Association between waist-to-height ratio and cardiovascular events in patients with chronic obstructive pulmonary disease: the mediating effect of prognostic nutritional index Authors: Yuli Cai, Enlin Ye, Wang Liu, Shuyue Zhou, Xiaomi Chen, Huan Li, Guanhuan Chen, Yihan Zhou, Heyu Liu, Mengya Yi, Yuxin Guo, Qianqian Pan, Chengheng Tan, Hui Lin, Junfen Cheng, Lijuan Zeng, Jinrong Xu, Riken Chen Published: 2025 ABSTRACT AbstractObjectivesTo explore the relationship between waist-to-height ratio (WHtR) and the risk of cardiovascular disease (CVD) in patients with chronic obstructive pulmonary disease (COPD), and to analyze the mediating role of prognostic nutritional index (PNI) in this association.MethodsThis study was based on the National Health and Nutrition Examination Survey (NHANES) database from 2007 to 2018, including a total of 1,013 eligible COPD patients. Multivariate logistic regression analysis was used to evaluate the association between WHtR and CVD, and restricted cubic spline regression was applied to explore their non-linear relationship. Meanwhile, mediating effect analysis (Hayes Process Macro Model 4) was conducted to investigate the mediating role of PNI between WHtR and CVD.ResultsWHtR was an independent risk factor for cardiovascular disease in COPD patients (OR = 5.76, P = 0.028). Restricted cubic spline results showed a positive correlation between WHtR and CVD risk without a significant non-linear relationship (P for non-linearity = 0.247). Mediation analysis revealed that PNI played a partial mediating role between WHtR and CVD (the indirect effect accounted for 7.98% of the total effect, P < 0.001), suggesting that WHtR may increase CVD risk by affecting the body’s nutritional status.ConclusionWHtR is an important predictor of CVD in COPD patients, and PNI exerts a partial mediating effect between them. Controlling WHtR and improving nutritional status may help reduce the risk of cardiovascular disease in COPD patients.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12872-025-05286-3.Keywords: Chronic obstructive pulmonary disease, Waist-to-Height ratio, Cardiovascular disease, Prognostic nutritional index, Mediation analysis Background Chronic Obstructive Pulmonary Disease (COPD) is one of the leading causes of global mortality and disability, characterized by persistent airflow limitation, chronic inflammatory response, and multisystem involvement. According to the World Health Organization, by 2020, COPD had become the third leading cause of death worldwide [1]. In addition to respiratory symptoms, COPD patients often have multiple comorbid cardiovascular diseases (CVD), such as coronary heart disease, heart failure, and arrhythmia, which increase the risk of morbidity and exacerbate disease burden [2]. In recent years, the mechanisms underlying COPD and CVD comorbidity have attracted extensive attention, with inflammation, oxidative stress, metabolic disorders, and abnormal nutritional status considered as important common pathological bases [3]. Among various metabolic indicators, waist-to-height ratio (WHtR), a simple index reflecting abdominal obesity, has gradually shown advantages over body mass index (BMI) in predicting cardiovascular risk [4]. Compared with BMI, WHtR is more sensitive to reflecting visceral fat accumulation, which has been confirmed to be closely related to atherosclerosis, insulin resistance, and chronic inflammation [5]. Some studies have indicated that elevated WHtR is closely associated with decreased lung function and increased risk of all-cause mortality in COPD patients, However, studies on the value of WHtR in assessing cardiovascular risk in the COPD population are still limited, especially lacking systematic verification in large, representative population data. On the other hand, nutritional and inflammatory status is an important bridge connecting COPD and CVD. The prognostic nutritional index (PNI), calculated from serum albumin concentration and peripheral blood lymphocyte count, is a comprehensive index for evaluating individual nutritional and immune status, and has been widely used in predicting the prognosis of patients with tumors, cardiovascular diseases, and chronic diseases [6–8]. Studies have suggested that decreased PNI is closely related to an increased risk of COPD exacerbation, higher hospitalization rates, and poor CVD prognosis [9, 10]. However, no studies have explored whether it plays a mediating role in the relationship between WHtR and COPD complicated with