 | < .001 | |||
| No | 11,414 (72.71) | 4,308 (64.88) | 7,106 (78.44) | ||
| Yes | 4,285 (27.29) | 2,332 (35.12) | 1,953 (21.56) | ||
| BMI, n(%) | χ²=296.50 | < .001 | |||
| 18.5–23.9 | 8,209 (52.29) | 3,878 (58.41) | 4,331 (47.81) | ||
| < 18.5 | 5,426(34.57) | 1,803 (27.15) | 3,623 (39.99) | ||
| 24–28 | 16,35 (10.41) | 797 (12.00) | 838 (9.25) | ||
| ≥ 28 | 429 (2.73) | 162 (2.44) | 267 (2.95) | ||
| Gastric or duodenal ulcer, n(%) | χ²=4.84 | 0.028 | |||
| No | 15,022 (95.69) | 6,326 (95.27) | 8,696 (95.99) | ||
| Yes | 677 (4.31) | 314 (4.73) | 363 (4.01) | ||
| Hypertension, n(%) | χ²=1.01 | 0.314 | |||
| No | 12,607 (80.30) | 5,357 (80.68) | 7,250 (80.03) | ||
| Yes | 3,092 (19.70) | 1,283 (19.32) | 1,809 (19.97) | ||
| Diabetes, n(%) | χ²=4.41 | 0.036 | |||
| No | 15,296 (97.43) | 6,449 (97.12) | 8,847 (97.66) | ||
| Yes | 403 (2.57) | 191 (2.88) | 212 (2.34) |
t: t-test, χ²: Chi-square test, SD: standard deviation
Table 2.
Characteristics of the study participants at baseline by cMIND diet scores
| Variables | Total (n = 15,699) |
cMIND diet scores | Statistic | P | |||
|---|---|---|---|---|---|---|---|
| Quartile 1 | Quartile 2 | Quartile 3 | Quartile 4 | ||||
| ≤ 4 (n = 5,278) | 4–5 (n = 4,062) | 5–6 (n = 3,458) | >6 (n = 2,937) | ||||
| Age, Mean ± SD | 87.75 ± 11.08 | 90.13 ± 10.47 | 87.99 ± 10.819 | 86.50 ± 11.15 | 84.63 ± 11.45 | F = 180.02 | < .001 |
| Sex, n(%) | χ²=218.48 | < .001 | |||||
| Male | 6,640 (42.30) | 1,905 (36.09) | 1,652 (41.03) | 1,541 (44.56) | 1,542 (52.50) | ||
| Female | 9,059 (57.70) | 3,373 (63.91) | 2,374 (58.97) | 1,917 (55.44) | 1,395 (47.50) | ||
| Ethnic group, n(%) | χ²=29.575 | < .001 | |||||
| Other | 926 (5.90) | 384 (7.28) | 192 (4.77) | 185 (5.35) | 165 (5.62) | ||
| Han | 14,773 (94.10) | 4,894 (92.72) | 3,834 (95.23) | 3,273 (94.65) | 2,772 (94.38) | ||
| Residence, n(%) | χ²=706.76 | < .001 | |||||
| Rural | 9,448 (60.18) | 3,812 (72.22) | 2,464 (61.20) | 1,897 (54.86) | 1,275 (43.41) | ||
| Urban | 6,251 (39.82) | 1,466 (27.78) | 1,562 (38.80) | 1,561 (45.14) | 1,662 (56.59) | ||
| Marital status, n(%) | χ²=303.00 | < .001 | |||||
| Other | 11,117 (70.81) | 4,105 (77.78) | 2,902 (72.08) | 2,342 (67.73) | 1,768 (60.20) | ||
| Married | 4,582 (29.19) | 1,173 (22.22) | 1,124 (27.92) | 1,116 (32.27) | 1,169 (39.80) | ||
| Living arrangements, n(%) | χ²=46.58 | < .001 | |||||
| Other | 2,702 (17.21) | 1,001 (18.97) | 740 (18.38) | 567 (16.40) | 394 (13.42) | ||
| In a nursing home | 12,997 (82.79) | 4,277 (81.03) | 3,286 (81.62) | 2,891 (83.60) | 2,543 (86.58) | ||
| Economic status, n(%) | χ²=618.94 | < .001 | |||||
| Rich | 2,087 (13.29) | 2,087 (13.29) | 488 (9.25) | 4937 (12.25) | 502 (14.52) | ||
| Common | 10,795 (68.76) | 10,795 (68.76) | 3,400 (64.42) | 2,803 (69.62) | 2,500 (72.30) | ||
| Poor | 2,817 (17.94) | 2,817 (17.94) | 1,390 (26.33) | 730 (18.13) | 456 (13.18) | ||
| Years of schooling, n(%) | χ²=736.68 | < .001 | |||||
| ≤ 0 year | 9,975 (63.54) | 9,975 (63.54) | 3,846 (72.87) | 2,679 (66.54) | 2,065 (59.72) | ||
| 0–6 years | 4,241 (27.01) | 1,143 (21.65) | 1,069 (26.55) | 1,039 (30.04) | 990 (33.71) | ||
| > 6 years | 1,483 (9.45) | 289 (5.48) | 278 (6.91) | 354 (10.24) | 562 (19.13) | ||
