Effects of a cardiometabolic risk–reducing dietary pattern health education intervention on cardiometabolic disease risk in community-dwelling older adults in China: a cluster-randomized clinical trial Authors: Qi Wu, Li Huang, Jie Yang, Liyang Zhang, Cheng Cheng, Xukun Chen, Tongtong Li, Huilian Duan, Zehao Wang, Qinghan Ren, Tongyang Wu, Yue Wang, Shanshan Bian, Zhenshu Li, Huan Liu, Jing Yan, Yue Du, Yongjie Chen, Huichao Qu, Ya Su, Baoguo Gao, Wei Liu, Lichun Cao, Wen Li, Fei Ma, Guowei Huang Published: 2026 ABSTRACT ObjectiveTo assess the efficacy of the reducing cardiometabolic disease risk (RCMDR) dietary pattern health education intervention on cardiometabolic disease (CMD) risk and carotid atherosclerosis in the Chinese older population.DesignThe cluster randomized clinical trial was conducted from April 2023 to December 2024. Participants according to residential areas were randomized to the intervention group (2 clusters) or control group (2 clusters).SettingSingle-center in Tianjin, China.Participants694 community older residents (61-80 years): intervention group (n = 349) and control group (n = 345).InterventionParticipants in the intervention group received the RCMDR dietary pattern health education intervention, while those in the control group received general health education. The intervention period lasted 12 months.MeasurementsThe primary outcome was clustered cardiometabolic risk score (CMRS). The secondary outcomes included common carotid artery ultrasound parameters and cardiometabolic risk indicators. Analysis was intention-to-treat principle.ResultsOf 694 participants, 411 (59.2%) were female; mean (SD) age was 68.2 (5.1) years. Compared with the control group, the RCMDR dietary pattern health education intervention resulted in a significantly lower CMRS (β = -0.25; 95% CI -0.31, -0.19); lower intima-media thickness (β = -0.05, 95% CI -0.08, -0.02); lower serum total cholesterol, triglyceride and low-density lipoprotein cholesterol, higher high-density lipoprotein cholesterol; lower fasting blood glucose, fasting insulin, fasting C-peptide and HOMA-IR, higher HOMA-β; lower serum homocysteine; lower systolic blood pressure and diastolic blood pressure; lower weight, body mass index, waist circumference, waist-to-hip ratio, body fat percentage, body fat mass and Chinese visceral adiposity index, higher lean body mass (all P < 0.05).ConclusionThe RCMDR dietary pattern health education intervention could reduce the CMD risk, delayed the progression of carotid atherosclerosis and improved cardiometabolic factors among Chinese older population. Introduction Cardiometabolic diseases (CMDs), including a series of interrelated diseases, such as hypertension, diabetes, dyslipidemia, coronary heart disease and stroke, are the leading causes of death and disability worldwide and also pose a serious threat to the health of Chinese residents [[1], [2], [3]]. Due to the accelerated aging of the population, the prevalence of one or more CMDs among Chinese older people aged 60 and above has exceeded 50% and is still rising [4,5]. Atherosclerosis refers to the formation of fibrofatty lesions in the arterial wall and is the major culprit of cardiovascular diseases [3,6]. Substantial evidence has supported the beneficial role of healthy dietary factors in preventing the occurrence and development of CMDs and atherosclerosis in the aging population [[7], [8], [9], [10]]. The analysis of holistic dietary patterns is a more powerful tool for assessing the synergistic and cumulative effects of various food groups or multiple nutrients on cardiometabolic health [11]. The reducing cardiometabolic disease risk (RCMDR) dietary pattern was designed based on Chinese dietary culture, focusing on the prevention of CMDs, which is characterized by increasing the intake of whole grains, vegetables, fruits, soybeans, nuts, fish, and low-fat, skimmed, or fermented dairy products; decreasing the intake of cooking oil, salt, added sugar, and animal-derived food, especially limiting the intake of red meat and banning processed meats [12,13]. In our previously published protocol article, we described the process of constructing the RCMDR dietary pattern using the Delphi method and detailed the content of the RCMDR dietary pattern [12]. In a 12-week randomized controlled trial (RCT) article that we have subsequently published, the cardiometabolic benefits of the RCMDR dietary pattern were evaluated and validated in 100 participants aged 34-45 with dyslipidemia using the intervention combining standard meals feeding and dietary guidance standard dietary feeding and dietary guidance [13]. However, due to the small sample size and short intervention period in RCT, larger-scale studies with larger sample sizes and longer intervention periods are needed to further validate the intervention and generalization effects of the RCMDR dietary pattern. Furthermore, the intervention effects of the RCMDR dietary pattern in the older adults in Chinese communities still need to be explored. Therefore, we performed a 12-month cluster-randomized controlled