The dietary inflammatory index (DII^®) and human papillomavirus infection: a cross-sectional study of US women

Abstract

Background

Diet may influence susceptibility of human papillomavirus (HPV) infection by affecting inflammation and immunity. However, the association between HPV infection and the inflammatory potential of diet has not been investigated. The research aimed to examine the correlation between HPV status and the dietary inflammatory index (DII^®).

Methods

We utilized data from the National Health and Nutrition Examination Survey (NHANES) 2003–2016 to investigate the correlation between DII and HPV status among 9,256 women aged 18–59 years. DII scores were calculated based on 24-hour dietary recall interviews. The association between HPV status and DII was analyzed using weighted logistic regression and restricted cubic spline (RCS).

Results

Women with HPV infection exhibited higher DII scores than those without HPV infection. An increased likelihood of HPV infection was found to be significantly associated with higher DII scores (OR = 1.05, 95% CI: 1.01–1.09, P = 0.021), after full multivariate adjustment. Compared with the lowest tertile of DII scores, the ORs (95% CIs) for HPV infection were 1.20 (1.01, 1.42) and 1.27 (1.07, 1.51) for the second and third tertiles, respectively (P for trend = 0.006). RCS analysis showed a U-shaped relationship between DII and HPV infection, with a breakpoint identified at 0.13.

Conclusions

Our findings suggest that a pro-inflammatory diet is associated with an increased likelihood of HPV infection among women in the United States. Dietary interventions to reduce inflammation may help prevent HPV infection and related diseases.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-024-20490-0.

Introduction

Human papillomavirus (HPV) infection is the most common sexually transmitted disease (STD) worldwide, contributing to 4.5% of newly diagnosed cancer cases [1]. It is also one of the etiological factors for benign lesions such as genital warts [2]. Almost 100% of cervical cancers and a considerable portion of oropharyngeal, anal, penile, vulvar, and vaginal cancers are caused by HPV infection [3].

Previous studies indicate that chronic inflammation and oxidative stress are associated with the risk of HPV infection, the formation of precancerous lesions, and the progression to cervical cancer [4–6]. Diet is one of the most crucial factors in regulating the body’s inflammatory status. Previous studies have shown that an unhealthy dietary pattern of consuming large amounts of red and processed meats, dips, chips, and snacks, as well as small amounts of olive oil, is associated with a higher risk of HPV infection, and can lead to increased inflammation, decreased control of infections, and an increased risk of auto-inflammatory diseases [7, 8]. However, healthy dietary patterns through the consumption of specific nutritious foods (including vegetables and fruits) and intake of nutritional antioxidants (such as vitamins A, B2, C, D, E, and folic acid) can reduce the likelihood of HPV infection, to prevent cervical cancer [8–11].

The dietary inflammatory index (DII^®) serves as an important epidemiological instrument for assessing the inflammatory potential of diets and predicting chronic diseases based on dietary patterns [12, 13]. It distinguishes between the anti-inflammatory and pro-inflammatory properties of diets. A higher DII score indicates that the dietary pattern has more pro-inflammatory components. Recent studies have shown the association between a high intake of proinflammatory diets and an increased risk of cervical cancer among women with cervical intraepithelial neoplasia (CIN) grades 2 or 3 [14, 15].

Despite the established links between diet, inflammation, and the progression of HPV-related conditions, the specific role of dietary inflammation—measured by the DII—in influencing HPV infection remains unclear. This study aims to address this gap by examining the association between dietary inflammation, as quantified by the DII, and HPV infection using data from the National Health and Nutrition Examination Survey (NHANES). By exploring this association, we hope to provide valuable insights that could contribute to developing more effective dietary-based preventive strategies for HPV-related diseases.

Methods

Study population

NHANES is a comprehensive cross-sectional survey using a stratified, multistage probabilistic methodology to generate population estimates concerning the nutritional and health status of adults and children in the United States. This is achieved through personal interviews, health screenings, and laboratory specimen analyses. Details of the NHANES are available through the NHANES website [16] (https://www.cdc.gov/nchs/nhanes/index.htm). In this study, we initially included 35,936 female participants from seven cycles of NHANES spanning from 2003 to 2016. Among them, 23,302 participants were excluded due to missing information on HPV vaginal swab testing. Additionally, 608 participants were excluded because they lacked dietary information, and 2770 participants were excluded due to no information on covariates of interest. Finally, 9,256 participants were included in the analysis (Supplementary Figure S1).