| Smoking, n(%) | χ²=56.50 | < .001 | |||||
| No | 13,055 (83.16) | 4,514 (85.52) | 3,389 (84.18) | 2,805 (81.12) | 2,347 (79.91) | ||
| Yes | 2,644 (16.84) | 764(14.48) | 637 (15.82) | 653 (18.88) | 590 (20.09) | ||
| Drinking, n(%) | χ²=80.14 | < .001 | |||||
| No | 13,067 (83.23) | 4,494 (85.15) | 3,438 (85.39) | 2,831 (81.87) | 2,304 (78.45) | ||
| Yes | 2,632 (16.77) | 784 (14.85) | 588 (14.61) | 627 (18.13) | 633 (21.55) | ||
| Exercise, n(%) | χ²=538.37 | < .001 | |||||
| No | 11,414 (72.71) | 4,254 (80.60) | 3,009 (74.74) | 2,475 (71.57) | 1,676 (57.07) | ||
| Yes | 4,285 (27.29) | 1,024 (19.40) | 1,017 (25.26) | 983 (28.43) | 1,261 (42.93) | ||
| BMI, n(%) | χ²=416.54 | < .001 | |||||
| 18.5–23.9 | 8,209 (52.29) | 2,542 (48.16) | 2,092 (51.96) | 1,908 (55.18) | 1,667 (56.76) | ||
| < 18.5 | 5,426 (34.56) | 2,239 (42.42) | 1,443 (35.84) | 1,058 (30.60) | 686 (23.36) | ||
| 24–28 | 1,635 (10.41) | 401 (7.60) | 387 (9.61) | 388 (11.22) | 459 (15.63) | ||
| ≥ 28 | 429 (2.73) | 96 (1.82) | 104 (2.58) | 104 (3.00) | 125 (4.25) | ||
| Gastric or duodenal ulcer, n(%) | χ²=2.048 | 0.562 | |||||
| No | 15,022 (95.69) | 5,042 (95.53) | 3,868 (96.08) | 3,303 (95.52) | 2,809 (95.64) | ||
| Yes | 677 (4.31) | 236 (4.47) | 158 (3.92) | 155 (4.48) | 128 (4.36) | ||
| Hypertension, n(%) | χ²=38.68 | < .001 | |||||
| No | 12,607 (80.30) | 4,356 (82.53) | 3,230 (80.23) | 2,763 (79.90) | 2,258 (76.88) | ||
| Yes | 3,092 (19.70) | 922 (17.47) | 796 (19.77) | 695 (20.10) | 679 (23.12) | ||
| Diabetes, n(%) | χ²=135.12 | < .001 | |||||
| No | 15,296 (97.43) | 5,218 (98.86) | 3,941 (97.89) | 3,355 (97.02) | 2,782 (94.72) | ||
| Yes | 403 (2.57) | 60 (1.14) | 85 (2.11) | 103 (2.98) | 155 (5.28) | ||
F: ANOVA, χ²: Chi-square test, SD: standard deviation
All-cause mortality
During the follow-up period from 2008 to 2018 (6.00 ± 3.84 years), there were a total of 9,288 deaths (59.20%), including 3,844 males and 5,444 females. Figure 2 illustrates that the cumulative all-cause mortality rate was decreased in males compared to females.
Fig. 2.
Kaplan - Meier survival analysis curves for all-cause mortality in female and male
Association between the cMIND diet and all-cause mortality
We employed Cox proportional hazards regression models to estimate the hazard ratios (HRs) of the cMIND diet related to mortality, adjusting for various sociodemographic, lifestyle, and health status factors. In Model 3, which was fully adjusted, in comparison with participants in the first quartile (Q1) of cMIND diet scores, those in the second quartile (Q2) (HR = 0.909, 95% CI: 0.858–0.962, P < 0.001), third quartile (Q3) (HR = 0.865, 95% CI: 0.816–0.917, P < 0.001), and fourth quartile (Q4) (HR = 0.767, 95% CI: 0.722–0.815, P < 0.001) were more likely to exhibit lower all-cause mortality. Among female and male participants, compared to Q1, the risk of mortality in Q4 was reduced by 19.0% (HR = 0.810, 95% CI: 0.748–0.878, P < 0.001) and 29.3% (HR = 0.707, 95% CI: 0.643–0.778, P < 0.001), respectively. Trend tests indicated a significant linear relationship between cMIND diet scores and all-cause mortality in the overall participant group, as well as in female and male participants (P < 0.001) (Table 3).
Table 3.