intervention trial to evaluate the effects of the RCMDR dietary pattern on clustered cardiometabolic risk, common carotid artery (CCA) atherosclerosis parameters, CMD risk factors (blood lipids, blood glucose, insulin sensitivity, homocysteine (Hcy), blood pressure) and body composition in community-dwelling older adults. Material and methods Study design The single-center, single-blind, cluster randomized controlled trial was conducted in Beichen District, Tianjin, China. The study protocol (Supplement 1) was approved by the Ethics Committees of Tianjin Medical University (Approval number: TMUhMEC2023003). The trial followed the Measures for Ethical Review of Life Sciences and Medical Research Involving Human Beings and the Declaration of Helsinki. All participants provided written informed consent. The study was registered (Chinese Clinical Trial Registry; registration number: ChiCTR2300075022). The recruitment period occurred from August, 2023 until September, 2023. The final follow-up of the last participant was completed on October 2024. Study participants Participants should meet the following inclusion criteria: (1) people aged over 60, belonging to registered residence residents of Dazhang Zhuang town, Beichen district, Tianjin, and having no intention of moving out within one year; (2) capable of organizing and expressing language, easy to communicate, willing to receive health education, and able to independently cooperate with the researchers to complete on-site tests related to this study; (3) sign an informed consent form. Exclusion criteria included: (1) the presence of late stage of chronic disease, mental disorders or malignant tumors; (2) with serious urinary, circulatory, respiratory, digestive, rheumatic, immune, and endocrine system diseases, unable to cooperate with research; (3) with cognitive impairment, illiteracy, and language communication barriers; (4) difficulty moving or lying in bed, unable to cooperate with research. Randomization and blinding Participants from four residential areas in 2 communities according to their residential areas were cluster randomized to the RCMDR dietary pattern health education intervention or general health education control group. In this study, a community was defined as an administrative village within an administrative division, and a residential area refers to a residential cluster within a community that possesses clear geographical boundaries and unified property management. Each residential area serves as a randomized unit. Four residential areas were randomized in a 1:1 ratio to either the intervention group or the control group by using a random number generator (R version 4.3.1). Sealed envelope method was employed for random concealment. The study was single-blinded. Participants were blinded to other residential assignments. Also, physical examinations and health educations were scheduled at different times for participants in different residential areas. The health educations included one-on-one personalized dietary guidance and health lectures conducted in groups based on residential areas. In this study, health education for participants from different residential areas was strictly conducted separately to minimize the possibility of communication between participants from different clusters. Interventions The interventions flowchart for two groups is shown in Fig. 1. Participants in the RCMDR dietary pattern health education intervention group learned and adhered to the RCMDR dietary pattern through health education for 12 months; detailed information on the RCMDR dietary pattern has been published previously by us [12,13]. The health education provided by researchers included personalized dietary guidance and health education lectures. Personalized dietary guidance for the RCMDR dietary pattern was provided through face-to-face and telephone counseling, including guidance on adjusting the daily diet according to the food groups and prescribed intake in the RCMDR dietary pattern, and energy control for overweight or obese individuals to reduce weight. To help participants better learn and master the contents of the RCMDR dietary pattern, participants received a self-developed brochure by the researchers, which included the dangers of CMD, the diagram of the RCMDR dietary pattern, and examples of standard meals, etc. Additionally, after each health education lecture, participants in the intervention group were rewarded with foods that were emphasized in the RCMDR meal pattern, such as oats, buckwheat noodles, and unsalted mixed nuts. The control group received general health education based on “Dietary Guidelines for Chinese Residents (2022)” [14], and consisted of personalized dietary guidance and lectures at the same frequency. To improve adherence in the control group participants received non-food items as an incentive after each lecture. The intervention did not target total physical activity (PA) or sleep habits in either group. And medications were prescribed during the participants' routine medical care and were not affected by the