Abstract
Objective
There has been some evidence that dietary fiber may be associated with diabetic nephropathy (DN), but the relationship is still unclear. The purpose of this study was to examine the association between dietary fiber intake and DN.
Methods
This cross-sectional study used National Health and Nutrition Examination Survey (NHANES) data collected between 2007 and 2020. Weighted multivariate logistic regression was used to examine the relation between dietary fiber intake and DN. In addition, fitted smoothed curves were used to explore potential non-linear relationships. If non-linearity was observed, inflection points were further calculated by a recursive algorithm.
Results
The study finally included 5964 subjects ≥20 years of age. The mean age was 60.8 ± 13.4 years with males (52.4 %), and non-Hispanic Whites (62.4 %), and the weighted prevalence of DN was 36.7 %. Dietary fiber was negatively associated with the risk of DN after controlling for all confounding variables (OR = 0.89, 95%CI: 0.80, 0.99). Smoothed curve fit plots of the dose relationship showed that dietary fiber intake was linearly related to DN, whereas males (inflection point of 8.0 g/d) and non-Hispanic Blacks (inflection point of 14.9 g/d) followed a non-linear inverted U-shaped curve relationship. In United States adults aged 20 and older, dietary fiber intake may be associated with a reduced risk of DN.
Conclusion
Appropriate increases in dietary fiber intake may offer potential benefits for DN. In conclusion, it appears that increasing dietary fiber intake may be one of the most effective strategies for the prevention and management of DN.
Keywords: Dietary fiber, Diabetic nephropathy, NHANES, Cross-sectional study
1. Introduction
In 2021, nearly 540 million people had diabetes mellitus (DM) globally, and by 2045, this number is expected to reach 780 million [1]. One of DM's major microvascular complications is diabetic nephropathy (DN). The main manifestations are decreased glomerular filtration rate (GFR), increased urinary albumin, and hypertension, which ultimately leads to end-stage renal disease (ESRD). This is the main cause of DM morbidity [2,3]. It is worth noting that approximately 30–40 % of DM in the United States (US) progresses to DN [4]. Many factors have been reported to contribute to the progression of DN, such as hypertension [5], obesity [6], and hyperglycemia [2]. Dietary habits may be one of the most favorable factors for DN [7,8]. However, as it stands, there are still areas of dietary management that are not fully addressed in clinical practice guidelines, and thus dietary interventions may be a future trend in the prevention and treatment of DN [7,9]. Therefore, it is necessary to learn more about dietary factors that may help prevent and manage DN.
In recent years, dietary fiber consumption has been increasingly studied for its health benefits. Dietary fiber is a plant polysaccharide that cannot be hydrolyzed and absorbed by human digestive enzymes, also known as non-absorbed plant carbohydrates. Nevertheless, intestinal flora can ferment it into short-chain fatty acids [10]. It has been reported that fiber intake can reduce the risk of cardiovascular disease, type 2 diabetes mellitus (T2DM), and cancer when consumed in moderation [11,12]. Increasing fiber intake in patients with chronic kidney disease reduces all-cause mortality, according to a recent prospective study [13]. These benefits may result from the important mechanisms of dietary fiber on intestinal motility, insulin sensitivity, glycemic control, chronic inflammation, and gut microbiology [14,15]. However, it is not clear if dietary fiber has benefits on DN.
As far as we know, large-sample epidemiological studies on dietary fiber and DN in the US have been very limited in the past decades. Thus, further research on the beneficial effects of dietary fiber on DN is necessary. We evaluated fiber intake and DN risk for the first time in this study. The study used large-sample real-world data from the National Health and Nutrition Examination Survey (NHANES) (2007–2020). Our research will provide ideas for subsequent basic clinical research and give new insights and evidence for nutritional guidelines and health policy.
2. Materials and methods
2.1. Study population
In this study, data were obtained from the NHANES database, an annual survey conducted by the Centers for Disease Control and Prevention in the US, which mainly consisted of questionnaires, nutritional information, physical examinations, and laboratory findings. An unbiased, stratified, multistage probability sampling was used to represent the non-institutionalized civilian population of the US over a period of two years [16]. Participants in the NHANES study provided written informed consent to the Ethics Review Board of the National Center for Health Statistics (NCHS). Data were selected from NHANES (2007–2020), which recruited 66,148 participants. Inclusion exclusion (Fig. 1): excluded were those less than 20 years of age, those not having DM, missing data on dietary fiber intake, missing data on urinary albumin creatinine ratio, missing data on glomerular filtration rate, and pregnant. The final 5964 DM participants took part in the study.
Fig. 1.
Diagram showing how the study population is selected.
2.2. Exposure variable
Food beverages and dietary supplements consumed during two 24-h dietary periods (midnight to midnight) were used to measure the intake of dietary fiber. The energy, nutrition, and other food components in food and beverages were also estimated [17]. In the Mobile Examination Center (MEC), the first dietary recall interview was conducted in person by trained professionals, and the second interview was conducted over the phone three to ten days later by trained professionals. If the participant completed two 24-h recalls, the average was used, and if only one was completed, the first 24-h recall data was used. The US Department of Agriculture was used to calculate the nutritional values of all foods, beverages, and supplements.
2.3. Outcome variable
NHANES self-report questionnaire and the American Diabetes Association (ADA) diagnostic criteria were used to diagnose participants with diabetes [18]. An individual must have glycated hemoglobin of at least 6.5 %, fasting blood glucose (FPG) of 126 mg/dL, 2-h plasma glucose of at least 200 mg/dL, and be taking an insulin or hypoglycemic medication currently. In the present study, all DM participants were included in the study. The outcome variable was whether or not they had DN. The participants were defined as having DN when the urinary albumin to creatinine ratio (UACR) was ≥ (30 mg/g) or when the estimated glomerular filtration rate (eGFR) was <60 mL/min/1.73 m2. Otherwise, they were defined as non-DN [19,20]. Documents describing data collection and laboratory testing methods are publicly available and free of charge on the NHANES website.