Association of cMIND diet scores with all-cause mortality
| Model | cMIND diet scores | P trend | |||
|---|---|---|---|---|---|
| Overall | Quartile 1 | Quartile 2 | Quartile 3 | Quartile 4 | P for trend |
| <4 | 4–5 | 5–6 | ≥ 6 | ||
| Model 1 | Ref. | 0.794(0.750, 0.840)*** | 0.692(0.653, 0.733)*** | 0.549(0.518, 0.581)*** | <0.001 |
| Model 2 | Ref. | 0.871(0.823, 0.921)*** | 0.806(0.761, 0.854)*** | 0.690(0.651, 0.732)*** | <0.001 |
| Model 3 | Ref. | 0.909(0.858, 0.962)** | 0.865(0.816, 0.917)*** | 0.767(0.722, 0.815)*** | <0.001 |
| Female | Quartile 1 | Quartile 2 | Quartile 3 | Quartile 4 | |
| <3.5 | 3.5–4.5 | 4.5–5.5 | ≥ 5.5 | ||
| Model 1 | Ref. | 0.766(0.706, 0.832)*** | 0.713(0.658, 0.772)*** | 0.554(0.513, 0.599)*** | <0.001 |
| Model 2 | Ref. | 0.843(0.776, 0.916)*** | 0.852(0.787, 0.923)*** | 0.745(0.689, 0.805)*** | <0.001 |
| Model 3 | Ref. | 0.869(0.800, 0.945)*** | 0.895(0.825, 0.970)** | 0.810(0.748, 0.878)*** | <0.001 |
| Male | Quartile 1 | Quartile 2 | Quartile 3 | Quartile 4 | |
| <4 | 4–5 | 5–6 | ≥ 6 | ||
| Model 1 | Ref. | 0.795(0.725, 0.873)*** | 0.695(0.634, 0.762)*** | 0.537(0.491, 0.587)*** | <0.001 |
| Model 2 | Ref. | 0.835(0.760, 0.916)*** | 0.747(0.681, 0.820)*** | 0.617(0.563, 0.676)*** | <0.001 |
| Model 3 | Ref. | 0.881(0.802, 0.967)** | 0.818(0.744, 0.898)*** | 0.707(0.643, 0.778)*** | <0.001 |
Data was weighted hazard ratio (95% confidence interval)
Model 1: Unadjusted variables
Model 2: Adjusted for age, sex (for all participants only), ethnic group, years of schooling, marital status
Model 3: Further adjusted for residence, living arrangements, smoking, drinking, exercise, BMI, hypertension, diabetes
After controlling for all covariates, we observed significant associations between the type of staple food, amount of staple food per day, grease, fresh fruits, fish, sugar, nuts, and mushroom or algae with all-cause mortality in the overall participant group (P < 0.05). Among female participants, grease, fresh fruits, fish, sugar, nuts, and mushroom or algae were significantly associated with all-cause mortality (P < 0.05). Among male participants, the type of staple food, the amount of staple food per day, fresh fruits, fish, sugar, nuts, and mushroom or algae were significantly associated with the risk of all-cause mortality (P < 0.05)(Supplementary Table S3).
RCS
In the fully adjusted Model 3, the results of restricted cubic spline (RCS) indicated a significant dose-response relationship between cMIND diet scores and all-cause mortality in the overall population (P for overall trend < 0.001; P for nonlinear association = 0.276). As the cMIND diet score increased by one standard deviation unit, the hazard ratio for all-cause mortality showed a decreasing trend. A dose-response relationship between cMIND diet scores and all-cause mortality was also observed among both male and female participants (Fig. 3).
Fig. 3.
Restricted cubic spline for testing the hypothesis of nonlinear relationship between cMIND diet scores and all-cause mortality in the overall group, female and male
Subgroup analyses
We considered 14 subgroups, including age, sex, ethnic group, marital status, years of schooling, residence, living arrangements, drinking, smoking, exercise, BMI, gastric or duodenal ulcer, hypertension, and diabetes, to further examine the consistency of the association of cMIND diet scores with all-cause mortality. In the overall population, as well as among female and male participants, the relationship between the cMIND dietary pattern and all-cause mortality remained statistically significant across all subgroup analyses (P < 0.05) (Fig. 4). Further interaction analyses revealed that, in the overall population, significant interaction effects were observed between cMIND diet scores and subgroups of living arrangements, years of schooling, exercise, and BMI (P for interaction < 0.05). Among females, significant interactions were observed between cMIND diet scores and subgroups of ethnic group, marital status, years of schooling, exercise, BMI, and diabetes. Notably, among male participants, only the residence subgroup showed a statistically significant interaction effect with cMIND diet scores (P for interaction < 0.05).
Fig. 4.
HRs and 95% CI for all-cause mortality per unit increment in adherence to cMIND diet scores, stratified by selected characteristics(n = 15,699). All models were multivariable adjusted for age, sex (for all participants only), ethnic group, years of schooling, marital status, residence, living arrangements, smoking, drinking, exercise, BMI, hypertension, diabetes
The results of random forest
The results of random forest analyses indicated that, among female participants, the amount of staple food per day was the most critical dietary item influencing all-cause mortality. The importance of other dietary components influencing all-cause mortality, in descending order, was as follows: garlic, food made from beans, sugar, fish, type of staple food, fresh fruits, mushroom or algae, tea, fresh vegetables, nuts, and grease. Among male participants, the amount of staple food per day was also the most critical dietary component influencing all-cause mortality. The importance of other dietary items affecting all-cause mortality, in descending order, was as follows: sugar, garlic, food made from beans, fish, fresh fruits, mushroom or algae, type of staple food, nuts, tea, fresh vegetables, and grease (Supplementary Fig. 2).
Secondary analyses
Using a 1:1 propensity score matching (PSM) with a nearest-neighbor matching algorithm, the study ultimately constructed a balanced cohort comprising 4,894 participants (2,447 in the lowest quartile of cMIND diet scores and 2,447 in the highest quartile). As shown in Supplementary Table S4, after PSM weighting, the two groups achieved statistical balance across all covariates (SMD < 0.1). The probability density distribution curves before and after PSM further demonstrated that the covariate distributions between the lowest and highest quartiles of cMIND diet scores were approximately aligned after matching (Supplementary Fig. 3). A significant relationship between cMIND diet scores and all-cause mortality was also observed after PSM (Supplementary Table S5).
Sensitivity analyses
The first time, missing data were handled using chained-equation multiple imputation (Supplementary Table S6). In the second time, participants with gastrointestinal diseases, hypertension, and diabetes were excluded. Both methods demonstrated that the statistical association between cMIND diet scores and all-cause mortality remained significant (P < 0.001) (Supplementary Table S7).
Discussion
In our 10-year prospective cohort study that focused on Chinese older adults (aged ≥ 65), males had a decreased cumulative all-cause mortality rate than females. This study is the first to show a substantial negative relationship between all-cause mortality in the Chinese population and cMIND diet scores. Both male and female participants showed the preventive impact of the cMIND diet on all-cause mortality, according to sex-stratified analyses, the benefit was more pronounced in males. Additionally, a significant dose-response relationship between cMIND diet scores and all-cause mortality was observed in both sexes. Subgroup analyses also identified significant interaction effects between cMIND diet scores and various subgroups in both females and males.