intervention.Fig. 1Interventions flowchart for two groups.Fig. 1 Interventions flowchart for two groups. Outcomes and measurements The primary outcome was the clustered cardiometabolic disease risk score (CMRS), calculated in the same way as in previously published study using waist circumference (WC), systolic blood pressure (SBP), diastolic blood pressure (DBP), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), and fasting glucose (FBG)[13]. The secondary outcomes included CCA ultrasound parameters [carotid artery intima-media thickness (CIMT); resistance index (RI); peak systolic flow velocity (PSV)], blood lipid [total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), TG, and HDL-C], blood glucose and insulin sensitivity [FBG, fasting insulin (FINS), fasting C-peptide, HOMA-IR, and HOMA of β cell function (HOMA-β)], serum folic acid and Hcy, blood pressure (SBP and DBP), body composition indices [weight, BMI (kg/m2), WC, waist-to-hip ratio (WHR), body fat percentage, Chinese visceral adiposity index (CVAI), body fat mass, lean body mass]. The HOMA-IR was calculated as (fasting glucose [mmol/L] × fasting insulin [mU/L])/ 22.5, HOMA-β was calculated as (20 × fasting insulin [mU/L])/ (fasting glucose [mmol/L] – 3.5).[15] The CIMT values used for the analysis were the mean of the left and right sides maximum IMT values measured at the proximal and distal walls of the scanned CCA region,[16] and the RI and PSV values on both sides were also averaged. Plaque in CCA was defined as discrete IMT ≥ 1.5 mm, or focal thickening ≥ 0.5 mm, or 50% greater than the adjacent IMT in CCA.[17] The CVAI was calculated as follows: CVAI males = -267.93 + 0.68 × age (y) + 0.03 × BMI (kg/m2) + 4.00 × WC (cm) + 22.00 × Lg TG (mmol/L) − 16.32 × HDL-C (mmol/L); CVAI females = -187.32 + 1.71 × age (y) + 4.23 × BMI (kg/m2) + 1.12 × WC (cm) + 39.76 × Lg TG (mmol/L) − 11.66 × HDL-C (mmol/L)[18]. The data of primary and secondary outcomes were measured at baseline and end of intervention. The details of outcomes’ measurement methods are available in the study protocol in Supplement 1. Dietary and adherence assessment At baseline and the end of the intervention, the average daily intake of each food group of participants was calculated based on the dietary intake of 50 foods collected using a validated food frequency questionnaire within the past month [13], and the average daily energy and nutrient intake was calculated according to the available Chinese Food Composition Table [19,20]. The dietary adherence for all participants were assessed at baseline, and at the 1 st, 2nd, 4th, 6th, 9th, and 12th months, according to the participants’ 24 -h dietary records on the day before the follow-up visits and the intake of fish, poultry, livestock, eggs and added sugars during the week. These dietary records were collected by researchers during personalized dietary instruction. The RCMDR dietary pattern adherence scores ranging from 0-23 [13]. The detailed methods of dietary adherence assessment are available in the study protocol in Supplement 1. Covariates In order to comprehensively collect potential confounders, a structured questionnaire was designed to obtain the demographic characteristics (age, sex, educational level, living alone), lifestyles (smoking status, alcohol drinkers), family history of CMD, the use of nutritional supplements, medical history (diabetes, hypertension, dyslipidemia, cardiovascular disease) and therapy (lipid-lowering drugs, antihypertensive drugs, oral antidiabetics drugs, insulin, other cardiovascular disease drugs). Besides, the 10-item Pittsburgh Sleep Quality Index (PSQI) [21] and the Physical Activity Questionnaire Short Form [22] were performed to evaluate sleep quality and total PA, respectively. The above covariates data were collection at baseline and end of intervention. ApoE ε4 carriers were defined as individuals with ApoE genotypes ε2/ε4, ε3/ε4 and ε4/ε4; the MTHFR C677 T genotype was categorized into CC, CT and TT genotypes. The APOE and MTHFR genotypes were detected at baseline. Among the 14 covariates mentioned above, except for age, PSQI and total PA, which were continuous variables, the rest were categorical variables. Statistical analysis Based on an intervention efficacy size of reducing CMRS by 0.43, the intracluster correlation coefficient of 0.01,[23,24] two-sided α of 0.05, statistical power of 0.90, dropout rate of 15%, the required sample size was a minimum of 300 participants in each group, for a total of at least 600 participants. Sample sizes were calculated using the PASS 15 software. The analyses of outcomes followed the intention-to-treat (ITT) principle. Multiple imputations were performed for missing data. Continuous variables were expressed as mean ± standard deviation (SD), and using the t tests for baseline comparison between two groups. Categorical variables were shown as frequencies and proportions (percentages), and the χ2 test was used for baseline comparison between two groups. A linear mixed effects model was used to evaluated the intervention efficacy on primary and secondary