Human papillomavirus infection

DNA samples from self-collected vaginal swabs of women aged 18–59 in seven NHANES surveys (2003–2016) were typed for 37 genotypes. Genotyping of all samples was conducted using the HPV L1 consensus polymerase chain reaction (PCR). More detailed HPV testing methods are publicly available on the NHANES website. The study’s outcome variable was the overall HPV infection status, categorized into three groups: no HPV, low-risk (LR) HPV, and high-risk (HR) HPV infection. Negative results for all 37 HPV types were classified as the no HPV group. Positive results for any of the 12 specified HPV genotypes (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, and 59) were classified as the HR-HPV group [17–19], while positive results for the remaining HPV genotypes were considered as the LR-HPV group.

Calculation of the DII

The dietary data used in this study were collected from NHANES using 24-hour dietary recall interviews. This method is widely employed in large-scale epidemiological studies, and to ensure accuracy, bilingual dietary interviewers conducted in-person interviews using the USDA’s Automated Multiple-Pass Method (AMPM) [20]. The dietary data collected from NHANES serves as forms the foundation for calculating the DII, a validated tool for assessing the inflammatory potential of diets, as demonstrated in previous research [12, 13]. The method assesses the inflammatory effects of dietary intake based on 45 nutrients. However, in this study, we calculated DII scores based on 28 of the 45 food and nutrient components due to data availability from the NHANES dietary recall interviews. These components included energy, alcohol, protein, fiber, cholesterol, carbohydrates, total fat, saturated fat, polyunsaturated fatty acid (PUFA), monounsaturated fatty acids (MUFA), n-6 fatty acids, n-3 fatty acids, vitamins (A, B6, B12, C, D, and E), thiamin, riboflavin, niacin, iron, magnesium, zinc, folic acid, selenium, caffeine and beta-carotene.

The DII calculation was performed following the standard methodology proposed by Hébert et al. [13]. Briefly, each food and dietary component was assigned an inflammation score based on its association with specific inflammatory markers (IL-1 β, IL-4, IL-6, IL-10, TNF-α, and CRP), resulting in a pooled score that reflects the pro-inflammatory or anti-inflammatory potential of an individual’s daily diet. The DII calculation was carried out as follows [12]: First, the Z-score for each food parameter was calculated using the formula: (daily average intake - global daily average intake)/standard deviation. Next, the Z-score was converted to a percentile score. Each percentile score was multiplied by 2, and 1 was subtracted from the result. The DII score for each specific food parameter was then calculated by multiplying this value by its corresponding inflammatory effect score. Finally, the total DII score was obtained by summing the DII scores for all food parameters.

Covariables

Covariates were included in this study, including age, race or ethnicity, marital status, poverty income ratio, education level, body mass index (BMI), physical activity, smoking status, drinking status, hypertension status (HTN), cardiovascular status (CVD), type 2 diabetes status (T2D) and number of lifetime sexual partners. Physical activity was categorized into three groups: the inactive group (no leisure-time physical activity, the insufficiently active group (1 to 5 moderate activities or 1 to 3 vigorous activities per week during leisure time), or the active group (more moderate or vigorous activity than mentioned above). There were three categories for smoking status: non-smoker, former smoker, or current smoker [21]. Non-smokers were people who had either never smoked or smoked less than 100 cigarettes in their lifetime. Former smokers were those who had smoked a minimum of 100 cigarettes in their lifetime but had since quit. Current smokers were those who had smoked a minimum of 100 cigarettes in their lifetime and were actively smoking. Drinking status was categorized as non-drinker, light/moderate drinker (≤ 2 drinks per day for males and ≤ 1 drink per day for females), and heavier drinker (> 2 drinks per day for males and > 1 drink per day for females) [22]. Hypertension was characterized by a systolic blood pressure of ≥ 130 mm Hg and/or a diastolic blood pressure of ≥ 80 mm Hg, taking anti-hypertensive medication, or being diagnosed with hypertension [23]. Diabetes was characterized by a glycated hemoglobin A1c level of 6.5% or higher, fasting glucose exceeding 7 mmol/L, a 2-hour post 75 g glucose load (OGTT) result of 11.1 mmol/L or more, self-reported diabetes diagnosis, or self-reported usage of insulin or other diabetic treatment [22, 24]. The cardiovascular condition was assessed using self-reported information on stroke, heart attack, congestive heart failure, angina or coronary heart disease.