CVD. Based on the above background, this study intends to use data from the National Health and Nutrition Examination Survey (NHANES) from 2007 to 2018 to explore the association between WHtR and the risk of CVD in COPD patients, and further analyze whether PNI plays a mediating role between them. The results of this study are expected to reveal the potential mechanism by which abdominal obesity affects cardiovascular risk through nutritional-inflammation pathways, and provide new ideas and evidence for the early identification and comprehensive intervention of COPD complicated with CVD. Materials and methods Data source Data were derived from the National Health and Nutrition Examination Survey (NHANES), a cross-sectional survey of U.S. residents. The NHANES database systematically collects health-related data from a nationally representative sample of the U.S. population [11]. NHANES is a collaboration between the U.S. Centers for Disease Control and Prevention (CDC) and the National Center for Health Statistics (NCHS). The NCHS Research Ethics Review Board granted ethical approval for the NHANES protocol, and written informed consent was obtained from each participant (Source: https://www.cdc.gov/nchs/nhanes/index.htm). Population selection To study the association between waist-to-height ratio and cardiovascular events in COPD patients, as well as the mediating effect of prognostic nutritional index, only participants diagnosed with COPD were included. Therefore, data were selected from six interview cycles of the NHANES database from 2007 to 2018, with 59,842 participants reviewed. After excluding non-COPD participants, 1,199 participants were initially included. In addition, participants under 20 years old and those with missing data on PNI, WHtR, etc., were further excluded. Finally, a total of 1,013 COPD patients (639 males and 374 females) were included in the statistical analysis (Fig. 1). Flowchart of participant inclusion Identification of COPD patients and cardiovascular disease COPD was determined by a combination of lung function or self-reported physician diagnosis and completion of a standardized medical condition questionnaire. COPD patients collected from the NHANES database were confirmed by a post-bronchodilator FEV1/FVC < 0.7 from 2007 to 2012, and by the medical condition questionnaire “Has a doctor ever told you that you have chronic obstructive pulmonary disease?” (MCQ160O) from 2013 to 2018 [12, 13]. CVD was determined by a combination of self-reported physician diagnosis and completion of a standardized medical condition questionnaire. Participants were asked five different questions: “Has a doctor or other healthcare professional ever told you that you have congestive heart failure/coronary heart disease/angina/heart attack/stroke?” Participants were identified as CVD patients if they answered “yes” to any of the five individual questions. We excluded all subjects who indicated “don’t know” or refused to answer.The survey questionnaires are all sourced from the NHANES database (Source: https://www.cdc.gov/nchs/nhanes/index.htm). Variable collection Data of the subjects were collected from the NHANES database, including general information: age, gender, race, education level, BMI, diabetes, hypertension; laboratory data: low-density lipoprotein, high-density lipoprotein, total cholesterol, triglycerides, lymphocytes, serum albumin. WHtR = waist circumference/height, which is an index commonly used to assess abdominal obesity and the risk of chronic diseases (such as cardiovascular disease) [14]. PNI is a new immunonutritional risk index calculated based on albumin and total lymphocyte count to evaluate disease prognosis, with the formula: PNI = serum albumin (g/L) + 5 * total lymphocyte count (10⁹/L) [10]. Statistical methods All analyses were performed using R software (version 4.4.3) and SPSS 25.0. Statistical analyses of data were weighted according to NHANES guidelines, and multiple imputation was used to address missing covariate data. Continuous variables with a normal distribution were expressed as mean [standard deviation (SD)], and independent samples t-test was used to evaluate differences between groups. Categorical variables were expressed as percentages, and chi-square test was used to assess differences between groups. Logistic regression models were used to determine the odds ratio (OR) and 95% confidence interval (CI) associated with WHtR and cardiovascular disease in COPD patients. The non-linear association