outcomes based on maximum likelihood estimation. Group, time and a group × time interaction as fixed effects, residential areas and participant IDs as random effects were included in the model. A significant group × time interaction indicated a significant treatment effect. The models adjusted for social demographics (age, sex, education level, living alone); lifestyles (smoking status, alcohol drinkers, total PA, global PSQI score); family history of CMD, medical history (diabetes, hypertension, dyslipidemia, cardiovascular disease), therapy (lipid-lowering drugs, antihypertensive drugs, oral antidiabetics drugs, insulin, other cardiovascular disease drugs), nutritional supplements, ApoE ε4, MTHFR C677 T genotype in analyzing CMRS and body composition. The models additionally adjusted for BMI at baseline and endpoint when analyzing cardiometabolic risk factors indicators and CCA ultrasound parameters. The per-protocol (PP) analysis excluded 43 participants who were unable to finish the study. All analyses were performed using SPSS 24.0, with statistical significance set at P < 0.05 (two-tailed). The Benjamini-Hochberg false discovery rate (FDR) method was applied to adjust the P-value. The FDR < 0.05 was used to control for multiple testing. The Benjamini-Hochberg FDR method was performed using R version 4.3.1. Supplement 1 provides more detailed information about the statistical analysis plan. Results Baseline characteristics of participants A total of 694 individuals from two communities (mean [SD] age 68.2, [5.1] years; 411 (59.2%) female and 283 (40.8%) male) were enrolled. 349 individuals were cluster randomly assigned to the RCMDR dietary pattern health education intervention group, while 345 individuals were cluster randomly assigned to the general health education control group (Fig. 2). The number of participants in the two clusters of the intervention group was 173 and 176, while the number of participants in the two clusters of the control group was 172 and 173. The 694 participants were included in ITT analysis. During the study, 43 participants (21 for the intervention group and 22 for the control group) withdrew from the study or had missing measurements for some indicators, resulting in a loss to follow-up rate of 6.2%. Ultimately, a total of 651 participants who completed the entire study (328 for the intervention group and 323 for the control group) were included in the PP analysis. All characteristics of two groups were balance (all P > 0.05, Table 1).Fig. 2Flowchart of participantsNote: RCMDR, Reducing Cardiometabolic Diseases Risk; ITT, intention-to-treat; PP, per-protocol.Fig. 2Table 1Baseline characteristics of participantsTable 1CharacteristicsIntervention group (n = 349)Control group (n = 345)P valuesAge, mean ± SD, years68.1 ± 5.168.4 ± 5.10.384Female, n (%)212 (60.7)199 (57.7)0.412BMI, mean ± SD, kg/m226.3 ± 3.526.2 ± 3.70.827Education level, n (%)0.938Elementary school and below176 (50.4)175 (50.7)Middle school and above173 (49.6)170 (49.3)Living alone, n (%)31 (8.9)33 (9.6)0.756Smoking status, n (%)0.662Current smokers80 (22.9)82 (23.8)Former smokers55 (15.8)46 (13.3)Never214 (61.3)217 (62.9)Alcohol drinkers, n (%)68 (19.5)80 (23.2)0.234Total PA, mean ± SD, MET min/week2276.1 ± 1782.62463.3 ± 1755.20.164Global PSQI score, mean ± SD9.2 ± 3.59.3 ± 3.40.801Family History of CMDs, n (%)156 (44.7)141 (40.9)0.308Medical History, n (%)Hypertension205 (58.7)218 (63.2)0.230Diabetes71 (20.3)82 (23.8)0.277Dyslipidemia240 (68.8)218 (63.2)0.121Cardiovascular diseases64 (18.3)72 (20.9)0.401Carotid plaque, n (%)196 (56.2)184 (53.3)0.454Therapy, n (%)Lipid-lowering drugs39 (11.2)29 (8.4)0.220Antihypertensive drugs177 (50.7)155 (44.9)0.127Oral antidiabetics drugs36 (10.3)43 (12.5)0.373Insulin28 (8.0)28 (8.1)0.964Other cardiovascular disease drugs54 (15.5)63 (18.3)0.327Nutritional supplements, n (%)65 (18.6)64 (18.6)0.980ApoE ε4, n (%)63 (18.1)60 (17.4)0.820MTHFR C677 T genotype, n (%)0.492CC90 (25.8)76 (22.0)CT154 (44.1)163 (47.2)TT105 (30.1)106 (30.7)Abbreviation: BMI, body mass index; CMDs, cardiometabolic diseases; PA, physical activity; MET min/week, metabolic equivalent minutes per week; PSQI, Pittsburgh sleep quality index; ApoE, apolipoprotein E; MTHFR, methylenetetrahydrofolate reductase. Flowchart of participants Note: RCMDR, Reducing Cardiometabolic Diseases Risk; ITT, intention-to-treat; PP, per-protocol. Table 1Baseline characteristics of participantsTable 1CharacteristicsIntervention group (n = 349)Control group (n = 345)P valuesAge, mean ± SD, years68.1 ± 5.168.4 ± 5.10.384Female, n (%)212 (60.7)199 (57.7)0.412BMI, mean ± SD, kg/m226.3 ± 3.526.2 ± 3.70.827Education level, n (%)0.938Elementary school and below176 (50.4)175 (50.7)Middle school and above173 (49.6)170 (49.3)Living alone, n (%)31 (8.9)33 (9.6)0.756Smoking status, n (%)0.662Current smokers80 (22.9)82 (23.8)Former smokers55 (15.8)46 (13.3)Never214 (61.3)217 (62.9)Alcohol drinkers, n (%)68 (19.5)80 (23.2)0.234Total PA, mean ± SD, MET min/week2276.1 ± 1782.62463.3 ± 1755.20.164Global PSQI score, mean ± SD9.2 ± 3.59.3 ± 3.40.801Family