2.4. Covariates
Previous studies helped us determine the covariates [[21], [22], [23]]. Population characteristics: age, gender, race, education, and poverty income ratio. Lifestyle factors: smoking behavior, drinking behavior, physical activity, dietary fiber intake, energy intake, carbohydrate intake, and fiber supplement. Biochemical examination: alanine aminotransferase (ALT), aspartate aminotransferase (AST), blood urea nitrogen (BUN), serum creatinine (SCR), total bilirubin (TBIL), total cholesterol (TC), triglyceride (TG), serum uric acid (SUA), HBA1C, and high-density lipoprotein cholesterol (HDL-C). Medical examination data: body mass index (BMI), and waist circumference (WC). Disease diagnosis: high blood pressure (HBP), hyperlipidemia, and cardiovascular disease (CVD). Free access to the NHANES data was provided on the website. HBP was defined as having HBP as diagnosed by a doctor or currently taking prescription drugs for HBP, and systolic blood pressure (SBP) ≥ 130 mmHg or diastolic blood pressure (DBP) ≥ 80 mmHg [24]. A BMI group is designated as underweight or normal weight (under 25.0 kg/m2), overweight (between 25.0 and 29.9 kg/m2), and obese (over 29.9 kg/m2). Hyperlipidemia was defined as having hyperlipidemia diagnosed by a doctor or TC ≥ 240 mg/dL, TG ≥ 200 mg/dL, HDL-C < 40 mg/dL, low-density lipoprotein-cholesterol (LDL-C) ≥ 160 mg/dL [22]. Among the CVDs identified were congestive heart failure (CHF), coronary heart disease (CHD), angina, heart attack, and stroke.
2.5. Statistical analyses
Given the representative nature of the study results, we performed weighting and variance estimation as recommended by the Centers for Disease Control and Prevention guidelines. R (version 4.2.0) and EmpowerStats (www.empowerstats.com; X&Y Solutions, Inc., Boston, MA, USA) were used for statistical analysis. Considering that participants' dietary fiber intake was skewed, we transformed the data by log2 to conform to a normal distribution and grouped them into four groups (Q1, Q2, Q3, Q4). For continuous variables, mean +SD or median (IQR, skewed distribution) were used, and Kruskal-Wallis tests or weighted linear regression models were used for comparisons between the groups. To examine the association between dietary fiber intake and DN, we implemented weighted univariate and multivariate logistic regression, following the guidelines recommended by Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) [25]. This observational study developed 3 models: Model 1 (unadjusted); Model 2 (adjusted for age, gender, race, education, and poverty income ratio); and Model 3 (adjusted for all covariates). Trend analysis was also conducted using the median dietary fiber intake quartiles as a continuous variable. In addition, we stratified the analysis by gender and race. A smoothed curve fitting algorithm and generalized additive modeling were used to examine the non-linear relationship between dietary fiber intake and DN. Recursive algorithms were used to calculate the inflection point of nonlinear relationships. Based on the log-likelihood ratio test, construct a two-segment linear model on either side of the inflection point. If p > 0.05, the log-likelihood ratio test indicates a linear relationship, otherwise, a non-linear relationship.
3. Results
3.1. Typical characteristics of the population
The data were selected from NHANES (2007–2020), which ultimately included 5964 subjects aged 20–85 years after passing a series of nadir criteria (Fig. 1). The mean age was 60.8 ± 13.4 years for males (52.4 %) and females (47.6 %). There were 62.4 % non-Hispanic Whites, 13.7 % non-Hispanic Blacks, 9.9 % Mexican Americans, and 14.1 % other races. Weighted population characteristics according to the presence or absence of DN (Table 1) revealed significant differences between the two groups for age, education, poverty income ratio, smoking behavior, drinking behavior, physical activity, energy intake, carbohydrate intake, dietary fiber intake, BMI, SBP, DBP, HBP, hyperlipidemia, CHF, CHD, angina, heart attack, stroke, CVD, BUN, SCR, TG, TC, SUA, HBA1C, ACR and eGFR. Gender, race, fiber supplement, ALT, AST, TBIL, WC, and HDL-C did not differ between the two groups.
Table 1.
Characteristics of the population describing the occurrence of diabetic nephropathy.
| Variable Features | Total |
Is there diabetic nephropathy? |
p-Value | |
|---|---|---|---|---|
| NO |
YES |
|||