Statistical analysis

Due to the complex survey design of the NHANES, all analyses utilized sample weights. The median and interquartile range (IQR) were used to characterize continuous variables, whereas the unweighted frequency and weighted percentage were used to represent categorical variables. For the purposes of analysis, DII scores were categorized into tertiles. The highest tertile (T3) represents the most pro-inflammatory diet, with higher DII scores indicating a greater inflammatory potential of the diet. Conversely, the lowest tertile (T1) represents a more anti-inflammatory diet, characterized by lower DII scores. To determine differences among different groups, the weighted chi-square or Kruskal-Wallis H test was used. The association between DII scores and HPV infection was analyzed using weighted logistic regression models. Using the lowest tertile as the reference, odds ratios (ORs) and 95% confidence intervals (CIs) were computed for each tertile of DII scores. A restricted cubic spline (RCS) was applied to visualize the probable non-linear correlation between HPV infection and DII. Model 3 was stratified by age group, race or ethnicity, marital status, education level, poverty income ratio, BMI group, physical activity, smoking status, drinking status, T2D status, HTN status, CVD status and number of sexual partners to evaluate the association between DII scores and HPV infection. A statistical interaction test was then used to assess the robustness of the observed modifications. R software version 4.4.0 was used for all analyses, and a two-sided p-value < 0.05 was regarded as statistically significant.

Results

Study participants and baseline characteristics

The baseline characteristics of the individuals based on their HPV infection status are shown in Table 1. There were 4100 participants infected with HPV, of which 1938 and 2162 were infected with HR-HPV and LR-HPV, respectively. Compared to the non-HPV infection group, participants in the HPV-positive group were younger, predominantly non-Hispanic black, less likely to be married, with lower levels of education and poverty income ratio, more likely to be current smokers and heavier drinkers, had 5 or more sexual partners during their lifetime and had a higher proportion of CVD status. Notably, participants with HPV infection tended to have a higher DII score than those without HPV (Fig. 1). Supplementary Table S1 displays the subject characteristics based on their DII score tertiles. Compared to participants in the lowest inflammation tertile (T1) of DII, those in the highest inflammation tertile (T3) were more likely to be younger, non-Hispanic black, never married, have lower physical activity levels, be smokers or drinkers, and have higher rates of obesity, T2D status, CVD status, and HPV positivity.

Table 1.