History of CMDs, n (%)156 (44.7)141 (40.9)0.308Medical History, n (%)Hypertension205 (58.7)218 (63.2)0.230Diabetes71 (20.3)82 (23.8)0.277Dyslipidemia240 (68.8)218 (63.2)0.121Cardiovascular diseases64 (18.3)72 (20.9)0.401Carotid plaque, n (%)196 (56.2)184 (53.3)0.454Therapy, n (%)Lipid-lowering drugs39 (11.2)29 (8.4)0.220Antihypertensive drugs177 (50.7)155 (44.9)0.127Oral antidiabetics drugs36 (10.3)43 (12.5)0.373Insulin28 (8.0)28 (8.1)0.964Other cardiovascular disease drugs54 (15.5)63 (18.3)0.327Nutritional supplements, n (%)65 (18.6)64 (18.6)0.980ApoE ε4, n (%)63 (18.1)60 (17.4)0.820MTHFR C677 T genotype, n (%)0.492CC90 (25.8)76 (22.0)CT154 (44.1)163 (47.2)TT105 (30.1)106 (30.7)Abbreviation: BMI, body mass index; CMDs, cardiometabolic diseases; PA, physical activity; MET min/week, metabolic equivalent minutes per week; PSQI, Pittsburgh sleep quality index; ApoE, apolipoprotein E; MTHFR, methylenetetrahydrofolate reductase. Baseline characteristics of participants Abbreviation: BMI, body mass index; CMDs, cardiometabolic diseases; PA, physical activity; MET min/week, metabolic equivalent minutes per week; PSQI, Pittsburgh sleep quality index; ApoE, apolipoprotein E; MTHFR, methylenetetrahydrofolate reductase. Dietary adherence, dietary intake, PA and sleep At baseline, the mean of RCMDR dietary pattern adherence score was 5.25 in the intervention group and 5.02 in the control group, with no significant difference between the two groups (P =0.059). Dietary changes in both groups primarily occurred during the first and second months. After two months of intervention, participants in the intervention group increased the RCMDR dietary pattern adherence scores by 14.99 points, with a mean score of 20.24, which was maintained at this level over the next 10 months (Fig. 3).Fig. 3Dietary adherence score of RCMDR dietary pattern during 1 year of follow-up.Note: N = 694 (n intervention = 349, n control = 345) at baseline, 1st month and 2nd month; N = 687 (n intervention = 349, n control = 338) at 4th month; N = 685 (n intervention = 347, n control = 338) at 6th month; N = 677 (n intervention = 342, n control = 335) at 9th month; N = 660 (n intervention = 336, n control = 324) at 12th month.Values are expressed as mean ± SD. *P < 0.05 for comparisons between groups at each visit. #P < 0.05 for comparisons with baseline in each group. RCMDR, Reducing Cardiometabolic Diseases Risk.Fig. 3 Dietary adherence score of RCMDR dietary pattern during 1 year of follow-up. Note: N = 694 (n intervention = 349, n control = 345) at baseline, 1st month and 2nd month; N = 687 (n intervention = 349, n control = 338) at 4th month; N = 685 (n intervention = 347, n control = 338) at 6th month; N = 677 (n intervention = 342, n control = 335) at 9th month; N = 660 (n intervention = 336, n control = 324) at 12th month. Values are expressed as mean ± SD. *P < 0.05 for comparisons between groups at each visit. #P < 0.05 for comparisons with baseline in each group. RCMDR, Reducing Cardiometabolic Diseases Risk. The average daily intake of nutrients and food groups at baseline between two groups were comparable (all P >0.05). The actual average nutrient intake and food intake of the intervention group participants were similar to the goals of the RCMDR dietary pattern. Compared to the control group, participants in the intervention group significantly increased consumption of whole grains, legumes, total vegetables, dark green leafy and cruciferous species vegetables, other dark vegetables, fruits, fish, dairy products, soybeans and nuts at the end of intervention, and significantly decreased consumption of cereals, other vegetables, food of animal origin, livestock meat, eggs, cooking oil, salt, added sugar and sugar-sweetened beverages (all P < 0.05, adjusted P < 0.05, Table S1 in Supplement 2). Compared to the control group, participants in the intervention group significantly increased consumption of energy distribution of carbohydrates and protein, fiber, potassium, magnesium, calcium and folate, and significantly decreased consumption of energy, energy distribution of fat, cholesterol, sodium (all P < 0.05, adjusted P < 0.05, Table S2 in Supplement 2). The total PA did not change in either group from baseline to the end of the intervention (all P >0.05). Both groups had lower PSQI at the end of the intervention than at baseline (both P Within-group <0.001), with the similar extent of the reduction (β = -0.22, 95% CI -0.79, 0.36; P = 0.461) (Supplementary Table S3 in Supplement 2). Efficacy of RCMDR diet on CMRS After 12 months intervention, both groups had significant reductions in CMRS (primary outcome) compared to baseline (both P Within-group <0.001), with mean reductions of -0.35 standardized units in the intervention group and -0.10 standardized units in the control group. Compared with control group, the intervention group had statistically significant decrease in CMRS (β = -0.25, 95% CI -0.31, -0.19; P < 0.001, adjusted P < 0.001), after adjusting for potential confounders (Fig. 4A).Fig. 4Longitudinal analysis of intervention on primary and