| n = 5964 | n = 3607 | n = 2357 | ||
| Population characteristics | ||||
| Age, mean ± SD, years | 60.8 ± 13.4 | 58.0 ± 13.2 | 65.1 ± 12.6 | <0.001 |
| Gender, n (%), (95%CI) | 0.277 | |||
| Male | 52.4 (50.6,54.3) | 53.3 (50.8,55.8) | 51.0 (48.0,54.0) | |
| Female | 47.6 (45.7,49.4) | 46.7 (44.2,49.2) | 49.0 (46.0,52.0) | |
| Race, n (%), (95%CI) | 0.081 | |||
| Mexican American | 9.9 (7.9,12.2) | 10.3 (8.2,12.7) | 9.1 (7.2,11.6) | |
| Non-Hispanic White | 62.4 (58.7,65.9) | 61.2 (57.3,65.0) | 64.3 (60.4,68.1) | |
| Non-Hispanic Black | 13.7 (11.8,15.9) | 13.9 (11.8,16.2) | 13.4 (11.4,15.7) | |
| Other race | 14.1 (12.6,15.7) | 14.6 (12.9,16.5) | 13.2 (11.4,15.1) | |
| Education, n (%), (95%CI) | 0.001 | |||
| <High school | 22.7 (21.0,24.4) | 21.1 (19.1,23.3) | 25.3 (23.3,27.3) | |
| High school | 25.7 (23.9,27.7) | 24.7 (22.2,27.4) | 27.5 (25.0,30.1) | |
| >High school | 51.6 (49.5,53.8) | 54.2 (51.2,57.1) | 47.3 (44.4,50.2) | |
| Poverty income ratio, n (%), (95%CI) | <0.001 | |||
| <1.3 | 23.3 (21.8,25.0) | 21.4 (19.7,23.2) | 26.7 (24.5,28.9) | |
| 1.3–3.5 | 36.5 (34.5,38.5) | 35.0 (32.8,37.3) | 39.0 (36.3,41.8) | |
| >3.5 | 33.0 (30.7,35.3) | 36.1 (33.4,39.0) | 27.5 (24.4,30.9) | |
| Missing | 7.2 (6.3,8.2) | 7.5 (6.5,8.7) | 6.8 (5.6,8.1) | |
| Lifestyle factors | ||||
| Smoking behavior, n (%), (95%CI) | 0.011 | |||
| Never smoked | 49.1 (47.2,51.1) | 50.4 (47.7,53.1) | 46.9 (43.9,50.0) | |
| Former smoker | 35.1 (33.3,36.9) | 33.0 (30.4,35.6) | 38.7 (35.9,41.6) | |
| Current smoker | 15.8 (14.6,17.1) | 16.7 (15.0,18.5) | 14.4 (12.5,16.4) | |
| Drinking behavior, n (%), (95%CI) | <0.001 | |||
| Never drank | 12.2 (11.0,13.4) | 11.2 (9.7,12.8) | 13.9 (12.2,15.9) | |
| Moderate drinker | 37.9 (35.7,40.0) | 38.4 (35.6,41.3) | 36.9 (34.5,39.4) | |
| Heavy drinker | 31.4 (29.3,33.6) | 33.9 (31.0,36.8) | 27.2 (24.5,30.1) | |
| Missing | 18.5 (17.2,19.9) | 16.6 (15.0,18.3) | 21.9 (19.8,24.3) | |
| Physical activity, n (%), (95%CI) | <0.001 | |||
| vigorous activity | 41.5 (39.5,43.4) | 37.4 (34.8,40.1) | 48.5 (46.0,51.0) | |
| moderate activity | 34.9 (33.0,36.9) | 36.2 (33.8,38.8) | 32.8 (30.1,35.5) | |
| light activity | 23.6 (21.9,25.4) | 26.4 (24.3,28.6) | 18.8 (16.6,21.1) | |
| Dietary fiber intake, median (IQR), g/d | 14.4 (10.2–20.6) | 15.2 (10.5–21.6) | 13.6 (9.6–18.9) | <0.001 |
| Energy intake, mean, mean ± SD, kcal/d | 1849.7 ± 782.3 | 1896.0 ± 788.7 | 1778.9 ± 767.2 | <0.001 |
| Carbohydrate intake, mean ± SD, g/d | 219.0 ± 97.0 | 224.8 ± 97.9 | 210.1 ± 94.9 | <0.001 |
| Dietary fiber supplement, n (%), (95%CI) | 3.5 (2.8,4.3) | 3.8 (2.9,5.0) | 2.9 (2.0,4.1) | 0.245 |
| Biochemical examination | ||||
| ALT, median (IQR), U/L | 21.0 (16.0–30.0) | 22.0 (17.0–31.0) | 20.0 (15.0–27.0) | 0.097 |
| AST, median (IQR), U/L | 22.0 (18.0–28.0) | 23.0 (19.0–28.0) | 22.0 (18.0–28.0) | 0.963 |
| BUN, median (IQR), mg/dL | 15.0 (11.0–19.0) | 13.0 (11.0–16.0) | 18.0 (13.0–24.0) | <0.001 |
| SCR, median (IQR), mg/dL | 0.9 (0.7–1.1) | 0.8 (0.7–0.9) | 1.1 (0.8–1.4) | <0.001 |
| TBIL, mean ± SD, mg/dL | 0.6 ± 0.3 | 0.6 ± 0.3 | 0.6 ± 0.3 | 0.262 |
| TG, median (IQR), mg/dL | 149.0 (102.0–221.0) | 145.0 (99.0–215.0) | 153.0 (105.0–232.0) | <0.001 |
| TC, mean ± SD, mg/dL | 185.1 ± 46.8 | 186.9 ± 45.9 | 182.2 ± 48.1 | 0.027 |
| HDL-C, mean ± SD, mg/dL | 48.1 ± 14.7 | 48.4 ± 14.3 | 47.7 ± 15.2 | 0.572 |
| SUA, mean ± SD, mg/dL | 5.8 ± 1.6 | 5.5 ± 1.4 | 6.2 ± 1.8 | <0.001 |
| HBA1C, mean ± SD, % | 7.3 ± 1.7 | 7.1 ± 1.6 | 7.5 ± 1.9 | <0.001 |
| ACR, median (IQR), mg/g | 12.6 (6.8–39.4) | 8.6 (5.7–13.9) | 59.0 (26.9–174.7) | <0.001 |
| eGFR, mean ± SD, mL/min/1.73 m2 | 83.7 ± 28.8 | 92.8 ± 22.4 | 69.7 ± 31.6 | <0.001 |
| Medical examination data | ||||
| BMI, n (%), (95%CI) | <0.001 | |||
| <25.0 kg/m2 | 11.3 (10.2,12.4) | 11.0 (9.6,12.6) | 11.7 (10.1,13.6) | |
| 25.0–29.9 kg/m2 | 25.9 (24.5,27.3) | 27.1 (25.2,29.2) | 23.7 (21.8,25.8) | |
| >29.9 kg/m2 | 61.7 (59.8,63.6) | 61.2 (58.7,63.7) | 62.6 (60.1,65.1) | |
| Missing | 1.1 (0.8,1.5) | 0.6 (0.4,1.0) | 1.9 (1.4,2.8) | |
| WC, mean ± SD, cm | 109.3 ± 15.6 | 108.8 ± 15.8 | 110.0 ± 15.2 | 0.062 |
| SBP, mean ± SD, mmHg | 131.0 ± 19.5 | 127.9 ± 17.2 | 135.7 ± 21.8 | <0.001 |
| DBP, mean ± SD, mmHg | 69.9 ± 14.3 | 70.8 ± 13.1 | 68.5 ± 15.7 | <0.001 |
| Disease diagnosis | ||||
| HBP, n (%), (95%CI) | 73.0 (71.1,74.7) | 67.6 (65.2,70.0) | 82.2 (80.1,84.2) | <0.001 |
| Hyperlipidemia, n (%), (95%CI) | 70.3 (68.7,71.9) | 67.7 (65.6,69.8) | 74.8 (72.5,77.0) | <0.001 |