Several previous studies conducted in Western countries have explored the potential link between the MIND dietary pattern and mortality, nearly unanimously confirming that the MIND diet reduces mortality risk, which aligns with our findings. For example, a United Kingdom 10-year cohort study revealed that higher MIND scores were significantly associated with a reduced risk of all-cause mortality [32]. In another cohort study of Australian residents aged 25 and above, researchers found that the risk of mortality caused by cardiovascular disease decreased by 10% and the chance of all-cause mortality decreased by 7% for every point rise in the MIND diet score [33]. Janie Corley found that participants with the highest tertile of MIND diet scores had a 37% lower mortality risk than those with the lowest tertile, based on data from the Lothian Birth Cohort in Edinburgh, Scotland [10]. It is worth noting that the health benefits of the MIND diet are not limited to mortality. It was designed to be neuroprotective, as shown in the pioneering study by Morris et al.: older adults who strictly followed the MIND diet had significantly slower cognitive decline [11]. This strong protective effect on cognitive health provides a possible explanation for the MIND diet’s impact on long-term survival by slowing the progression of neurodegenerative diseases.
The cMIND diet is rich in various bioactive compounds, including polyphenols (e.g., fresh fruits and vegetable oils [34, 35]), carotenoids (e.g., fresh vegetables [34]), unsaturated fatty acids (e.g., nuts [36]), and long-chain omega-3 polyunsaturated fatty acids (e.g., fish [37]). The synergistic effects of these components may mitigate cellular aging, reduce DNA damage, and lower mortality risk by inhibiting NF-κB and MAPK/ERK signaling pathways, reducing Toll-like receptor and pro-inflammatory gene expression, suppressing enzymes associated with reactive oxygen species production, and upregulating endogenous antioxidant enzymes [38, 39]. Furthermore, the cMIND diet may reduce all-cause mortality by influencing the onset of chronic diseases. Previous researches have confirmed that the MIND diet can lower the risk of dementia, cardiovascular diseases, disability, and mental disorders, all of which are closely linked to increased mortality risk [15, 40, 41]. Another possible explanation is that individuals with higher adherence to the cMIND diet often adopt healthier lifestyles, such as regular exercise, non-smoking, moderate alcohol consumption, healthy eating habits, and regular sleep patterns. The combined effects of these health behaviors may promote overall well-being, reduce chronic disease burden, and thereby lower mortality risk.
Although our findings highlight the benefits of the cMIND diet for both males and females, the protective effects may differ between sexes due to physiological and metabolic differences. In this study, we observed that the protective effect of the cMIND diet on all-cause mortality was better in males than in females. Sex differences in dietary nutrition primarily manifest in nutrient intake, absorption, and metabolism. First, from the perspective of nutrient intake, differences were found in the quantity and types of food consumed by males and females. Our study found that the amount of staple food per day was the most critical factor influencing all-cause mortality in both sexes, but its protective effect was stronger in males than in females. This may partly explain why the protective effect of the cMIND diet on all-cause mortality was more pronounced in males. Possible explanations include: (1) females generally consume fewer total calories than males due to lower energy requirements [42]; (2) estradiol regulates food intake through estrogen receptor signaling, and elevated estradiol levels are associated with reduced food intake [43]. Due to the influence of estrogen during the menstrual cycle, females may consume a higher proportion of high-sugar and high-fat foods during specific periods compared to males [44]. It is well-known that excessive sugar intake adversely affects health [45], and our study also observed that sugar increases all-cause mortality risk. According to research, after consuming significant amounts of refined and quickly absorbed carbohydrates for a long time, females are more likely than males to develop glucose intolerance [21]. Females may be especially susceptible to the adverse health effects of sugar because of this double burden. The use of even little amounts of sugar may increase the risk of all-cause mortality in females relative to males, even if the cMIND diet promotes reduced sugar intake. The exposure theory, which suggests that variations in exposure to health-related variables may be the cause of sex differences, can also be used to explain this sex difference [46]. Males have a higher likelihood than females of participating in risky activities like smoking, drinking excessively, sleeping irregularly, and having poor dietary habits, making them more susceptible to a range of health issues. Therefore, male data tend to be more heterogeneous, and the benefits of dietary patterns are more easily detectable in males.
In the subgroup analyses, we observed significant interaction effects between the cMIND diet and all-cause mortality across certain characteristics. In females, significant interactions were observed for marital status, years of schooling, exercise, BMI, and diabetes. Marital status may be a key factor influencing health-damaging behaviors (e.g., substance use, non-adherence) and health-enhancing behaviors (e.g., physical activity, diet, adherence). A stable marital status may imply better adherence to the cMIND diet. However, females may be more physiologically influenced by marital status than males [47], which could explain why the interaction between marital status and the cMIND diet was only observed in females. Recent research has shown that lower education levels are associated with abnormal (either high or low) micronutrient intake and poorer overall diet quality [48]. Multimodal interventions combining nutritional guidance with physical exercise have gained increasing attention, with several studies demonstrating that combining exercise with specific dietary patterns can further improve disease outcomes and lower mortality risk [49, 50]. The relationship between the MIND diet and obesity has been extensively explored [51, 52], supporting the significant interaction between BMI and the cMIND diet in relation to all-cause mortality. Our study provides new insights and evidence for understanding how BMI influences the association between dietary patterns and all-cause mortality. Stephanie E. Tison et al. highlighted the significant association between the MIND diet and diabetes [53], and diabetes directly increases mortality risk [54]. Notably, in males, the relationship between the cMIND diet and mortality revealed a significant interaction only with residence. Previous studies have indicated that individuals living in rural areas are more likely to have poorer diet quality compared to urban residents [55], possibly due to limited access to balanced and healthy foods in rural settings.