secondary outcomes between two groups.Note: Panel A, Longitudinal analysis of intervention on CMRS, CCA ultrasound parameters and blood indicators between two groups. Panel B, Longitudinal analysis of intervention on blood pressure and body composition between two groups.The figure was analyzed using the ITT principle.a Linear mixed-effects models adjusted for social demographics (age, sex, education level, living alone, BMI); lifestyles (smoking status, alcohol drinkers, total PA, global PSQI score); family history of CMD, medical history (diabetes, hypertension, dyslipidemia, cardiovascular disease), therapy (lipid-lowering drugs, antihypertensive drugs, oral antidiabetics drugs, insulin, other cardiovascular disease drugs), nutritional supplements, ApoE ε4, MTHFR C677 T genotype. β (95% CI) and p values represent the interaction term (group × time) in the model.b Linear mixed-effects models adjusted the above variables except for BMI.c The data was transformed by multiplying by 10.d The data was transformed by multiplying by 0.1.e The data were square root transformed.f The data was transformed by multiplying by 100.gP value was adjusted by Benjamini-Hochberg FDR method.Abbreviation: CMRS, cardiometabolic risk score; CCA, common carotid artery; CIMT, carotid artery intima-media thickness; RI, resistance index; PSV, peak systolic flow velocity; TC, total cholesterol; TG, triglyceride; HDL-C, High-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; FBG, fasting blood glucose; FINS, fasting insulin; HOMA-IR, homeostasis model assessment–estimated insulin resistance; HOMA-β, HOMA of β-cell function; Hcy, homocysteine; SBP, systolic blood pressure; DBP, diastolic blood pressure; CVAI, Chinese visceral adiposity index, WHR, waist-to-hip ratio; CI, confidence intervals; BMI, body mass index; CMD, cardiometabolic disease; PA, physical activity; PSQI, Pittsburgh sleep quality index; ApoE, apolipoprotein E; MTHFR, methylenetetrahydrofolate reductase; ITT, intention-to-treat; FDR, false discovery rate.Boldface indicates statistical significance (P < 0.05).Fig. 4 Longitudinal analysis of intervention on primary and secondary outcomes between two groups. Note: Panel A, Longitudinal analysis of intervention on CMRS, CCA ultrasound parameters and blood indicators between two groups. Panel B, Longitudinal analysis of intervention on blood pressure and body composition between two groups. The figure was analyzed using the ITT principle. a Linear mixed-effects models adjusted for social demographics (age, sex, education level, living alone, BMI); lifestyles (smoking status, alcohol drinkers, total PA, global PSQI score); family history of CMD, medical history (diabetes, hypertension, dyslipidemia, cardiovascular disease), therapy (lipid-lowering drugs, antihypertensive drugs, oral antidiabetics drugs, insulin, other cardiovascular disease drugs), nutritional supplements, ApoE ε4, MTHFR C677 T genotype. β (95% CI) and p values represent the interaction term (group × time) in the model. b Linear mixed-effects models adjusted the above variables except for BMI. c The data was transformed by multiplying by 10. d The data was transformed by multiplying by 0.1. e The data were square root transformed. f The data was transformed by multiplying by 100. gP value was adjusted by Benjamini-Hochberg FDR method. Abbreviation: CMRS, cardiometabolic risk score; CCA, common carotid artery; CIMT, carotid artery intima-media thickness; RI, resistance index; PSV, peak systolic flow velocity; TC, total cholesterol; TG, triglyceride; HDL-C, High-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; FBG, fasting blood glucose; FINS, fasting insulin; HOMA-IR, homeostasis model assessment–estimated insulin resistance; HOMA-β, HOMA of β-cell function; Hcy, homocysteine; SBP, systolic blood pressure; DBP, diastolic blood pressure; CVAI, Chinese visceral adiposity index, WHR, waist-to-hip ratio; CI, confidence intervals; BMI, body mass index; CMD, cardiometabolic disease; PA, physical activity; PSQI, Pittsburgh sleep quality index; ApoE, apolipoprotein E; MTHFR, methylenetetrahydrofolate reductase; ITT, intention-to-treat; FDR, false discovery rate. Boldface indicates statistical significance (P < 0.05). Efficacy of RCMDR diet on CCA ultrasound parameters Compared to the control group, the RCMDR dietary pattern intervention significantly prevented the increment in CIMT (β = -0.05, 95% CI -0.08, -0.02; P = 0.001, adjusted P = 0.001) after 12 months of intervention (Fig. 4A). Compared with baseline, the mean CIMT increased from 1.51 to 1.56 mm in the control group (P Within-group < 0.001), while no difference was observed in the intervention group (P Within-group = 0.742) (Table S4 in Supplement 2). However, there were no significant differences in the changes of RI and PSV from baseline to the end of the intervention between the intervention group and the control group (all P > 0.05) (Fig. 4A). Efficacy of RCMDR diet on cardiometabolic risk factors The RCMDR dietary pattern intervention reduced in TC (β = -0.28, 95% CI -0.43, -0.13; P < 0.001, adjusted P < 