| CHF, n (%), (95%CI) | 7.8 (6.8,8.9) | 4.2 (3.3,5.3) | 13.9 (11.9,16.1) | <0.001 |
| CHD, n (%), (95%CI) | 10.7 (9.5,12.0) | 8.1 (6.7,9.7) | 15.2 (13.4,17.3) | <0.001 |
| Angina, n (%), (95%CI) | 6.9 (6.1,7.8) | 5.7 (4.5,7.1) | 9.0 (7.5,10.7) | 0.005 |
| Heart attack, n (%), (95%CI) | 9.9 (8.9,11.0) | 7.7 (6.6,9.0) | 13.7 (11.9,15.8) | <0.001 |
| Stroke, n (%), (95%CI) | 7.4 (6.5,8.3) | 4.8 (3.9,5.8) | 11.8 (10.0,13.9) | <0.001 |
| CVD, n (%), (95%CI) | 21.3 (19.9,22.9) | 16.0 (14.3,17.9) | 30.4 (27.7,33.4) | <0.001 |
3.2. DN correlated univariate analysis
A weighted one-way analysis of variance (Table 2) showed that dietary fiber intake, age, education, poverty income ratio, energy intake, carbohydrate intake, SBP, DBP, HBP, smoking behavior, drinking behavior, physical activity, BUN, SCR, TC, TG, SUA, HBA1C, hyperlipidemia, and CVD were significantly correlated to DN. However, gender, race, fiber supplement, WC, BMI, ALT, AST, TBIL, and HDL-C were not statistically significant with DN. Having HBP was associated with a higher risk compared to not having HBP (OR = 2.22, 95%CI: 1.88, 2.63). Having hyperlipidemia was associated with a higher risk compared to not having hyperlipidemia (OR = 1.42, 95%CI: 1.22, 1.65). Having CVD was associated with a higher risk compared to not having CVD (OR = 2.29, 95%CI: 1.89, 2.77). In addition, dietary fiber intake, and DBP, were negatively associated with the risk of developing DN, playing as protective factors. On the contrary, age, energy intake, carbohydrate intake, SBP, BUN, SCR, TC, TG, SUA, and HBA1C were positively associated with the risk of developing DN, acting as risk factors.
Table 2.
Univariate analysis of risk factors for diabetic nephropathy, weighted.
| Variable | N | DN |
|
|---|---|---|---|
| OR (95%CI) | p-value | ||
| Fiber intake (log2 transform) | 5964 | 0.78 (0.72, 0.84) | <0.001 |
| Age | 5964 | 1.05 (1.04, 1.05) | <0.001 |
| Gender | |||
| Male | 3126 | Ref. | |
| Female | 2838 | 1.10 (0.93, 1.29) | 0.280 |
| Race | |||
| Mexican American | 1048 | Ref. | |
| Non-Hispanic White | 2121 | 1.18 (0.99, 1.40) | 0.065 |
| Non-Hispanic Black | 1495 | 1.08 (0.90, 1.29) | 0.405 |
| Other race | 1300 | 1.01 (0.83, 1.24) | 0.924 |
| Education | |||
| <High school | 1946 | Ref. | |
| High school | 1403 | 0.93 (0.77, 1.13) | 0.456 |
| >High school | 2615 | 0.73 (0.61, 0.87) | 0.001 |
| Poverty income ratio | |||
| <1.31 | 1910 | Ref. | |
| 1.31–3.50 | 2147 | 0.89 (0.79, 1.01) | 0.070 |
| >3.50 | 1334 | 0.61 (0.50, 0.74) | <0.001 |
| Missing | 573 | 0.72 (0.59, 0.89) | 0.003 |
| Fiber supplement | |||
| No | 5806 | Ref. | |
| Yes | 158 | 0.76 (0.47, 1.21) | 0.249 |
| Energy intake | 5964 | 1.00 (1.00, 1.00) | <0.001 |
| Carbohydrate intake | 5964 | 1.00 (1.00, 1.00) | <0.001 |
| SBP | 5964 | 1.02 (1.02, 1.03) | <0.001 |
| DBP | 5964 | 0.99 (0.98, 0.99) | <0.001 |
| HBP | |||
| No | 1497 | Ref. | |
| Yes | 4467 | 2.22 (1.88, 2.63) | <0.001 |
| WC | 5964 | 1.00 (1.00, 1.01) | 0.063 |
| BMI | |||
| <25.0 kg/m2 | 762 | Ref. | |
| 25.0–29.9 kg/m2 | 1683 | 0.82 (0.63, 1.07) | 0.146 |
| >29.9 kg/m2 | 3441 | 0.96 (0.76, 1.22) | 0.738 |
| Missing | 78 | 2.98 (1.58, 5.61) | 0.001 |
| Smoking behavior | |||
| Never smoked | 3027 | Ref. | |
| Former smoker | 1972 | 1.26 (1.04, 1.53) | 0.020 |
| Current smoker | 965 | 0.92 (0.74, 1.16) | 0.492 |
| Drinking behavior | |||
| Never drank | 791 | Ref. | |
| Moderate drinker | 2078 | 0.77 (0.62, 0.95) | 0.019 |
| Heavy drinker | 1729 | 0.64 (0.50, 0.83) | 0.001 |
| Missing | 1366 | 1.06 (0.83, 1.35) | 0.641 |
| Physical activity | |||
| Vigorous activity | 2745 | Ref. | |
| Moderate activity | 1901 | 0.70 (0.59, 0.83) | <0.001 |
| Light activity | 1318 | 0.55 (0.46, 0.65) | <0.001 |
| ALT | 5964 | 1.00 (0.99, 1.00) | 0.397 |
| AST | 5964 | 1.00 (1.00, 1.00) | 0.964 |
| BUN | 5964 | 1.14 (1.12, 1.15) | <0.001 |
| SCR | 5964 | 45.54 (32.84, 63.15) | <0.001 |
| TBIL | 5964 | 0.89 (0.73, 1.09) | 0.261 |
| TC | 5964 | 1.00 (1.00, 1.00) | 0.024 |
| TG | 5964 | 1.00 (1.00, 1.00) | 0.032 |
| SUA | 5964 | 1.37 (1.31, 1.43) | <0.001 |
| HBA1C | 5964 | 1.18 (1.14, 1.23) | <0.001 |
| HDL-C | 5964 | 1.00 (0.99, 1.00) | 0.589 |
| Hyperlipidemia | |||
| No | 1856 | Ref. | |
| Yes | 4108 | 1.42 (1.22, 1.65) | <0.001 |
| CVD | |||
| No | 4656 | Ref. | |
| Yes | 1308 | 2.29 (1.89, 2.77) | <0.001 |
3.3. Multiple logistic regression analysis of dietary fiber intake and DN