between WHtR and cardiovascular disease was analyzed by restricted cubic spline regression in R 4.4.3 software. The mediating effect model was implemented using the SPSS macro program Process to analyze the mediating effect of PNI. The test level α = 0.05, and p < 0.05 was considered statistically significant. Results General characteristics of participants A total of 59,842 participants were initially included in this study. After excluding non-COPD participants, those under 20 years old, and those with missing key variables such as PNI and WHtR, 1,013 subjects were finally included in the statistical analysis. Baseline analysis of the study population is shown in Table 1. The average age of the participants was 61.54 ± 12.82 years, with 639 males (63.08%) and 374 females (36.92%). Among them, there were 297 CVD participants and 716 non-CVD participants. Statistically significant differences were observed between the CVD and non-CVD groups in terms of average age, height, weight, education level, BMI, WHtR, LDL, HDL, TC, PNI, smoking, diabetes, and hypertension (p < 0.05). Table 1Basic characteristics and differential analysisVariablesTotal(n = 1013)Non-CVD(n = 716)CVD(n = 297)StatisticPAge(years), n(%)61.54 ± 12.8259.52 ± 13.0566.40 ± 10.81χ²=46.29< 0.001 <60407 (40.18)336 (46.93)71 (23.91) ≥ 60606 (59.82)380 (53.07)226 (76.09)Gender, n(%)χ²=0.230.632 Male639 (63.08)455 (63.55)184 (61.95) Female374 (36.92)261 (36.45)113 (38.05) Height (cm)169.31 ± 9.90169.97 ± 9.82167.73 ± 9.92t = 3.300.001 Weight (kg)83.32 ± 22.2782.24 ± 21.5185.93 ± 23.84t=−2.410.016Race, n(%)χ²=2.060.724 Mexican48 (4.74)37 (5.17)11 (3.70) Hispanics46 (4.54)33 (4.61)13 (4.38) Non-Hispanic White686 (67.72)478 (66.76)208 (70.03) Non-Hispanic Black159 (15.70)117 (16.34)42 (14.14) Others74 (7.31)51 (7.12)23 (7.74)Education, n(%)χ²=6.520.011 ≤High school558 (55.08)376 (52.51)182 (61.28) >high school455 (44.92)340 (47.49)115 (38.72)BMI(kg/m2), n(%)29.02 ± 7.2628.41 ± 6.8530.50 ± 7.99χ²=13.66< 0.001 < 25307 (30.31)237 (33.10)70 (23.57) ≥ 25 和 < 30339 (33.46)243 (33.94)96 (32.32) ≥ 30367 (36.23)236 (32.96)131 (44.11)LDL(mmol/L)2.84 ± 0.712.90 ± 0.682.71 ± 0.75t = 3.82< 0.001HDL (mmol/L)1.35 ± 0.461.38 ± 0.481.28 ± 0.42t = 3.260.001TC (mmol/L)4.91 ± 1.135.04 ± 1.094.58 ± 1.16t = 6.01< 0.001TG(mmol/L)1.78 ± 1.151.74 ± 1.161.89 ± 1.13t=−1.850.064WHtR1.03 ± 1.060.96 ± 1.061.18 ± 1.06t=−3.010.003PNI51.76 ± 5.3552.33 ± 5.1250.37 ± 5.63t = 5.38< 0.001Smoking, n(%)χ²=8.500.004 Yes820 (80.95)563 (78.63)257 (86.53) No193 (19.05)153 (21.37)40 (13.47)Drinking, n(%)χ²=0.830.361 Yes875 (86.38)623 (87.01)252 (84.85) No138 (13.62)93 (12.99)45 (15.15)Diabetes, n(%)χ²=45.33< 0.001 Yes218 (21.52)114 (15.92)104 (35.02) No767 (75.72)581 (81.15)186 (62.63) Borderline28 (2.76)21 (2.93)7 (2.36)Hypertension, n(%)χ²=97.54< 0.001 Yes527 (52.02)301 (42.04)226 (76.09) No486 (47.98)415 (57.96)71 (23.91)BMI body mass index, LDL low-density lipoprotein cholesterol, HDL high-density lipoprotein cholesterol, TC total cholesterol, TG triglycerides, WHtR waist-to-height ratio, PNI prognostic nutritional Index Basic characteristics and differential analysis Relationship between WHtR and incidence of cardiovascular disease Variables with P < 0.05 in the baseline analysis were included in the logistic regression analysis, and indicators were screened using the Wald chi-square test (two-tailed: Wald). Finally, a multivariate logistic regression model was constructed. Since BMI, as a comprehensive index of body size, was included in the regression model, height and weight were only used for baseline characteristic description and not included in the model alone; similarly, given that this study focused on exploring the mediating pathway of PNI between WHtR and CVD, PNI was only used for mediating effect analysis to avoid over-adjustment of the model or collinearity between variables interfering with the evaluation of the main effect. As shown in Table 2, in the unadjusted analysis, factors associated with the incidence of cardiovascular disease included age, education level, BMI, WHtR, LDL, HDL, TC, smoking, diabetes, and hypertension. After adjusting for important variables in the unadjusted analysis, WHtR remained an important predictor of complicated cardiovascular disease. WHtR (p = 0.028; OR: 5.76) was considered an independent risk factor for complicated cardiovascular disease in COPD patients. Table 2Risk of cardiovascular disease in all study participantsVariablesUnadjusted modeladjusted modelβS.EZPOR (95%CI)βS.EZPOR (95%CI)Age <601.00 (Reference)1.00 (Reference) ≥ 601.030.166.66< 0.0012.81(2.08 ~ 3.82)0.660.173.92< 0.0011.94(1.39 ~ 2.69)Education ≤High school1.00 (Reference)1.00 (Reference) >high school−0.360.14−2.550.0110.70(0.53 ~ 0.92)−0.280.15−1.830.0670.75(0.56 ~ 1.02)BMI < 251.00 (Reference) ≥ 25 和 < 300.290.181.600.1101.34(0.94 ~ 1.91) ≥ 300.630.173.62< 0.0011.88(1.34 ~ 2.64) WHtR3.980.675.93< 0.00153.29(14.31 ~ 198.36)1.750.802.200.0285.76(1.21 ~ 27.44) LDL−0.410.11−3.78< 0.0010.66 (0.54 ~ 0.82) HDL−0.550.17−3.230.0010.58(0.42 ~ 0.81) TC−0.390.07−5.82< 0.0010.68(0.59 ~ 0.77)−0.240.07−3.37< 0.0010.79(0.68 ~ 0.90)Smoking Yes1.00 (Reference)1.00 (Reference) No−0.560.19−2.890.0040.57(0.39 ~ 0.84)−0.650.21−3.060.0020.52(0.35 ~ 0.79)Diabetes Yes1.00 (Reference)1.00 (Reference) No−1.050.16−6.56< 0.0010.35(0.26 ~ 0.48)−0.430.19−2.280.0230.65(0.45 ~ 0.94)Borderline−1.010.46−2.200.0280.37(0.15 ~ 0.89)−0.920.49−1.890.0580.40(0.15 ~ 1.03)Hypertension Yes1.00 (Reference)1.00 (Reference) No−1.480.16−9.50< 0.0010.23(0.17 ~ 0.31)−1.150.17−6.89< 0.0010.32(0.23 ~ 0.44) Risk of cardiovascular disease in all study participants 2.81 (2.08 ~ 3.82) 1.94 (1.39 ~ 2.69) 0.70 (0.53 ~ 0.92) 0.75 (0.56 ~ 1.02) 1.34 (0.94 ~ 1.91) 1.88 (1.34 ~ 2.64) 53.29 (14.31 ~ 198.36) 5.76 (1.21 ~ 27.44) 0.58 (0.42 ~ 0.81) 0.68 (0.59 ~ 0.77) 0.79 (0.68 ~ 0.90) 0.57 (0.39 ~ 0.84) 0.52 (0.35 ~ 0.79) 0.35 (0.26 ~ 0.48) 0.65 (0.45 ~ 0.94) 0.37 (0.15 ~ 0.89) 0.40 (0.15 ~ 1.03) 0.23 (0.17 ~ 0.31) 0.32 (0.23 ~ 0.44) Restricted cubic spline plot of WHtR and CVD in COPD patients The abscissa represents WHtR in COPD patients, and the ordinate represents the relative risk of COPD patients complicated with CVD. The results showed that the overall P < 0.001 and the non-linear P = 0.247 > 0.05, indicating no non-linear relationship between WHtR and CVD risk in COPD patients. According to the results in Fig. 2, when WHtR < 0.601, the risk of COPD patients complicated with CVD changed slowly with the increase of WHtR; when WHtR = 0.601, the cut-off value of OR ≈ 1; when WHtR > 0.601, the risk of CVD increased significantly with the increase of WHtR. Therefore, it is appropriate to control the WHtR of COPD patients below 0.601. Restricted cubic spline plot of waist-to-height ratio(WHtR)and COPD patients complicated with CVD Mediating role of PNI between WHtR and CVD in COPD patients To explore the potential mechanism of the significant positive effect of WHtR on CVD, PNI was further introduced as a mediating variable to replace the structural equation model of this study. Model 4 in the SPSS macro program Process was used to test the mediating effect, and the bootstrap method by Hayes was used to verify and analyze the mediating role of PNI between WHtR and COPD patients complicated with CVD. The PNI path coefficients between WHtR and cardiovascular risk in COPD patients are shown in Fig. 3. Path analysis confirmed the existence of this mediating mechanism: WHtR significantly negatively predicted PNI (β = −5.17, P = 0.001), suggesting that elevated waist-to-height ratio is associated with decreased individual nutritional status; meanwhile, the effect of PNI on CVD was also significantly negative (β = −0.07, P < 0.001), indicating that poor nutritional status may increase the risk of CVD. Therefore, PNI played a partial mediating role between WHtR and CVD, which, although accounting for a relatively small proportion, was statistically significant. Path coefficient diagram of WHtR, prognostic nutritional index(PNI), and CVD According to Table 3, the analysis results of the mediating effect of PNI on cardiovascular disease (CVD) and PNI in COPD patients showed that the 95% confidence interval of Bootstrap did not contain 0, indicating that the effect of WHtR on CVD not only had a direct effect but also a significant indirect effect through PNI. Specifically, the direct effect of WHtR was 0.56, accounting for 92.02% of the total effect (0.61); the indirect effect was 0.05, accounting for 7.98% of the total effect. The effects of both paths were statistically significant (both p < 0.001). The results further confirmed that PNI played a partial mediating role between WHtR and CVD, suggesting that abnormal body size may indirectly increase the risk of cardiovascular disease by affecting the body’s nutritional and inflammatory status. Table 3Breakdown of total effect, direct effect, and mediating effectEffectEfficiency Value95%CIPEffect sizeIndirect0.050.02 ~ 0.08< 0.0017.98%Direct0.560.42 ~ 0.68< 0.00192.02%Total0.610.49 ~ 