Baseline characteristics of participants with different HPV infection status

Characteristic	Overall N = 92561	HPV negative N = 5156 (59%)1	HPV positive N = 4100 (41%)1	p-Value2
Age	41 (30, 50)	42 (31, 51)	38 (28, 48)	< 0.001
Age group				< 0.001
20–34 years	4,752 (48%)	2,498 (44%)	2,254 (52%)
35–49 years	2,410 (26%)	1,404 (27%)	1,006 (25%)
50–59 years	2,094 (26%)	1,254 (29%)	840 (22%)
Race/ethnicity				< 0.001
Hispanic	2,414 (14%)	1,408 (14%)	1,006 (14%)
Non-Hispanic White	3,993 (67%)	2,376 (71%)	1,617 (62%)
Non-Hispanic Black	2,036 (12%)	810 (8.0%)	1,226 (19%)
Other races/ethnicities	813 (6.6%)	562 (7.5%)	251 (5.3%)
Marital status				< 0.001
Never married	3,048 (29%)	1,339 (23%)	1,709 (39%)
Married	4,549 (54%)	3,131 (65%)	1,418 (37%)
Widowed/Divorced/Separated	1,659 (17%)	686 (12%)	973 (24%)
Education level				< 0.001
< High school graduate	1,768 (13%)	929 (12%)	839 (15%)
High school graduate/GED	1,920 (20%)	962 (19%)	958 (23%)
College Graduate or above	5,568 (67%)	3,265 (70%)	2,303 (62%)
Poverty income ratio				< 0.001
< 1(poverty)	2,150 (16%)	990 (13%)	1,160 (22%)
≥ 1(non-poverty)	7,106 (84%)	4,166 (87%)	2,940 (78%)
BMI group				0.577
Underweight (< 18.5)	196 (2.2%)	105 (2.0%)	91 (2.5%)
Normal (18.5 to < 25)	2,812 (34%)	1,596 (35%)	1,216 (34%)
Overweight (25 to < 30)	2,468 (27%)	1,379 (27%)	1,089 (26%)
Obese (30 or greater)	3,780 (37%)	2,076 (36%)	1,704 (37%)
Physical activity				0.069
Inactive	2,026 (18%)	1,084 (17%)	942 (20%)
Insufficient	3,036 (34%)	1,688 (35%)	1,348 (33%)
Recommended	4,194 (48%)	2,384 (48%)	1,810 (47%)
Smoking status				< 0.001
Never smoker	5,826 (60%)	3,529 (65%)	2,297 (53%)
Former smoker	1,429 (18%)	820 (19%)	609 (16%)
Current smoker	2,001 (23%)	807 (17%)	1,194 (31%)
Drinking status				< 0.001
Never drinker	1,483 (12%)	980 (15%)	503 (8.3%)
Light/moderate drinker	7,015 (79%)	3,878 (79%)	3,137 (80%)
Heavier drinker	758 (8.4%)	298 (6.4%)	460 (11%)
T2D status	935 (8.3%)	537 (8.9%)	398 (7.5%)	0.085
HTN status	3,186 (34%)	1,748 (35%)	1,438 (34%)	0.379
CVD status	367 (3.7%)	165 (2.8%)	202 (4.9%)	< 0.001
No. of sexual partners				< 0.001
NO	166 (1.3%)	111 (1.7%)	55 (0.7%)
≤ 5	5,356 (55%)	3,467 (65%)	1,889 (42%)
> 5	3,734 (43%)	1,578 (34%)	2,156 (57%)
DII score	1.84 (0.38, 2.93)	1.63 (0.15, 2.81)	2.13 (0.68, 3.11)	< 0.001
DII score categories3				< 0.001
T1 (-1.33, 0.38)	2,832 (33%)	1,765 (37%)	1,067 (28%)
T2 (1.39, 2.23)	3,169 (33%)	1,771 (33%)	1,398 (34%)
T3 (2.93, 3.70)	3,255 (33%)	1,620 (30%)	1,635 (38%)

¹Median (IQR); n (unweighted) (%)

²Wilcoxon rank-sum test for complex survey samples; chi-squared test with Rao & Scott’s second-order correction

³DII tertile ranges. T1: low DII scores, T2: middle DII scores, T3: higher DII scores

Abbreviations: BMI, body mass index; T2D: type 2 diabetes; HTN: hypertension; CVD: cardiovascular disease; No.: number; DII: Dietary Inflammation Index

Fig. 1.

Distribution of DII score. (A) The distribution of DII in the population with non-HPV infection and HPV infection. (B) The distribution of DII in the population with different HPV genotypes

Association between DII and HPV infection

To investigate the relationship between DII and HPV infection, we performed multivariable logistic regression analysis (Table 2). In the unadjusted model, we found that the DII score was positively associated with HPV infection, with each unit increase in the DII score potentially leading to a 13% higher likelihood of HPV infection (OR = 1.13; 95% CI: 1.08, 1.17; P < 0.0001). After adjustment for age, race or ethnicity, marital status, education level and poverty income ratio in model 1, the relationship between DII and HPV infection remained significant and positive (OR = 1.06; 95% CI: 1.02–1.10; P = 0.006). In Model 2, which further adjusted for BMI group, physical activity, smoking status, drinking status, HTN status, CVD status, T2D status, and number of sexual partners, consistent results were observed (OR = 1.05; 95% CI: 1.01–1.09; P = 0.021). Similarly, participants in the intermediate (T2) and highest inflammation groups (T3) showed greater odds of HPV infection compared to those in the lowest inflammation group (T1 of DII scores), with ORs of 1.33 (95%CI: 1.13, 1.55) and 1.71 (95%CI: 1.47, 1.98), respectively, in the unadjusted model. After adjusting for all covariates in Model 2, the OR with 95% CI for HPV infection across the tertiles was 1.20 (1.01, 1.42) and 1.27 (1.07, 1.51), respectively. The study further examined the correlation between different HPV genotypes and DII scores. The results indicated that both low-risk HPV status and high-risk HPV status were significantly and positively associated with DII scores in the unadjusted model (Supplementary Figure S2). After adjusting for all covariates in Model 2, among those with low-risk HPV infection, the T3 group showed a significant positive association with DII scores compared to the T1 group (OR = 1.36; 95% CI: 1.12, 1.66; P = 0.003).