0.001), TG (β = -0.26, 95% CI -0.45, -0.08; P = 0.006, adjusted P = 0.007), LDL-C (β = -0.29, 95% CI -0.42, -0.16; P < 0.001, adjusted P < 0.001) and increased HDL-C (β = 0.10, 95% CI 0.04, 0.16; P = 0.001, adjusted P = 0.001), compared to no change of blood lipid in the control group (all P within-group > 0.05) (Fig. 4A and Table S4 in Supplement 2). The RCMDR dietary pattern intervention also exhibited superior effects on in lowering FBG (β = -0.40, 95% CI -0.52, -0.28; P < 0.001, adjusted P < 0.001), FINS (β = -0.64, 95% CI -0.85, -0.44; P < 0.001, adjusted P < 0.001), fasting C-peptide (β = -0.24, 95% CI -0.29, -0.19; P < 0.001, adjusted P < 0.001), HOMA-IR (β = -0.33, 95% CI -0.40, -0.25; P < 0.001, adjusted P < 0.001) and increasing HOMA-β (β = 0.38, 95% CI 0.21, 0.55; P < 0.001, adjusted P < 0.001) compared with the control group (Fig. 4A). Compared with control group, the greater rise in serum folate (β = 0.27, 95% CI 0.19, 0.35; P < 0.001, adjusted P < 0.001) and decrease in Hcy (β = -0.33, 95% CI -0.39, -0.26; P < 0.001, adjusted P < 0.001) were observed in the intervention group. (Fig. 4A). Moreover, at the end of the intervention, both groups presented reductions in SBP from baseline, with a greater decline in the intervention group (β = -0.36, 95% CI -0.63, -0.10; P = 0.008, adjusted P = 0.009); DBP decreased only in the intervention group (β = -0.20, 95% CI -0.36, -0.05; P = 0.011, adjusted P = 0.012) (Fig. 4B and Table S5 in Supplement 2). Efficacy of RCMDR diet on body composition Compared to the control group, the intervention group had statistically significant decreases in weight (β = -1.52, 95% CI -1.94, -1.10; P < 0.001, adjusted P < 0.001), BMI (β = -0.53, 95% CI -0.73, -0.33; P < 0.001, adjusted P < 0.001), WC (β = -2.19, 95% CI -2.85, -1.52; P < 0.001, adjusted P < 0.001), WHR (β = -1.54, 95% CI -2.14, -0.93; P < 0.001, adjusted P < 0.001), body fat percentage (β = -2.68, 95% CI -3.20, -2.16; P < 0.001, adjusted P < 0.001), body fat mass (β = -2.41, 95% CI -2.87, -1.95; P < 0.001, adjusted P < 0.001), CVAI (β = -0.87, 95% CI -1.11, -0.63; P < 0.001, adjusted P < 0.001) (Fig. 4B). Meanwhile, compared to the control group, the intervention group had statistically significant increases in lean body mass (β = 0.93, 95% CI 0.68, 1.18; P < 0.001, adjusted P < 0.001) (Fig. 4B). Sensitivity analysis After excluding the 43 participants who failed to complete the study, the PP analysis was performed. In the PP analysis, the results of both the crude model and the model adjusted for confounders remained robust and were all in accordance with the results reported in the ITT analysis (Figure S1 in Supplement 2). Discussion In this cluster randomized intervention study of community-dwelling older people, we demonstrated that the RCMDR dietary pattern had better effects on reducing CMRS, delaying the progression of atherosclerosis in the CCA, and improving multiple cardiometabolic risk indicators, including blood lipids, blood glucose, insulin resistance, Hcy and blood pressure, compared with general health education. In addition, following the RCMDR dietary pattern was more effective at reducing abdominal obesity, lowering body fat, especially visceral fat, while increasing muscle mass. After our previously published randomized controlled intervention study showing the cardiometabolic health benefits of the RCMDR dietary pattern for people aged 35-45 with dyslipidemia [13], this trial further suggests that the RCMDR dietary pattern might be an effective and potential nutritional tool to prevent the occurrence and development of CMDs in Chinese populations. In this study, cardiometabolic protective effects were still observed in Chinese community-dwelling older adults following the RCMDR dietary pattern. In addition to the favorable effects of the RCMDR dietary pattern in reducing the risk of CMDs, improving blood lipids, Hcy, blood pressure and abdominal obesity as reported in the previous intervention study [13], the present study found that the RCDMR diet also had a better lowering of glycemic and insulin resistance effects. Several clinical studies have also found similar association between healthy diet and glucose metabolism, such as DASH diet [25], healthy plant-based diet [26], low-fat vegan diet [27]. The protective effect of the RCMDR dietary pattern in glucose metabolism might be mainly attributed to the fact that the RCMDR dietary pattern increased intake of dietary fiber [28], magnesium [29] and vitamin C [30] by increasing intake of plant foods such as whole grains, vegetables and fruits. Our study found that following the RCMDR dietary pattern slowed the growth of IMT in CCA. CIMT is an ultrasound biomarker of early atherosclerosis and a strong predictor of future vascular events [31]. A large number of studies have reported similar associations between healthy dietary factors on markers related to atherosclerosis progression as the results of the present study, such as the Mediterranean diet, high-fiber dietary patterns rich in plant foods, restriction of fats and processed foods [[32], [33], [34], [35]]. The