0.72< 0.001100.00% Breakdown of total effect, direct effect, and mediating effect Discussion Based on the U.S. NHANES database from 2007 to 2018, this study explored the relationship between waist-to-height ratio (WHtR) and the risk of cardiovascular disease (CVD) in patients with chronic obstructive pulmonary disease (COPD), and for the first time evaluated the mediating role of prognostic nutritional index (PNI) in this relationship. The results showed that WHtR was an independent risk factor for COPD patients complicated with CVD, and PNI played a partial mediating role between WHtR and CVD, suggesting that nutritional status may be one of the important mechanisms by which abdominal obesity affects CVD risk. Firstly, this study found a positive correlation between WHtR and CVD risk, i.e., the higher the WHtR, the greater the risk of COPD patients complicated with CVD. After adjusting for confounding factors, WHtR remained an independent predictor of CVD (OR = 5.76, P = 0.028), which was consistent with previous studies. Previous studies have pointed out that WHtR can more accurately reflect visceral fat accumulation and cardiometabolic risk than BMI, and is particularly clinically sensitive in patients with respiratory diseases [4]. Our restricted cubic spline analysis further revealed that although the relationship between WHtR and CVD was linear, there was a risk turning point at 0.601, which is consistent with the control logic recommended in the “Global Consensus Definition of Metabolic Syndrome” released by the International Diabetes Federation (IDF) that “WHtR < 0.5 is considered low risk “. It is suggested that controlling the WHtR of COPD patients below 0.601 may have important preventive value. Secondly, this study further revealed the important role of the nutritional prognosis index (PNI) in the association between waist-to-height ratio (WHtR) and cardiovascular events (CVD) in patients with chronic obstructive pulmonary disease (COPD) through rigorous mediation analysis, indicating that PNI plays a partial mediating role in this association. Specific data show that the indirect effect accounts for 7.98% of the total effect, and the statistical test results are significant (P < 0.001). This finding provides strong empirical support for the hypothesis that “WHtR may indirectly increase the risk of CVD by influencing the body’s nutritional-immune status.” As a composite indicator combining serum albumen levels and peripheral blood lymphocyte counts, PNI not only reflects the body’s nutritional reserve status (albumin is the core indicator for assessing protein nutritional status) but also reflects immune function status (lymphocytes are key cellular components of the immune system). Therefore, it is widely used to assess the nutritional-immune status of patients with chronic diseases [10]. When PNI decreases, it often indicates that the body may be in a state of chronic inflammation, malnutrition, or immune suppression, which are particularly common in the long-term progression of chronic diseases [15–18]. From a physiological and pathological perspective, a high WHtR ratio, as a key indicator of abdominal obesity, is closely associated with metabolic disorders in the body. Abdominal obesity can induce insulin resistance through multiple pathways, and insulin resistance further leads to abnormalities in glucose and lipid metabolism. Additionally, it stimulates adipose tissue (especially visceral fat) to release large amounts of pro-inflammatory factors (such as tumor necrosis factor-α and interleukin-6), resulting in a persistent state of low-grade chronic inflammation [19–23]. This metabolic disorder interacts with chronic inflammation, directly affecting the body’s nutrient-immune balance: on one hand, chronic inflammation increases energy expenditure and protein breakdown, leading to increased nutrient requirements. If intake is insufficient or absorption is impaired (e.g., COPD patients often experience inadequate nutrient intake due to dyspnea and weakened digestive function), it can result in reduced albumin synthesis and exacerbate malnutrition [24]; on the other hand, prolonged inflammation suppresses lymphocyte proliferation and function, reducing the body’s immune defense capacity [25]. The combined effects of these two factors lead to a decrease in PNI. A reduced PNI implies weakened protection of vascular endothelium (albumin plays a role in maintaining vascular colloid osmotic pressure and antioxidant functions) [26] and impaired