To obtain the relationship between dietary fiber and DN, we used three models to analyze the effect values (Table 3). Unadjusted modeling results indicated (Model 1) that dietary fiber was negatively associated with the risk of DN (OR = 0.78, 95%CI: 0.72, 0.84). This indicated that for every 1 unit of dietary fiber (log2 transform), there was a 22 % reduction in the risk of developing DN. The results of the micro-adjustment model showed (Model 2) that dietary fiber was negatively associated with the risk of DN (OR = 0.77, 95%CI: 0.72, 0.83). The equivalent of each 1-unit in dietary fiber (log2 transform) led to a 23 % reduction in the risk of developing DN. The results of the fully adjusted model showed (Model 3) that dietary fiber was negatively associated with the risk of DN (OR = 0.89, 95%CI: 0.80, 0.99). The equivalent of each 1 unit in dietary fiber (log2 transform) resulted in an 11 % reduction in the risk of developing DN. To verify the stability of the results, dietary fiber was grouped into quartiles (Q1, Q2, Q3, Q4) and validated in three models (Table 3). In Model 1, the risk of DN was negatively correlated in Q3 and Q4 compared to Q1, respectively (OR = 0.75, 95%CI: 0.63, 0.91; OR = 0.56, 95%CI: 0.48, 0.67). In Model 2, the risk of DN was negatively correlated in Q3 and Q4 compared to Q1, respectively (OR = 0.71, 95%CI: 0.59, 0.85; OR = 0.56, 95%CI: 0.47, 0.67). In Model 3, the risk of DN was negatively correlated in Q4 compared to Q1. Overall, the risk of developing DN tended to decrease with increasing dietary fiber intake (Q1-Q4) in all 3 models, and the test for trend was consistently significant (model 1, p for trend <0.001; model 2, p for trend <0.001; model 3, p for trend = 0.003), with results consistent with those for continuous variables. We plotted a smoothed curve fit (p for non-linearity = 0.064) through Model 3 to better see if there was a non-linear relationship between dietary fiber intake and DN (Fig. 2).
Table 3.
The relationship between fiber intake and diabetic nephropathy, weighted.
| Exposure variables | Model 1 |
Model 2 |
Model 3 |
|---|---|---|---|
| OR(95 % CI) | OR(95 % CI) | OR(95 % CI) | |
| Fiber intake (log2 transform) per 1 mg/dL increase | 0.78 (0.72, 0.84) | 0.77 (0.72, 0.83) | 0.89 (0.80, 0.99) |
| Fiber intake (log2 transform) (quartiles) | |||
| Q1 | Ref. | Ref. | Ref. |
| Q2 | 0.93 (0.78, 1.12) | 0.89 (0.74, 1.07) | 1.02 (0.81, 1.28) |
| Q3 | 0.75 (0.63, 0.91) | 0.71 (0.59, 0.85) | 0.83 (0.66, 1.05) |
| Q4 | 0.56 (0.48, 0.67) | 0.56 (0.47, 0.67) | 0.73 (0.57, 0.93) |
| P for trend | <0.001 | <0.001 | 0.003 |
Model 1: unadjusted.
Model 2: adjust for age, gender, race, education, and poverty income ratio.
Model 3: adjust for age, gender, race, education, poverty income ratio, smoking behavior; drinking behavior; physical activity, fiber supplement, energy intake, carbohydrate intake, SBP, DBP, HBP, BMI, WC, ALT, AST, BUN, SCR, TBIL, HBA1C, TG, TC, HDL-C, SUA, hyperlipidemia, CVD.
Fig. 2.
Relationship between fiber intake and risk of diabetic nephropathy (Model 3). The area between the two blue curves represents the 95 % CI.
3.4. Analyzing the correlation between dietary fiber and DN in subgroups
We further conducted a subgroup analysis stratified by gender and race (Table 4). In the study, females and males exhibited negative correlations (OR = 0.87, 95%CI: 0.74, 1.01; OR = 0.91, 95%CI: 0.78, 1.06), but the results were not significant. Non-Hispanic White, non-Hispanic Black, and other race were all negatively correlated with dietary fiber intake (OR = 0.88, 95%CI: 0.76, 1.01; OR = 0.94, 95%CI: 0.79, 1.12; OR = 0.77, 95%CI: 0.60, 0.99), and only other race has a significant correlation. Mexican Americans were positively correlated (OR = 1.09, 95%CI: 0.85, 1.40). In addition, we drew smooth curves for gender and race-stratified subgroups to further observe non-linear relationships (Fig. 3, Fig. 4). We found a non-linear inverted U-shaped relationship among male and non-Hispanic blacks (p for log-likelihood ratio tests <0.05). A two-section model was used to calculate the inflection point (Table 5). For males (the inflection point was 8.0 g/d), when the dietary fiber intake was >8.0 g/d, the risk of DN was reduced by 23 % (OR = 0.77, 95%CI: 0.64, 0.91). For non-Hispanic Black (the turning point is 14.9 g/d), when the dietary fiber intake of non-Hispanic Black was >14.9 g/d, the risk of DN was reduced by 44 % (OR = 0.56, 95%CI: 0.35, 0.89).
Table 4.