protectiveness of the RCMDR dietary pattern against atherosclerosis might be explained by several mechanisms. First, the RCMDR dietary pattern advocated low-fat or skimmed dairy products, cooking with less oils, limiting animal-derived foods, especially red meat, banning processed meat and animal oils, and restricting the intake of added sugars, which greatly decreased the intake of exogenous advanced glycation end-products (AGEs) [36]. Many studies have proven that the accumulation of AGEs provoked vascular damage and exacerbated the development of atherosclerosis and plaque formation [37,38]. Second, the RCMDR dietary pattern highlighted increasing the intake of vegetables and fruits, especially dark and cruciferous vegetables. A cohort study of older women reported the higher intakes of vegetable and cruciferous vegetables were associated with lower IMT in CCA [39]. Sulforaphane, a naturally-occurring isothiocyanate enriched in cruciferous vegetables, might produce vasculoprotective effects by modulating the antioxidant and anti-inflammatory pathways in endothelial cells [40]. Dietary intake of polyphenols, carotenoids, vitamin C and vitamin E, which are abundant in dark vegetables and fruits, have been reported to be negatively correlated with CIMT. These phytochemicals and vitamins had antioxidant and anti-inflammatory effects [[41], [42], [43]], and reduced the production and accumulation of endogenous AGEs [44,45]. Third, The RCMDR dietary pattern increases the intake of soy beans and its products. A Chinese cohort study of middle-aged and older adults showed that the serum concentrations of isoflavones, the objective biomarkers of soy intake, were negatively correlated with increments in CIMT [46]. Strengths and limitations This study has several strengths. First, contamination effects were minimized by the trial's cluster-randomized interventions and single-blind design. Second, the dual-channel intervention approach, using a combination of personalized dietary guidance and health education lectures, could better assist participants in learning and mastering the RCMDR dietary pattern and improve the applicability and generalization of the RCMDR dietary pattern in the population. Third, throughout the intervention, we provided strict and standardized supervision, resulting in high adherence rates and low dropout rates among the participants. Besides, possible confounding factors were adjusted, which could reduce confounding bias to some extent. However, this study has limitations that need to be pointed out. First, the study was conducted at a single center, which would limit extrapolation of these results. Second, considering implementability, we evaluated only the mean maximum IMT of the CCA and did not assess the mean IMT and the number and area of plaques, so future studies involving more indicators of carotid atherosclerosis are needed. Third, longer follow-up is needed because metabolic and atherosclerotic changes were slow processes. Finally, some covariates were based on self-reports from participants through questionnaires, including lifestyle factors such as smoking status and alcohol consumption, which may introduce reporting bias. Conclusion In this trial of Chinese community-dwelling older people, following the RCMDR dietary pattern could reduce their risk of CMD; delay the progression of CCA atherosclerosis; and improve cardiometabolic risk factors including blood lipids, blood glucose, Hcy and blood pressure. In addition, following the RCMDR dietary pattern could reduce body and visceral fat while increasing muscle mass. Authors’ contribution Qi Wu, Li Huang: Writing - Original Draft, Investigation, Writing - Review & Editing. Jie Yang: Investigation, Writing - Review & Editing. Liyang Zhang, Cheng Cheng, Xukun Chen, Tongtong Li, Tongyang Wu, Yue Wang, Huichao Qu, Ya Su, Baoguo Gao, Wei Liu, Lichun Cao: Investigation. Huilian Duan, Zehao Wang, Qinghan Ren: Data Curation, Formal analysis. Shanshan Bian, Zhenshu Li, Huan Liu, Jing Yan, Yue Du: Writing - Review & Editing. Yongjie Chen: Methodology, Writing - Review & Editing. Wen Li, Fei Ma: Conceptualization, Project administration, Writing - Review & Editing. Guowei Huang: Conceptualization, Methodology, Project administration, Supervision, Funding acquisition, Writing - Review & Editing. All authors read and approved the final version of the manuscript. Declaration of Generative AI and AI-assisted technologies in the writing process No Generative AI or AI -assisted technologies were used in the writing process. Funding information This research was supported by the National Nutrition Science Research Grant of Chinese Nutrition Society (grant number: CNS-NNSRG2022-149). Statement of ethics This study followed the Helsinki declaration. Participants from the study signed an informed consent form. The study was approved by the Ethics Committees of Tianjin Medical University (Approval number: TMUhMEC2023003). Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Data Sharing Statement The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.