Weight-adjusted-waist index is positively associated with fecal incontinence: a cross-sectional study

Abstract

Introduction

Fecal incontinence (FI), a prevalent condition affecting approximately 7.7% of the global population and 8.3% of Americans, significantly impairs quality of life. Although FI is closely associated with obesity, the specific impact of weight-adjusted waist index (WWI) on FI remains unclear. This study aimed to investigate the association between this novel anthropometric indicator, WWI, and FI among American adults.

Methods

This cross-sectional study enrolled 12,922 participants from the National Health and Nutrition Examination Survey (NHANES). WWI was calculated as waist circumference (cm) divided by the square root of weight (kg). FI was defined as any involuntary loss of mucus, liquid, or solid stool in the past month, assessed via the NHANES Bowel Health Questionnaire. Weighted multivariable logistic regression analyses were performed to evaluate the association between WWI and FI. Furthermore, we utilized smoothing curve fitting to elucidate potential linear relationships. The predictive performance of WWI, body mass index (BMI), and waist circumference (WC) in relation to FI was assessed using the receiver operating characteristic curve analysis and DeLong’s non-parametric test.

Results

The overall prevalence of FI was 8.14%. Weighted multivariable logistic regression analyses indicated that each one-unit increase in WWI was associated with a 36% higher prevalence of FI (OR = 1.36, 95% CI:1.20–1.55; P < 0.001). When WWI was categorized into tertiles and compared to the lowest tertile, the highest tertile maintained a positive association with FI (OR = 1.63, 95% CI:1.31–2.02; P < 0.001). Smoothing curve fitting revealed a linear dose–response relationship between WWI and FI. Subgroup analysis indicated no significant interactions (all P > 0.05). Additionally, our results suggested that the correlation between WWI and FI was stronger than that between BMI or WC and FI.

Conclusions

WWI is independently associated with FI, suggesting its potential utility in clinical assessment. WWI may refine risk stratification in obesity management strategies.

Supplementary Information

The online version contains supplementary material available at 10.1186/s41043-025-01050-7.

Introduction

Fecal incontinence (FI) is a prevalent and debilitating condition that affects individuals worldwide. It continues to pose a significant global burden, impacting approximately 7.7% of the global population [1], and highly affecting 8.3% of American population [2]. FI is recognized as a substantial public health challenge, with millions of individuals suffering from its consequences. These consequences can lead to decreased quality of life and reduced life expectancy. Despite its widespread impact, there has been a lack of research focusing on identifying protective factors against FI, making it a persistent challenge in the field.

FI, characterized by symptoms such as involuntary or uncontrollable bowel movements, typically results from a combination of anorectal sensorimotor dysfunctions, pelvic floor anatomical disturbances, and disrupted bowel habits [2]. Obesity, especially central obesity, significantly raises FI through multiple pathways [3, 4]. Excess abdominal fat heightens intra-abdominal pressure, causing pelvic floor dysfunction and weakening anal sphincter muscles [5]. Previous studies have supported the link between obesity and pelvic floor disorders. Overweight and obese women with urinary incontinence (UI) and FI who completed a weight loss program had an improvement of 13% overall in FI severity and frequency [6]. Abdominal obesity also brings systemic inflammation and oxidative stress, impairing anorectal functions and pelvic anatomy [3]. Moreover, obesity-related issues like diarrhea and rapid colonic transit can lead to FI [7]. While traditional measures like body mass index (BMI) and waist circumference (WC) poorly distinguish central adiposity from overall body mass [8]. Previous studies have aimed to evaluate the relationship between obesity-related body measures, such as BMI, and FI, yielded differing and inconsistent results [8–10]. The weight-adjusted waist index (WWI), calculated as WC divided by the square root of weight (WWI = WC/$\sqrt{\text{weight}}$), addresses the limitations of BMI and WC by differentiating between adipose and muscle tissues [11]. WWI addresses central obesity issues that aren't directly related to overall body weight, isolating central obesity as a key risk factor [12]. This characteristic positions WWI as a potentially superior predictor of FI. Additionally, the simplicity of WWI (requiring only waist and weight measurements) makes it a cost-effective tool for FI screening in primary care, particularly in resource-limited settings.

Previous studies have demonstrated that WWI is a more accurate predictor of obesity-related health outcomes, such as cardiovascular disease, diabetes, hypertension, digestive and urinary system diseases, cognitive impairment and certain types of cancer in medical research [12–19], compared to BMI and WC. This was attributed to WWI’s ability to better reflect central obesity, which was more closely associated with metabolic risks and clinical prognosis. Although WWI has emerged as a better predictor of obesity compared to other obesity-related indicators, the available evidence regarding the association between WWI and FI remains insufficient.