A weighted study of fiber intake and diabetic nephropathy stratified by race, gender, and high blood pressure.
| Exposure variables | Model 1 |
Model 2 |
Model 3 |
|---|---|---|---|
| OR(95 % CI) | OR(95 % CI) | OR(95 % CI) | |
| Subgroup analysis stratified by | |||
| Gender | |||
| Male | 0.73(0.65, 0.82) | 0.73 (0.65, 0.82) | 0.87 (0.74, 1.01) |
| Female | 0.83 (0.74, 0.93) | 0.81 (0.73, 0.91) | 0.91 (0.78, 1.06) |
| Race | |||
| Mexican American | 0.79 (0.68, 0.92) | 0.88 (0.74, 1.03) | 1.09 (0.85, 1.40) |
| Non-Hispanic White | 0.77 (0.68, 0.86) | 0.75 (0.67, 0.84) | 0.88 (0.76, 1.01) |
| Non-Hispanic Black | 0.88 (0.76, 1.01) | 0.87 (0.75, 1.02) | 0.94 (0.79, 1.12) |
| Other race | 0.68 (0.57, 0.80) | 0.68 (0.57, 0.80) | 0.77 (0.60, 0.99) |
Model 1: unadjusted.
Model 2: adjust for age, gender, race, education, and poverty income ratio.
Model 3: adjust for age, gender, race, education, poverty income ratio, smoking behavior; drinking behavior; physical activity, fiber supplement, energy intake, carbohydrate intake, SBP, DBP, HBP, BMI, WC, ALT, AST, BUN, SCR, TBIL, HBA1C, TG, TC, HDL-C, SUA, hyperlipidemia, CVD.
When stratifying subgroups for gender race and HBP, the stratification variables are not adjusted in the model.
Fig. 3.
Association between fiber intake and diabetic nephropathy stratified by gender (Model 3).
Fig. 4.
Association between fiber intake and diabetic nephropathy stratified by race (Model 3).
Table 5.
Based on a two-piecewise linear regression model, the results indicate that fiber intake is related to Diabetic Nephropathy.
| Fiber intake (log2 transform) | Adjusted OR (95 % CI) |
|---|---|
| Male | |
| The inflection points of fiber intake (g/d) (log2 transform) | 3.0 (log2 8.0) |
| Regression coefficients (≤ inflection point) | 1.22 (0.85, 1.74) |
| Regression coefficients (> inflection point) | 0.77 (0.64, 0.91) |
| P for log-likelihood ratio tests | 0.033 |
| Non-Hispanic Black | |
| Inflection points of fiber intake (g/d) (log2 transform) | 3.9 (log2 14.9) |
| Regression coefficients (≤ inflection point) | 1.20 (0.93, 1.56) |
| Regression coefficients (> inflection point) | 0.56 (0.35, 0.89) |
| P for log-likelihood ratio tests | 0.006 |
adjust for age, gender, race, education, poverty income ratio, smoking behavior; drinking behavior; physical activity, fiber supplement, energy intake, carbohydrate intake, SBP, DBP, HBP, BMI, WC, ALT, AST, BUN, SCR, TBIL, HBA1C, TG, TC, HDL-C, SUA, Hyperlipidemia, CVD. When stratifying subgroups for gender and race, the stratification variables are not adjusted in the model.
4. Discussion
An investigation of the relationship between dietary fiber intake and DN was undertaken in this study. Results showed dietary fiber intake is negatively related to DN risk (OR = 0.89, 95%CI: 0.80, 0.99). By increasing dietary fiber by 1 unit, the risk of developing DN was reduced by 11 % (log2 transformed), suggesting that there may be an association between dietary fiber intake and DN, but a causal relationship could not be established. In addition, dietary fiber intake showed a nonlinear U-shaped association with the prevalence of DN in male and non-Hispanic blacks. From the results, we have gained important insights into the possible relationship between dietary fiber intake and DN. Further confirmation is needed by causal studies.
According to this study, dietary fiber intake is negatively correlated with DN, which means that the more fiber you consume, the lower your risk for DN. Dietary fiber has been strongly associated with T2DM in recent epidemiological studies [26,27]. A study from a Danish cohort was conducted including 55,465 participants aged 50–65 years and followed them for 15 years. According to the researchers, eating whole grains reduced the risk of T2DM [28]. There was also a negative association between the total dietary fiber, soluble fiber, and insoluble fiber and the risk of T2DM in another prospective cohort study from France [29]. Studies have shown that dietary fiber reduces fasting blood glucose [30,31], which reduces the risk of albuminuria in diabetic patients and mortality in patients with DN [32,33]. In addition, high dietary fiber intake can reduce BUN and SCR levels [34,35], improve eGFR, and reduce the risk of chronic kidney disease [36,37]. A study suggests dietary fiber prevents DN through modulation of the gut microbiota's production of short-chain fatty acids (SCFA) and activation of G-protein-coupled receptors GPR43 and GPR109A [38,39]. In addition, dietary fiber may further protect renal function by altering the gut microbial community and reducing inflammation [15]. This probably partly explains the consistency of our findings. Therefore, patients who have been diagnosed with DM, by adjust their diet, especially by increasing the intake of dietary fiber. May help to better control blood glucose levels, thereby reducing the risk of complications such as DN. This is particularly important in the daily management of DM patients.
The second finding is that male, non-Hispanic Black showed a non-linear inverted U-shaped association. Appropriately increasing dietary fiber intake may reduce DN risk if it is on the right side of the inflection point. For males, there are differences in metabolism, hormonal and body composition, and activity, which may result in better maintaining stable blood glucose and kidney function after dietary fiber intake [40]. For non-Hispanic blacks, it may be related to genetic and environmental factors. The results of this study provide evidence of a complex relationship between dietary fiber and DN. To better understand this association, we need to further study different populations and the underlying biological mechanisms. We also emphasize the importance of personalized nutritional programs, which may require different dietary fiber intake recommendations for different populations.
This study has some limitations despite utilizing NHANES data and containing broad representative data. The first limitation of this study is that it is a cross-sectional study and cannot determine whether dietary fiber or DN are causally related. Although we observed an association, this does not mean that high dietary fiber intake directly led to a reduction in the prevalence of DN. Second, the 24-h recall method was used to collect dietary fiber intake and the diagnosis of DM was self-reported, so recall bias could have been introduced. Third, the findings may not apply to populations outside of the exclusion criteria, since we performed a series of nadir exclusion criteria. In addition, due to the complexity of the confounders, we could not rule out the influence of other potential confounders on the results of the study.