immune surveillance (inability to clear abnormal substances from blood vessels) [10]. Additionally, the persistent chronic inflammatory state ultimately contributes to cardiovascular pathophysiological processes such as atherosclerosis and thrombosis [27, 28], thereby increasing the risk of CVD. For patients with abdominal obesity, under the premise of avoiding excessive weight loss (to prevent a decline in respiratory muscle strength), an individualized diet plan that is low in calories and high in fiber should be formulated (such as reducing daily calories by 10%−15% compared to the conventional COPD diet, giving priority to high-quality protein and complex carbohydrates, and supplementing vitamin D when necessary). At the same time, low-intensity resistance exercises (such as sitting leg raises and resistance band training) should be incorporated into pulmonary rehabilitation training to control abdominal circumference and lower WHtR, while avoiding increasing the burden on breathing. This study has certain advantages: (1) Using a nationally representative sample of NHANES, the results have high generalizability; (2) Multivariate Logistic regression and restricted cubic spline analysis were used to ensure the robustness of the results; (3) PNI was introduced as a mediating variable for the first time, providing a new mechanistic explanation from the nutritional perspective. However, this study also has several limitations: (1) NHANES is a cross-sectional study, so a causal relationship cannot be established; (2) The diagnosis of COPD and CVD relies on self-reporting and some indicators, which may have information bias; (3) Although PNI is a commonly used nutritional indicator, it is also affected by various chronic inflammations and metabolic states, and needs to be further verified in combination with inflammatory markers; (4) Some covariates are missing, and although multiple imputation has been used, it may still affect the stability of the results. Conclusion This study indicates that WHtR is an important predictor of COPD patients complicated with CVD, and its role is partially achieved through the nutritional status represented by PNI. Controlling WHtR levels and improving nutritional status may help reduce the cardiovascular risk of COPD patients. This finding provides a theoretical basis for the early identification and multi-dimensional intervention of COPD complicated with CVD, suggesting that clinical practice should pay attention to the synergistic value of WHtR and nutritional indicators in disease management. Future prospective studies should be conducted to further verify this mediating mechanism and explore the potential role of nutritional intervention in reducing CVD risk in COPD patients. Supplementary Information Supplementary Material 1. Acknowledgements We would also like to thank everyone who has helped with our research. Authors’ contributions YLC, ELY and WL are the guarantors of the manuscript and take responsibility for the content of this manuscript. LJZ, JRX and RKC contributed to the design of the study. XMC, HL, GHC, YHZ and HYL, were involved in the data analysis. MYY, YXG, QQP, CHT and HL contributed to the acquisition of primary data. JFC wrote the initial draft of the manuscript. SYZ contributed significantly to the revision of the manuscript. All authors read and approved the final manuscript. Funding Guangdong Provincial Natural Science Foundation (2022A1515012375), Guangdong Provincial Medical Research Foundation (A2024728, A2024723), Zhanjiang Municipal Science and Technology Development Special Fund Competitive Allocation Project (2021A05088), Special Project for Clinical and Basic Sci & Tech Innovation of Guangdong Medical University (GDMULCJC2024075, GDMULCJC2024076, GDMULCJC2024083), National College Student Innovation and Entrepreneurship Training Program (202310571006). Data availability The NHANES dataset is publicly available at: https://www.cdc.gov/nchs/nhanes/index.htm. Declarations Ethics approval and consent to participate Not applicable. 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. Yuli Cai, Enlin Ye, Wang Liu and Shuyue Zhou contributed equally to this work. Contributor Information Lijuan Zeng, Email: ZLJ13729063609@163.com. Jinrong Xu, Email: zjeyxjr@163.com. Riken Chen, Email: chenriken@126.com. References Associated Data Supplementary Materials Supplementary Material 1. Data Availability Statement The NHANES dataset is publicly available at: https://www.cdc.gov/nchs/nhanes/index.htm.