This study was based on cohort data from 2008 to 2018, during which Chinese society was undergoing a rapid transformation. The overall dietary structure of the population showed a trend of shifting from a grain-based dietary structure to a dietary pattern with a higher proportion of non-staple foods and animal-derived foods [56]. This change was mainly affected by urbanization, rising incomes, globalization of food supply, and westernization of dietary culture [57]. Changes in the macro-dietary environment have a dual impact on individual dietary behavior: while broadening food choices and increasing dietary diversity, they also increase the accessibility of high-energy, low-nutrient-density foods, which may pose a challenge to maintaining traditional dietary patterns such as cMIND, which emphasizes plant-based foods, whole grains, and healthy fats. In this context, the association between cMIND dietary adherence and health outcomes in older adults observed in this study particularly highlights the special health protection value that may be possessed by adhering to this type of dietary pattern in an era of rapid evolution of dietary structure. In the future, long-term dynamic dietary monitoring studies are still needed to further reveal the evolution path of dietary behavior with social changes and its continuous impact on health outcomes.
Advantage and limitation
In this study, prospective design, sizable sample size, and accurate statistical analysis techniques are among its strong points. However, there are several limitations to our research as well. First, all the data collected in this study came from self-reports. Even though the investigators received expert training to reduce reporting errors, recall bias could still influence the findings. Second, a non-quantitative food frequency questionnaire was used to collect dietary data. However, previous research has shown that food frequency surveys can effectively capture dietary patterns in Asian populations. It’s possible that the findings won’t apply to other populations [58], the lack of precise quantitative assessment may limit the accuracy of food intake estimates. Future studies should consider more precise measurement methods to improve the accuracy of dietary assessments. Third, our findings are limited to Chinese older adults, and thus, it’s possible that the findings won’t apply to other populations. Fourth, due to the absence of specific cause-of-death data in the database, we were unable to further evaluate the relationship between the cMIND diet and cause-specific mortality. Fifth, due to the design of the original questionnaire and the data integrity of the sample, some meaningful variables (such as the frequency of smoking and drinking, and the use of supplements) were not included in this study, which to some extent limited the in-depth exploration of the relationship between related exposure factors and outcomes. Future prospective studies can consider including these variables in the evaluation system at the design stage.
Conclusion
This study is the first to examine the relationship between the cMIND diet and all-cause mortality in Chinese older adults, further exploring sex differences in this association. The results demonstrated that higher cMIND diet scores were related to a reduced risk of all-cause mortality among Chinese older adults, with this association being more pronounced in males. The findings highlight the limitations of a “one-size-fits-all” style to health and underscore the importance of paying greater attention to sex differences in the benefits of dietary interventions. This could enhance personalized healthcare and precisely improve health outcomes for all individuals. However, more research is required to fully comprehend the effect of the cMIND diet on all-cause mortality and its sex differences in the Chinese population.
Supplementary Information
Acknowledgements
We thank each and every participant for their support of our study. At the same time, we would like to express our sincere thanks to the staff of Chinese Longitudinal Healthy Longevity Survey(CLHLS) databased for collecting the data and making the data publicly accessible.
Authors’ contributions
XX: Conceptualization, Methodology, Data analysis, Draft Writing.SC: Data cleaning, Draft Writing.XW: data collection, Data preparation.ZZ: Visualization.XY: Visualization.LZ: Data preparation.YW: Data preparation.KS: Data preparation, Data analysis, Chart modification.QW: Data preparation, Data analysis, Chart modification.All authors read and approved the final manuscript.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Data availability
The data sets used and analyzed in this study are available from the corresponding author upon request. And all data can find in this link. 10.18170/DVN/WBO7LK.
Declarations
Ethics approval and consent to participate
This study was approved by the Biomedical Ethics Committee of Peking University, China (IRB00001052–13074). All participants provided written informed consent to participate in this study. The study adheres to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.
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.
Contributor Information
Kun Shen, Email: Ckoon1022@163.com.
Qian Wang, Email: 13996352313@163.com.