5. Conclusions
Dietary fiber intake may be associated with the risk of developing DN, and the relationship may depend on population characteristics. Our findings may emphasize that increasing dietary fiber intake may be an effective strategy for the management of DN. Our wish is that future studies, especially cohort studies or randomized controlled trials, will need to further validate this association and establish causality.
Funding
This research was funded by the National Key Clinical Specialties Construction Program.
Institutional review board Statement
The National Centre for Health Statistics’ research ethics review board approved the NHANES study (https://www.cdc.gov/nchs/nhanes/irba98.htm).
Informed consent Statement
All the NHANES participants provided their written informed consent.
Data Availability Statement
The data used in this study are publicly available online (https://wwwn.cdc.gov/nchs/nhanes/, accessed on November 1, 2023).
CRediT authorship contribution statement
Hao Yang: Writing – original draft, Software. Hong Lin: Writing – original draft, Methodology. Xiaorong Liu: Validation. Haoran Liu: Validation. Ting Chen: Software. Zhaohui Jin: Writing – review & editing, Funding acquisition, Conceptualization.
Declaration of competing interest
I, [Hao Yang ], declare that I have no conflicts of interest regarding the research presented in the manuscript entitled "[Title of Association Between Dietary Fiber Intake and Diabetic Nephropathy among Adults in the United States: A Cross-Sectional Study]." I confirm that I have no financial or personal relationships with individuals or organizations that could inappropriately influence or bias my work. Additionally, I have no competing interests, including but not limited to employment, consultancies, stock ownership, honoraria, paid expert testimony, patent applications/registrations, and grants or other funding.
Should any potential conflicts of interest arise during the review process or after publication, I will promptly disclose this information to the appropriate parties.
Acknowledgment
The authors gratefully thank the NHANES of Centers for Disease Control and Prevention for providing the publicly available data.
References
- 1.Wu Y.L., et al. Epigenetic regulation in metabolic diseases: mechanisms and advances in clinical study. Signal Transduct Target Ther. 2023;8(1):98. doi: 10.1038/s41392-023-01333-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Patel D.M., et al. Glucose and blood pressure-Dependent Pathways-the progression of diabetic kidney disease. Int. J. Mol. Sci. 2020;21(6) doi: 10.3390/ijms21062218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Lavoz C., et al. Interleukin-17A blockade reduces albuminuria and kidney injury in an accelerated model of diabetic nephropathy. Kidney Int. 2019;95(6):1418–1432. doi: 10.1016/j.kint.2018.12.031. [DOI] [PubMed] [Google Scholar]
- 4.Umanath K., Lewis J.B. Update on diabetic nephropathy: Core Curriculum 2018. Am. J. Kidney Dis. 2018;71(6):884–895. doi: 10.1053/j.ajkd.2017.10.026. [DOI] [PubMed] [Google Scholar]
- 5.Cardoso C., et al. Prognostic importance of visit-to-visit blood pressure variability for micro- and macrovascular outcomes in patients with type 2 diabetes: the Rio de Janeiro Type 2 Diabetes Cohort Study. Cardiovasc. Diabetol. 2020;19(1):50. doi: 10.1186/s12933-020-01030-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Garofalo C., et al. A systematic review and meta-analysis suggests obesity predicts onset of chronic kidney disease in the general population. Kidney Int. 2017;91(5):1224–1235. doi: 10.1016/j.kint.2016.12.013. [DOI] [PubMed] [Google Scholar]
- 7.Ko G.J., et al. Dietary Approaches in the management of diabetic patients with kidney disease. Nutrients. 2017;9(8) doi: 10.3390/nu9080824. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Ma R., et al. Retinol intake is associated with the risk of chronic kidney disease in individuals with type 2 diabetes mellitus: results from NHANES. Sci. Rep. 2023;13(1) doi: 10.1038/s41598-023-38582-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Sulaiman M.K. Diabetic nephropathy: recent advances in pathophysiology and challenges in dietary management. Diabetol Metab Syndr. 2019;11:7. doi: 10.1186/s13098-019-0403-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Schroeder B.O., et al. Bifidobacteria or fiber protects against diet-Induced microbiota-Mediated Colonic Mucus Deterioration. Cell Host Microbe. 2018;23(1):27–40. doi: 10.1016/j.chom.2017.11.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Reynolds A., et al. Carbohydrate quality and human health: a series of systematic reviews and meta-analyses. Lancet. 2019;393(10170):434–445. doi: 10.1016/S0140-6736(18)31809-9. [DOI] [PubMed] [Google Scholar]
- 12.Miller V., et al. Defining diet quality: a synthesis of dietary quality metrics and their validity for the double burden of malnutrition. Lancet Planet. Health. 2020;4(8):e352–e370. doi: 10.1016/S2542-5196(20)30162-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Kwon Y.J., et al. Association between dietary fiber intake and all-cause and cardiovascular mortality in middle aged and Elderly adults with chronic kidney disease. Front. Nutr. 2022;9 doi: 10.3389/fnut.2022.863391. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Barber T.M., et al. The health benefits of dietary fibre. Nutrients. 2020;12(10) doi: 10.3390/nu12103209. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ma W., et al. Dietary fiber intake, the gut microbiome, and chronic systemic inflammation in a cohort of adult men. Genome Med. 2021;13(1):102. doi: 10.1186/s13073-021-00921-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Chen T.C., et al. National health and nutrition examination survey, 2015-2018: sample Design and estimation Procedures. Vital Health Stat. 2020;2(184):1–35. [PubMed] [Google Scholar]
- 17.Ahluwalia N., et al. Update on NHANES dietary data: Focus on collection, Release, Analytical Considerations, and Uses to Inform public policy. Adv. Nutr. 2016;7(1):121–134. doi: 10.3945/an.115.009258. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.ElSayed N.A., et al. 2. Classification and diagnosis of diabetes: Standards of Care in diabetes-2023. Diabetes Care. 2023;46(Suppl 1):S19–S40. [Google Scholar]