Table 2.

Odds ratios (ORs) and 95% confidence interval (CI) of the dietary inflammatory index for HPV infection

DII	Model 0		Model 1		Model 2
	OR (95% CI)	p-Value	OR (95% CI)	p-Value	OR (95% CI)	p-Value
Continuous	1.13(1.08, 1.17)	< 0.001	1.06(1.02, 1.10)	0.006	1.05(1.01, 1.09)	0.021
Categories
T1
T2	1.33(1.13, 1.55)	< 0.001	1.18(1.00, 1.40)	0.055	1.20(1.01, 1.42)	0.039
T3	1.71(1.47, 1.98)	< 0.001	1.32(1.12, 1.55)	0.001	1.27(1.07, 1.51)	0.006
Trend test		< 0.001		< 0.001		0.006

Model 0: no covariates were adjusted

Model 1: age, race or ethnicity, marital status, education level and poverty income ratio were adjusted

Model 2: adjusted for age, race or ethnicity, marital status, education level, poverty income ratio, BMI group, physical activity, smoking status, drinking status, T2D status, HTN status, CVD status and No. of sexual partners

Figure 2 illustrates a non-linear relationship between DII scores and HPV infection (P for nonlinearity = 0.0118). The likelihood of HPV infection decreased gradually with an increasing DII score until it reached around 0.13, after which it started to increase rapidly. In addition, restricted cubic spline revealed a U-shaped association with low-risk HPV status (P for nonlinearity = 0.0003, Supplementary Figure S3A), whereas a positive linear association between DII and high-risk HPV status (P for nonlinearity = 0.9177, Supplementary Figure S3B).

Fig. 2.

Restricted spline curve for the association between DII and HPV infection. The red represents estimates of odd ratios and dashed lines represent 95% confidence intervals

Subgroup analysis

As shown in Fig. 3, subgroup analysis revealed inconsistent associations between DII scores and HPV infection, despite a predominantly positive relationship in most categories. The results of the interaction test indicated that marital status and number of sexual partners modified the association between DII scores and HPV infection (P for interaction < 0.05).

Fig. 3.

Forest plots of stratified analysis of the associations between DII and HPV infection

Discussion

This study examined the association between HPV infection and the DII score utilizing data from NHANES 2003–2016. Our findings suggest a significant positive correlation between a pro-inflammatory diet and HPV infection. This association remained statistically significant even after full multivariable adjustment, with a significantly higher OR for HPV infection in those in the highest DII tertile (T3) compared with those in the lowest tertile (T1) at 1.27 (95% CI: 1.07, 1.51, P = 0.006). Restricted cubic spline demonstrated a significant U-shaped correlation between DII and HPV infection. Initially, the prevalence of HPV infection decreased as the DII score increased up to approximately 0.13. However, beyond this threshold, the prevalence of HPV infection rose sharply with higher DII scores, suggesting that a more pro-inflammatory diet may compromise immune function and increase susceptibility to HPV infection. Notably, there was a linear positive correlation between DII and high-risk HPV infection, emphasizing the impact of pro-inflammatory dietary patterns on increasing the risk of infection.

Over the period from 2003 to 2016 in the United States, there have been notable shifts in dietary patterns and food variety. While there has been a general trend towards healthier eating, with increased focus on whole foods and reduced consumption of processed foods, the intake of certain pro-inflammatory nutrients, particularly sugars and low-quality carbohydrates, has remained high [25–27]. Our study demonstrates that a higher DII score, indicative of a pro-inflammatory diet, is significantly associated with an increased risk of HPV infection. This finding aligns with previous research that emphasizes the role of diet-induced inflammation in modulating immune responses and disease progression [28].

Diets characterized by high levels of calories, salt, processed foods, red meats, and refined sugars —components that contribute to higher DII scores—are known to exacerbate systemic inflammation, thereby potentially weakening the immune system’s ability to clear HPV infections [29–31]. Conversely, a healthy diet that includes a regular intake of antioxidant nutrients, such as vitamins A, C, and E, has been shown to lower the risk of HPV infection and even cervical cancer [10, 32, 33]. For example, regular consumption of fruits and dark green vegetables has been shown to significantly reduce the risk of HR-HPV infection, highlighting the importance of diet in modulating inflammatory responses [9, 18]. Vitamin A is essential for the regulation of local immunity and the reduction of inflammatory responses [34], and its insufficient intake has been linked to HPV infection, HR-HPV infection, and cervical cancer [9, 35, 36]. Similarly, vitamin E has been shown to protect the integrity of immune cell membranes, suppress the production of prostaglandin E2 (PGE2), and inhibit the enzymatic activity of cyclooxygenase-2 (COX-2), which may enhance the body’s ability to clear HPV infection [37–39]. Folic acid also plays a crucial role in the production of erythrocytes, synthesis and repair of DNA, methylation processes, and cell proliferation [40]. Insufficient folic acid levels can elevate the risk of cervical cancer and HR-HPV infection [11, 41–43].