References
- 1.Downer S, Berkowitz SA, Harlan TS, Olstad DL, Mozaffarian D. Food is medicine: actions to integrate food and nutrition into healthcare. BMJ-BRIT MED J. 2020;369:m2482. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Health effects of dietary risks in 195 countries. 1990–2017: a systematic analysis for the global burden of disease study 2017. Lancet. [Journal Article; Research Support, Non-U.S. Gov’t]. 2019;393(10184):1958–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Aune D, Keum N, Giovannucci E, Fadnes LT, Boffetta P, Greenwood DC, et al. Whole grain consumption and risk of cardiovascular disease, cancer, and all cause and cause specific mortality: systematic review and dose-response meta-analysis of prospective studies. BMJ-BRIT MED J. 2016;353:i2716. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Cui Y, Zhou HL, Wei MH, Song WJ, Di DS, Zhang RY, et al. Multiple vitamin co-exposure and mortality risk: A prospective study. CLIN NUTR. [Journal Article; Research Support, Non-U.S. Gov’t]. 2022;41(2):337–47. [DOI] [PubMed] [Google Scholar]
- 5.Ren Z, Amakye WK, Yao M. Protein intake and Cause-Specific Mortality. JAMA INTERN MED. 2021;181(3):407–8. [DOI] [PubMed] [Google Scholar]
- 6.Mozaffarian D, Rosenberg I, Uauy R. History of modern nutrition science-implications for current research, dietary guidelines, and food policy. BMJ-BRIT MED J. [Historical Article; Journal Article; Research Support, N.I.H., Extramural]. 2018;361:k2392. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Carballo-Casla A, Stefler D, Ortolá R, Chen Y, Knuppel A, Kubinova R, et al. The Southern European Atlantic diet and all-cause and cause-specific mortality: a European multicohort study. EUR J PREV CARDIOL. 2024;31(3):358–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Shan Z, Wang F, Li Y, Baden MY, Bhupathiraju SN, Wang DD et al. Healthy eating patterns and risk of total and Cause-Specific Mortality. JAMA INTERN MED. [Journal Article; Research Support, N.I.H., Extramural; Research Support, Non-U.S. Gov’t]. 2023;183(2):142–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Chen H, Wang X, Ji JS, Huang L, Qi Y, Wu Y, et al. Plant-based and planetary-health diets, environmental burden, and risk of mortality: a prospective cohort study of middle-aged and older adults in China. LANCET PLANET HEALTH. 2024;8(8):e545–53. [DOI] [PubMed] [Google Scholar]
- 10.Corley J. Adherence to the MIND diet is associated with 12-year all-cause mortality in older adults. PUBLIC HEALTH NUTR. [Journal Article; Research Support, N.I.H., Extramural; Research Support, Non-U.S. Gov’t]. 2022;25(2):358–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Morris MC, Tangney CC, Wang Y, Sacks FM, Barnes LL, Bennett DA, et al. MIND diet slows cognitive decline with aging. ALZHEIMERS DEMENT. [Journal Article; Research Support, N.I.H., Extramural]. 2015;11(9):1015–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Dempsey DA, Agarwal P, Fernandez S, Brosch JR, Gao S, Clark DO, et al. Validation of a MIND diet screener in older adults. ALZHEIMERS DEMENT. [Journal Article; Validation Study]. 2025;21(10):e70766. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Hosking DE, Eramudugolla R, Cherbuin N, Anstey KJ. MIND not mediterranean diet related to 12-year incidence of cognitive impairment in an Australian longitudinal cohort study. ALZHEIMERS DEMENT. [Journal Article; Research Support, Non-U.S. Gov’t; Review]. 2019;15(4):581–9. [DOI] [PubMed] [Google Scholar]
- 14.Golzarand M, Mirmiran P, Azizi F. Adherence to the MIND diet and the risk of cardiovascular disease in adults: a cohort study. FOOD FUNCT. 2022;13(3):1651–8. [DOI] [PubMed] [Google Scholar]
- 15.Talegawkar SA, Jin Y, Simonsick EM, Tucker KL, Ferrucci L, Tanaka T. The Mediterranean-DASH intervention for neurodegenerative delay (MIND) diet is associated with physical function and grip strength in older men and women. AM J CLIN NUTR. [Journal Article; Research Support, N.I.H., Extramural]. 2022;115(3):625–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Song Y, Chang Z, Jia L, Song W, Wang H, Dong Q, et al. Better adherence to the MIND diet is associated with lower risk of all-cause death and cardiovascular death in patients with atherosclerotic cardiovascular disease or stroke: a cohort study from NHANES analysis. FOOD FUNCT. 2023;14(3):1740–9. [DOI] [PubMed] [Google Scholar]
- 17.Zhang R, Wang Z, Fei Y, Zhou B, Zheng S, Wang L, et al. The difference in nutrient intakes between Chinese and Mediterranean, Japanese and American Diets. NUTRIENTS. [Comparative study; Journal Article; Research Support, Non-U.S. Gov’t]. 2015;7(6):4661–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Wang Y, Zhang Y, Zeng X, Xian X, Chen J, Niu T. Association between cMIND diet and hypertension among older adults in china: a nationwide survey. AGING CLIN EXP RES. 2024;36(1):182. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Huang X, Aihemaitijiang S, Ye C, Halimulati M, Wang R, Zhang Z. Development of the cMIND diet and its association with cognitive impairment in older Chinese people. J NUTR HEALTH AGING. [Journal Article; Research Support, Non-U.S. Gov’t]. 2022;26(8):760–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Wang R, Ye C, Huang X, Halimulati M, Sun M, Ma Y, Fan R, Zhang Z. cMIND Diet, indoor air pollution, and depression: a cohort study based on the CLHLS from 2011 to 2018. Nutrients. 2023;15(5):1203. https://doi.org/10.3390/nu15051203. PMID: 36904202; PMCID: PMC10005708. [DOI] [PMC free article] [PubMed]