Building upon the aforementioned studies, We hypothesize that WWI, owing to its focus on central obesity and operational simplicity, may serve as a more practical and accessible predictor of FI compared to BMI or WC in clinical settings. The primary aim of this study was to investigate the association between WWI and FI, with the goal of establishing WWI as a novel indicator for FI. This could refine obesity management strategies to mitigate FI in clinical practice.

Materials and methods

Data source

The data were from the NHANES, which was conducted by the National Center for Health Statistics (NCHS), a division of the Centers for Disease Control and Prevention (CDC) [20]. NHANES is a cross-sectional survey that assesses the health and nutritional status of the noninstitutionalized civilian population in the United States. Physical examinations, interviews, and laboratory assessments were performed at mobile examination centers (MECs), following the completion of demographic, socioeconomic, and medical health interviews were conducted at homes. Participants were interviewed in a private room within the MECs if the subject matter was sensitive, such as FI. The stratified multistage probability strategy employed in the NHANES study design ensures a representative sample of the participants. The study protocol was approved by the NCHS Ethics Review Board, and informed consent was obtained from all participants prior to their involvement in the study. The protocol number of NHANES 2005 to 2010 was Protocol#2005–06.

This study utilized data from the NHANES databases spanning the 2 years cycles of 2005–2006, 2007–2008, and 2009–2010, as these were the only cycles that specifically included measurements from the Bowel Health Questionnaire. Our study concentrated on the adults who are aged 20 years and above. Participants with incomplete Bowel Health Questionnaire data or lacking available information on BMI, WC, and WWI assessments were excluded in this research. Participants with incomplete data on demographic information, socioeconomic factors, comorbid conditions, and risk behaviors were excluded from the study.

Fecal incontinence (FI) assessment

The major outcome of interest in this study was the reported FI, which was assessed utilizing data derived from the Bowel Health Questionnaire (BHQ). This questionnaire includes four items that pertain to adult fecal incontinence, closely following the criteria of Rockwood's Fecal Incontinence Severity Index (FISI) [21]. These items evaluate the frequency and type of incontinence episodes such as gas, mucus, liquid, and solid [21, 22]. The frequency of fecal incontinence episodes was categorized based on their occurrence: never, 1–3 times per month, once per week, 2 or more times per week, once per day, or 2 or more times per day. The FI was defined by the presence of any involuntary loss of mucus, liquid, or solid stool within the preceding one month. For the purposes of analysis, we classified the responses into two categories: ‘No’ for the absence of symptoms, and ‘Yes’ for the presence of symptoms at any frequency [23, 24].

Weight-adjusted-waist index (WWI) assessment

The WWI was regarded as an exposure variable based on WC and weight, which are integral in assessing central obesity. Comprehensive body measurement data, encompassing both WC and weight, were collected by certified health technicians from the NHANES, ensuring the accuracy and reliability of our analysis. The WWI of each participant was computed using the following formula (WWI = WC/$\sqrt{\text{weight}}$, with WC in centimeters and weight in kilograms). The WWI, recognized as a novel adiposity index, indicates a higher degree of obesity with increasing values [8]. In analyses, WWI was used as a continuous variable as well as a categorical variable. The categorical WWI was separated into three subgroups (T1-T3) based on the WWI tertile distribution.

Covariates of interest

Based on current scientific literature and clinical practice, covariates included demographics, socioeconomic factors, comorbid conditions, physical activity, BMI, and WC [12, 25, 26]. Demographic and socioeconomic variables collected including age (continuous variable), sex (male, female), and race/ethnicity (categorized as non-Hispanic white, non-Hispanic Black, Mexican American, other race), marital status (classified as married or living with a partner, living alone), education level (less than 9 years, 9 to 12 years, and more than 12 years). Family income was categorized into three distinct groups by the poverty income ratio (PIR): low (PIR ≤ 1.3), medium (1.3 < PIR ≤ 3.5), and high (PIR > 3.5). According to US PA guidelines. Physical activity was classified as inactive (Physical activity < 150 min/week), active (Physical activity ≥ 150 min/week) [27]. Alcohol intake was classfied as drinkers (Yes) and non-drinkers (No) (whether intaked 12 drinks or more per year). Smoking status was classified into three distinct categories: never smokers (smoked less than 100 cigarettes), current smokers, and former smokers (quit after smoking more than 100 cigarettes). A dietary recall interview at MECs to collect participants’ 24 h nutritional data, including total dietary calories, protein, carbohydrates, and fat. The determination of comorbid conditions (hypertension, diabetes, and coronary heart disease, stroke, congestive heart failure, cancer) was based on participant self-reporting in NHANES. BMI was determined utilizing a standardized technique that incorporates measurements of weight and height, with populations defined as normal weight (< 25 kg/m²), overweight (25 kg/m² ≤ BMI < 30 kg/m²), and obese (≥ 30 kg/m²).