Furthermore, diet plays a crucial role in regulating the diversity, activity, and associated metabolites of human gut flora [44, 45]. Several studies have found that different nutrients in the diet and gut microbes can modulate the balance of pro-inflammatory and anti-inflammatory responses [46–48]. Existing studies suggest that bacterial strains can be transferred from the gut to the vagina [49–51]. The vaginal microbiota plays a crucial role in preventing inflammation and reducing susceptibility to microbial infections, including HPV infection [52–54]. For example, administering probiotic bacteria and yeast orally can hinder the progression and elimination of HPV infection by enabling probiotic microorganisms to travel from the gut to the vagina [55–57]. These mechanisms may underlie the connection between DII and HPV infection and warrant further exploration.

Notably, our study is the inaugural examination of the U-shaped correlation between DII and HPV infection in a representative sample of adult women in the United States. The likelihood of HPV infection significantly increased when the diet showed a pro-inflammatory pattern, highlighting the need for future longitudinal studies to confirm these findings and establish causative relationships. This finding is consistent with previous studies on the link between the inflammatory potential of diets and cervical carcinogenesis. For example, in a case-control study conducted by Sreeja et al. among Korean women, higher DII scores were significantly associated with increased odds of developing CIN2/3 and cervical cancer [14]. The study reported median DII scores of 1.1 (-0.7, 2.5) for women with CIN2/3 and 1.5 (-0.6, 3.0) for those with cervical cancer. Similarly, our study found that the prevalence of HPV infection increased in participants with DII scores above 0.13, indicating that even moderate shifts toward a pro-inflammatory diet can elevate infection risk. Furthermore, Maugeri et al. reported women with higher DII scores had elevated odds of developing high-grade CIN lesions in an Italian cohort [15]. These findings collectively suggest that a pro-inflammatory diets, as measured by DII, are linked to a higher risk of HPV infection and potentially cervical cancer.

However, this study has several potential limitations. First, being a cross-sectional study, it only provides a snapshot of data at a single point in time, which limits our ability to infer causality and may include unmeasured confounders. Second, the DII score was calculated based on a 24-hour dietary recall method. This method relies on participants’ memory and honesty, which can introduce recall bias and potentially inaccurate dietary assessments. The absence of long-term dietary data also limits our ability to capture habitual dietary patterns, which may affect the observed relationship between DII and HPV infection. Additionally, NHANES was designed to be primarily representative of the United States population, and certain population subgroups with different dietary habits or health conditions may be underrepresented, both of which may limit the generalization of our findings to more diverse or global populations.

Conclusion

In conclusion, our results suggest that among adult women in the United States, a pro-inflammatory diet is significantly and positively associated with HPV infection. The findings might offer useful insights for large-scale prospective studies to emphasize the significance of dietary health, especially for individuals following pro-inflammatory diets. Individuals with such dietary patterns are encouraged to increase their intake of anti-inflammatory foods to lower the likelihood of HPV infection.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Conception and design: Qian Li and Yu Chen; Collection and data curation: Yu Chen and Mei Yang; Data analysis and interpretation: Qian Li and Peibo Li; Manuscript writing: All authors; Final approval of manuscript: All authors.

Funding

This research was funded by 2023 key Disciplines on Public Health Construction in Chongqing, grant number 2023 − 00304.

Data availability

The datasets for this study can be found in the NHANES database (https://www.cdc.gov/nchs/nhanes/index.htm).

Declarations

Ethics approval and consent to participate

The studies involving humans were approved by the Centers for Disease Control and Prevention (CDC) National Increase for Health Statistics Research (NCHS) Ethics Review Board. Ethical review and approval were waived for this study because this study was a secondary analysis of public-use data available without personal identifiers. The patients/participants provided their written informed consent to participate in this study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.