- 21.Andrews RR, Anderson KR, Fry JL. Sex-specific variation in metabolic responses to diet. Nutrients. 2024;16(17):2921. 10.3390/nu16172921. PMID: 39275236; PMCID: PMC11397081. [DOI] [PMC free article] [PubMed]
- 22.Senior AM, Nakagawa S, Raubenheimer D, Simpson SJ. Global associations between macronutrient supply and age-specific mortality. P NATL ACAD SCI USA. [Journal Article; Research Support, Non-U.S. Gov’t]. 2020;117(48):30824–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Swaminathan S, Dehghan M, Raj JM, Thomas T, Rangarajan S, Jenkins D, et al. Associations of cereal grains intake with cardiovascular disease and mortality across 21 countries in Prospective Urban and Rural Epidemiology study: prospective cohort study. BMJ-BRIT MED J. [Journal Article; Multicenter Study; Research Support, Non-U.S. Gov’t]. 2021;372:m4948. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.van den Brandt PA. The impact of a mediterranean diet and healthy lifestyle on premature mortality in men and women. AM J CLIN NUTR. [Journal Article; Research Support, Non-U.S. Gov’t]. 2011;94(3):913–20. [DOI] [PubMed] [Google Scholar]
- 25.Knoops KT, de Groot LC, Kromhout D, Perrin AE, Moreiras-Varela O, Menotti A, et al. Mediterranean diet, lifestyle factors, and 10-year mortality in elderly European men and women: the HALE project. JAMA-J AM MED ASSOC. [Journal Article; Multicenter Study; Research Support, Non-U.S. Gov’t]. 2004;292(12):1433–9. [DOI] [PubMed] [Google Scholar]
- 26.Bonaccio M, Costanzo S, Di Castelnuovo A, Persichillo M, Magnacca S, De Curtis A, et al. Ultra-processed food intake and all-cause and cause-specific mortality in individuals with cardiovascular disease: the Moli-sani study. EUR HEART J. [Journal Article; Research Support, Non-U.S. Gov’t]. 2022;43(3):213–24. [DOI] [PubMed] [Google Scholar]
- 27.Park Y, Farhat Z, Liao LM, Inoue-Choi M, Loftfield E. Multivitamin use and all-cause and cause-specific mortality in cancer survivors. BRIT J CANCER. 2024;130(1):82–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Chen H, Shen J, Xuan J, Zhu A, Ji JS, Liu X, et al. Plant-based dietary patterns in relation to mortality among older adults in China. Nat Aging [Journal. 2022;2022(3/12):224–30. Article; Research Support, Non-U.S. Gov’t]. [DOI] [PubMed] [Google Scholar]
- 29.Lv X, Li W, Ma Y, Chen H, Zeng Y, Yu X, et al. Cognitive decline and mortality among community-dwelling Chinese older people. BMC MED. [Journal Article; Research Support, Non-U.S. Gov’t]. 2019;17(1):63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Yin R, Wang Y, Li Y, Lynn HS, Zhang Y, Jin X, et al. Changes in physical activity and all-cause mortality in the oldest old population: findings from the Chinese longitudinal healthy longevity survey (CLHLS). PREV MED. 2023;175:107721. [DOI] [PubMed] [Google Scholar]
- 31.Toi PL, Anothaisintawee T, Chaikledkaew U, Briones JR, Reutrakul S, Thakkinstian A. Preventive role of diet interventions and dietary factors in type 2 diabetes mellitus: An Umbrella Review. Nutrients. 2020;12(9):2722. 10.3390/nu12092722. PMID: 32899917; PMCID: PMC7551929. [DOI] [PMC free article] [PubMed]
- 32.Song Y, Chang Z, Song C, Cui K, Shi B, Zhang R, et al. Association between MIND diet adherence and mortality: insights from diabetic and Non-Diabetic Cohorts. NUTR DIABETES. [Journal Article; Research Support, Non-U.S. Gov’t]. 2023;13(1):18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Livingstone KM, Milte CM, Torres SJ, Hart MJ, Dingle SE, Shaw JE, et al. Nineteen-Year associations between three diet quality indices and All-Cause and cardiovascular disease mortality: the Australian Diabetes, Obesity, and lifestyle study. J NUTR. [Journal Article; Research Support, Non-U.S. Gov’t]. 2022;152(3):805–15. [DOI] [PubMed] [Google Scholar]
- 34.Miller V, Mente A, Dehghan M, Rangarajan S, Zhang X, Swaminathan S, et al. Fruit, vegetable, and legume intake, and cardiovascular disease and deaths in 18 countries (PURE): a prospective cohort study. Lancet. [Comparative Study; Journal Article; Multicenter Study]. 2017;390(10107):2037–49. [DOI] [PubMed] [Google Scholar]
- 35.Imamura F, Micha R, Wu JH, de Oliveira OM, Otite FO, Abioye AI, et al. Effects of saturated Fat, polyunsaturated Fat, monounsaturated Fat, and carbohydrate on Glucose-Insulin homeostasis: A systematic review and Meta-analysis of randomised controlled feeding Trials. PLOS MED. [Journal Article; Meta-Analysis; Review; Systematic Review]. 2016;13(7):e1002087. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Kris-Etherton PM, Hu FB, Ros E, Sabaté J. The role of tree nuts and peanuts in the prevention of coronary heart disease: multiple potential mechanisms. J Nutr. 2008;38(9):1746S–51S. 10.1093/jn/138.9.1746S. PMID: 18716180. [DOI] [PubMed]
- 37.Zhang Y, Zhuang P, He W, Chen JN, Wang WQ, Freedman ND, et al. Association of fish and long-chain omega-3 fatty acids intakes with total and cause-specific mortality: prospective analysis of 421 309 individuals. J INTERN MED. [Journal Article; Research Support, N.I.H., Intramural; Research Support, Non-U.S. Gov’t]. 2018;284(4):399–417. [DOI] [PubMed] [Google Scholar]
- 38.Liu Y, Fang M, Tu X, Mo X, Zhang L, Yang B, Wang F, Kim YB, Huang C, Chen L, Fan S. Dietary Polyphenols as anti-aging agents: targeting the Hallmarks of aging. Nutrients. 2024;16(19):3305. 10.3390/nu16193305. PMID: 39408272; PMCID: PMC11478989. [DOI] [PMC free article] [PubMed]