Statistical analyses

According to NHANES analysis guidelines, statistical analyses in our study were meticulously conducted using the appropriate NHANES sampling weights. This approach accounted for the intricate sampling design of NHANES, which encompasses a complex, multistage, probability sampling design featuring cluster sampling at multiple stages. All participants underwent a descriptive analysis. According to the type of data, continuous variables were reported as mean (standard error [SE]), and categorical variables were shown as an unweighted number (weighted percentage). To compare differences among three groups divided by WWI tertiles, one-way ANOVA was employed for data that followed a normal distribution, the Kruskal–Wallis test was used for skewed data, and the chi-square test was applied to categorical variables.

A weighted multivariable logistic regression model analysis was performed to evaluate the association between WWI and FI. Model 1 served as the baseline which was unadjusted and served to provide a foundational comparison for subsequent analyses. Model 2 included adjustments for age, sex, and race/ethnicity. Model 3 further incorporated adjustments for age, sex, race/ethnicity, marital status, education level, family income, smoking status, alcohol intake, physical activity, hypertension, diabetes, coronary heart disease, congestive heart failure, cancer, calorie consumption, protein consumption, carbohydrate consumption, fat consumption, and body mass index.

Trend analyses were performed using weighted multivariate logistic regression models by incorporating the median value of each tertile of WWI as a continuous variable. To explore potential linear relationships and the dose–response curve between WWI and FI, we utilized smooth curve fitting after adjusting Model 3.

The predictive abilities of WWI, BMI and WC for FI were used to assess by the receiver operating characteristic (ROC) curve and the area under the curve (AUC). To compare the AUC values between WWI and the other two indices, DeLong’s test was utilized, which was a statistical method suitable for comparing the predictive abilities of two or more different tests [26, 28].

Subgroup analysis was performed to examine the association between WWI and FI based on subgroup variables, such as age, sex, BMI, diabetes, hypertension, smoking status, and alcohol intake. Heterogeneity across subgroups was assessed by introducing an interaction term into the model. Furthermore, we performed sensitivity analyses to assess the robustness of our findings after excluding participants aged 75 and older and addressing incomplete data through multiple imputations.

Statistical significance was assessed by examining whether the adjusted odds ratios (ORs) differed from 1.0 and by providing 95% confidence intervals. All statistical analyses were conducted using R Statistical Software (Version 4.2.2; R Foundation; http://www.R-project.org), and Free Statistics software (Version 2.0; Beijing, China; http://www.clinicalscientists.cn/freestat-istics) were used for analyses. A two-sided P value < 0.05 was considered statistically significant.

Results

Study population

A total of 12,922 patients were included in the study following a thorough screening process according to the established inclusion and exclusion criteria. Of the 17,132 participants aged ≥ 20 years, 4,210 were excluded due to incomplete FI data (n = 2,415), missing WWI data (n = 390), or covariates (n = 1,405), leaving 12,922 for analysis (Fig. 1).

Fig. 1.

The flowchart of sample selection

Baseline characteristics

The overall prevalence of fecal incontinence (FI) was 8.14%. Groups categorized by WWI tertiles exhibited significant differences in demographics, including age, sex, race/ethnicity, marital status, education level, family income, smoking status, alcohol intake, physical activity, hypertension, diabetes, coronary heart disease, and BMI (all P < 0.05, Table 1).

Table 1.

Characteristics of participants by tertiles of the WWI in the NHANES 2005–2010 cycles