- 19.Afkarian M., et al. Clinical manifestations of kidney disease among US adults with diabetes, 1988-2014. JAMA. 2016;316(6):602–610. doi: 10.1001/jama.2016.10924. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.ElSayed N.A., et al. 11. Chronic kidney disease and risk management: Standards of Care in diabetes-2023. Diabetes Care. 2023;46(Suppl 1):S191–S202. doi: 10.2337/dc23-S011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Liu L., et al. Association between the triglyceride-glucose index and diabetic nephropathy in patients with type 2 diabetes: a cross-sectional study. J Diabetes Investig. 2021;12(4):557–565. doi: 10.1111/jdi.13371. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Guo W., et al. Systemic immune-inflammation index is associated with diabetic kidney disease in Type 2 diabetes mellitus patients: evidence from NHANES 2011-2018. Front. Endocrinol. 2022;13 doi: 10.3389/fendo.2022.1071465. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Wang J., et al. Association between Globulin and diabetic nephropathy in Type2 diabetes mellitus patients: a cross-sectional study. Front. Endocrinol. 2022;13 doi: 10.3389/fendo.2022.890273. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Whelton P.K., et al. ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA guideline for the prevention, Detection, evaluation, and management of high blood pressure in adults: Executive summary: a report of the American College of Cardiology/American heart association Task Force on clinical practice guidelines. Hypertension. 2018;71(6):1269–1324. doi: 10.1161/HYP.0000000000000066. 2017. [DOI] [PubMed] [Google Scholar]
- 25.Vandenbroucke J.P., et al. Strengthening the reporting of observational studies in Epidemiology (STROBE): explanation and elaboration. Int. J. Surg. 2014;12(12):1500–1524. doi: 10.1016/j.ijsu.2014.07.014. [DOI] [PubMed] [Google Scholar]
- 26.Piepoli M.F., et al. European guidelines on cardiovascular disease prevention in clinical practice: the Sixth Joint Task Force of the European Society of Cardiology and other Societies on cardiovascular disease prevention in clinical practice (constituted by representatives of 10 societies and by invited experts)Developed with the special contribution of the European association for cardiovascular prevention & Rehabilitation (EACPR) Eur. Heart J. 2016;37(29):2315–2381. doi: 10.1093/eurheartj/ehw106. 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Jin F., et al. Association of dietary fiber intake with newly-diagnosed type 2 diabetes mellitus in middle-aged Chinese population. Nutr. J. 2021;20(1):81. doi: 10.1186/s12937-021-00740-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Kyrø C., et al. Higher whole-Grain intake is associated with lower risk of type 2 diabetes among middle-aged men and Women: the Danish diet, cancer, and health cohort. J. Nutr. 2018;148(9):1434–1444. doi: 10.1093/jn/nxy112. [DOI] [PubMed] [Google Scholar]
- 29.Partula V., et al. Associations between consumption of dietary fibers and the risk of cardiovascular diseases, cancers, type 2 diabetes, and mortality in the prospective NutriNet-Santé cohort. Am. J. Clin. Nutr. 2020;112(1):195–207. doi: 10.1093/ajcn/nqaa063. [DOI] [PubMed] [Google Scholar]
- 30.Reynolds A.N., et al. Dietary fibre and whole grains in diabetes management: systematic review and meta-analyses. PLoS Med. 2020;17(3) doi: 10.1371/journal.pmed.1003053. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Xie Y., et al. Effects of soluble fiber supplementation on glycemic control in adults with type 2 diabetes mellitus: a systematic review and meta-analysis of randomized controlled trials. Clin Nutr. 2021;40(4):1800–1810. doi: 10.1016/j.clnu.2020.10.032. [DOI] [PubMed] [Google Scholar]
- 32.Ruospo M., et al. Glucose targets for preventing diabetic kidney disease and its progression. Cochrane Database Syst. Rev. 2017;6(6) doi: 10.1002/14651858.CD010137.pub2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Yoon J.J., et al. Dianthus superbus improves glomerular Fibrosis and renal Dysfunction in diabetic nephropathy model. Nutrients. 2019;11(3) doi: 10.3390/nu11030553. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Salmean Y.A., et al. Foods with added fiber lower serum creatinine levels in patients with chronic kidney disease. J. Ren. Nutr. 2013;23(2):e29–e32. doi: 10.1053/j.jrn.2012.04.002. [DOI] [PubMed] [Google Scholar]
- 35.Chiavaroli L., et al. Dietary fiber effects in chronic kidney disease: a systematic review and meta-analysis of controlled feeding trials. Eur. J. Clin. Nutr. 2015;69(7):761–768. doi: 10.1038/ejcn.2014.237. [DOI] [PubMed] [Google Scholar]
- 36.Krishnamurthy V.M., et al. High dietary fiber intake is associated with decreased inflammation and all-cause mortality in patients with chronic kidney disease. Kidney Int. 2012;81(3):300–306. doi: 10.1038/ki.2011.355. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Mirmiran P., et al. Dietary fibre intake in relation to the risk of incident chronic kidney disease. Br. J. Nutr. 2018;119(5):479–485. doi: 10.1017/S0007114517003671. [DOI] [PubMed] [Google Scholar]
- 38.Zhao L., et al. Gut bacteria selectively promoted by dietary fibers alleviate type 2 diabetes. Science. 2018;359(6380):1151–1156. doi: 10.1126/science.aao5774. [DOI] [PubMed] [Google Scholar]
- 39.Li Y.J., et al. Dietary fiber protects against diabetic nephropathy through short-chain fatty acid-Mediated activation of G protein-coupled receptors GPR43 and GPR109A. J. Am. Soc. Nephrol. 2020;31(6):1267–1281. doi: 10.1681/ASN.2019101029. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Whelton P.K., et al. ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA guideline for the prevention, Detection, evaluation, and management of high blood pressure in adults: Executive summary: a report of the American College of Cardiology/American heart association Task Force on clinical practice guidelines. Circulation. 2018;138(17):e426–e483. doi: 10.1161/CIR.0000000000000597. 2017. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data used in this study are publicly available online (https://wwwn.cdc.gov/nchs/nhanes/, accessed on November 1, 2023).