- 39.Guan L, Eisenmenger A, Crasta KC, Sandalova E, Maier AB. Therapeutic effect of dietary ingredients on cellular senescence in animals and humans: A systematic review. AGEING RES REV. [Journal Article; Research Support, Non-U.S. Gov’t; Review]. 2024;95:102238. [DOI] [PubMed] [Google Scholar]
- 40.Salari-Moghaddam A, Keshteli AH, Mousavi SM, Afshar H, Esmaillzadeh A, Adibi P. Adherence to the MIND diet and prevalence of psychological disorders in adults. J AFFECT DISORDERS. [Journal Article; Research Support, Non-U.S. Gov’t]. 2019;256:96–102. [DOI] [PubMed] [Google Scholar]
- 41.Barnes LL, Dhana K, Liu X, Carey VJ, Ventrelle J, Johnson K, et al. Trial of the MIND diet for prevention of cognitive decline in older Persons. NEW ENGL J MED. [Comparative Study; journal Article; multicenter Study; randomized controlled Trial]. 2023;389(7):602–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Wu BN, O’Sullivan AJ. Sex differences in energy metabolism need to be considered with lifestyle modifications in humans. J NUTR METAB. 2011;2011:391809. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Rasic-Markovic A, Djuric E, Skrijelj D, Bjekic-Macut J, Ignjatovic Đ, Sutulovic N, et al. Neuroactive steroids in the neuroendocrine control of food intake, metabolism, and reproduction. Endocrine [Journal Article; Review]. 2024;85(3):1050–7. [DOI] [PubMed] [Google Scholar]
- 44.Asarian L, Geary N. Modulation of appetite by gonadal steroid hormones. PHILOS T R SOC B. [Journal Article; Research Support, N.I.H., Extramural; Research Support, Non-U.S. Gov’t; Review]. 2006;361(1471):1251–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Huang Y, Chen Z, Chen B, Li J, Yuan X, Li J, et al. Dietary sugar consumption and health: umbrella review. BMJ-BRIT MED J [Journal Article; Syst Review]. 2023;381:e71609. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Denton M, Prus S, Walters V. Gender differences in health: a Canadian study of the psychosocial, structural and behavioural determinants of health. SOC SCI MED. [Comparative Study; journal Article]. 2004;58(12):2585–600. [DOI] [PubMed] [Google Scholar]
- 47.Robles TF, Slatcher RB, Trombello JM, McGinn MM. Marital quality and health: a meta-analytic review. PSYCHOL BULL [Journal Article; Meta-Analysis; Research Support, N, Non- IH. P.H.S.; Review]. 2014;140(1):140–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Rippin HL, Hutchinson J, Greenwood DC, Jewell J, Breda JJ, Martin A, et al. Inequalities in education and National income are associated with poorer diet: pooled analysis of individual participant data across 12 European countries. PLoS ONE. [Journal Article; Research Support, Non-U.S. Gov’t]. 2020;15(5):e232447. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Levak N, Lehtisalo J, Thunborg C, Westman E, Andersen P, Andrieu S, et al. Nutrition guidance within a multimodal intervention improves diet quality in prodromal Alzheimer’s disease: Multimodal Preventive Trial for Alzheimer’s Disease (MIND-AD(mini)). ALZHEIMERS RES THER. [Journal Article; Multicenter Study; Randomized Controlled Trial; Research Support, Non-U.S. Gov’t] 2024;16(1):147. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Vilar-Gomez E, Vuppalanchi R, Gawrieh S, Pike F, Samala N, Chalasani N. Significant Dose-Response association of physical activity and diet quality with mortality in adults with suspected NAFLD in a population study. AM J GASTROENTEROL. 2023;118(9):1576–91. [DOI] [PubMed] [Google Scholar]
- 51.Asgari E, Chamary M, Bellissimo N, Azadbakht L. Association between adherence to the MIND diet and overweight and obesity in children: an exploratory study. CLIN NUTR ESPEN. 2022;51:313–8. [DOI] [PubMed] [Google Scholar]
- 52.Fateh HL, Muhammad SS, Kamari N. Associations between adherence to MIND diet and general obesity and lipid profile: A cross-sectional study. FRONT NUTR. 2023;10:1078961. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Tison SE, Shikany JM, Long DL, Carson AP, Cofield SS, Pearson KE, et al. Differences in the Association of Select Dietary Measures With Risk of Incident Type 2 Diabetes. Diabetes Care. 2022;45(11):2602–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54.Magliano DJ, Sacre JW, Harding JL, Gregg EW, Zimmet PZ, Shaw JE. Young-onset type 2 diabetes mellitus - implications for morbidity and mortality. NAT REV ENDOCRINOL. 2020;16(6):321–31. [DOI] [PubMed] [Google Scholar]
- 55.Tharrey M, Bohn T, Klein O, Bulaev D, Van Beek J, Nazare JA, et al. Local retail food environment exposure and diet quality in rural and urban adults: A longitudinal analysis of the ORISCAV-LUX cohort study. HEALTH PLACE. 2024;87:103240. [DOI] [PubMed] [Google Scholar]
- 56.Lam H, Remais J, Fung M, Xu L, Sun SS. Food supply and food safety issues in China. Lancet. 2013;381(9882):2044–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57.Xiong X, Zhang L, Hao Y, Zhang P, Chang Y, Liu G. Urban dietary changes and linked carbon footprint in china: A case study of Beijing. J ENVIRON MANAGE. 2020;255:109877. [DOI] [PubMed] [Google Scholar]
- 58.Sun R, Liu Y. Mortality of the oldest old in China. J AGING HEALTH. 2006;2006(2/1):37–55. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data sets used and analyzed in this study are available from the corresponding author upon request. And all data can find in this link. 10.18170/DVN/WBO7LK.