Variables	Total (n = 12,992)	T1 (n = 4327) (8.11–10.64)	T2(n = 4137) (10.65–11.37)	T3 (n = 4458) (11.37–15.7)	P value
Age(years), Mean(SE)	46.39 (0.35)	38.84 (0.29)	47.86 (0.34)	55.59 (0.44)	< 0.001
Sex, n (%)					< 0.001
Male	6,342 (48.66%)	2,447 (54.00%)	2,127 (50.93%)	1,768 (38.26%)
Female	6,580 (51.34%)	1,880 (46.00%)	2,010 (49.07%)	2,690 (61.74%)
Race/ethnicity,n(%)					< 0.001
Non-Hispanic white	6,543 (72.17%)	2,213 (72.49%)	2,053 (71.42%)	2,277 (72.60%)
Non-Hispanic black	3,067 (15.85%)	1,330 (18.56%)	926 (14.83%)	811 (13.13%)
Mexican American	2,311 (7.90%)	489 (5.27%)	835 (9.51%)	987 (9.83%)
Others	1,001 (4.08%)	295 (3.68%)	323 (4.24%)	383 (4.45%)
Marital status,n(%)					< 0.001
Married or living with partners	8,013 (65.51%)	2,511 (61.59%)	2,751 (70.62%)	2,751 (65.19%)
Living alone	4,909 (34.49%)	1,816 (38.41%)	1,386 (29.38%)	1,707 (34.81%)
Education level(year), n(%)					< 0.001
< 9	1,405 (5.51%)	214 (2.89%)	421 (5.42%)	770 (9.41%)
9–12	5,164 (36.27%)	1,553 (29.90%)	1,664 (38.06%)	1,947 (43.39%)
> 12	6,353 (58.22%)	2,560 (67.21%)	2,052 (56.51%)	1,741 (47.20%)
Family income, n(%)					< 0.001
Low	3,770 (18.91%)	1,083 (16.13%)	1,158 (18.07%)	1,529 (23.92%)
Medium	4,978 (36.03%)	1,550 (33.03%)	1,610 (36.54%)	1,818 (39.78%)
High	4,174 (45.06%)	1,694 (50.83%)	1,369 (45.40%)	1,111 (36.30%)
Physical activity, n(%)					< 0.001
Inactive	6,316 (45.68%)	1,706 (37.71%)	1,991 (46.48%)	2,619 (56.27%)
Active	6,606 (54.32%)	2,621 (62.29%)	2,146 (53.52%)	1,839 (43.73%)
Alcohol intake, n(%)					< 0.001
No	3,623 (23.64%)	938 (18.02%)	1,129 (23.94%)	1,556 (31.43%)
Yes	9,299 (76.36%)	3,389 (81.98%)	3,008 (76.06%)	2,902 (68.57%)
Smoking status, n(%)					< 0.001
Never	6,772 (52.77%)	2,375 (56.23%)	2,138 (51.15%)	2,259 (49.66%)
Current	3,292 (24.94%)	791 (19.36%)	1,101 (27.16%)	1,400 (30.39%)
Former	2,858 (22.29%)	1,161 (24.41%)	898 (21.68%)	799 (19.95%)
Coronary heart disease, n(%)					< 0.001
No	12,418 (96.88%)	4,279 (99.19%)	3,984 (96.76%)	4,155 (93.68%)
Yes	504 (3.12%)	48 (0.81%)	153 (3.24%)	303 (6.32%)
Diabetes,n(%)					< 0.001
No	11,519 (92.26%)	4,198 (97.97%)	3,769 (93.51%)	3,552 (82.54%)
Yes	1,403 (7.74%)	129 (2.03%)	368 (6.49%)	906 (17.46%)
Hypertension, n(%)					< 0.001
No	9,286 (75.15%)	3,737 (88.10%)	3,025 (74.62%)	2,524 (57.02%)
Yes	3,636 (24.85%)	590 (11.90%)	1,112 (25.38%)	1,934 (42.98%)
Stroke, n(%)					< 0.001
No	12,661 (97.19%)	4,339 (99.02%)	4,070 (97.68%)	4,252 (93.97%)
Yes	487 (2.81%)	59 (0.98%)	134 (2.32%)	294(6.03%)
Congestive heart failure, n(%)					< 0.001
No	12,557 (98.02%)	4,287 (99.48%)	4,027 (98.12%)	4,243 (95.80%)
Yes	365 (1.98%)	40 (0.52%)	110 (1.88%)	215 (4.20%)
Cancer, n(%)					< 0.001
No	11,693 (90.91%)	4,100 (94.71%)	3,765 (90.57%)	3,828 (85.80%)
Yes	1,229 (9.09%)	227 (5.29%)	372 (9.43%)	630 (14.20%)
Calorie consumption (kcal/d),
Mean (SE)	2,207.99 (15.42)	2,389.65 (22.89)	2,205.92 (24.89)	1,947.46 (21.72)	< 0.001
Protein consumption(g/d),
Mean(SE)	84.67 (0.64)	91.14 (0.83)	84.30 (1.02)	75.73 (0.92)	< 0.001
Carbohydrate consumption(g/d),
Mean(SE)	263.72 (1.89)	286.31 (2.83)	261.34 (3.16)	233.82 (2.38)	< 0.001
Fat consumption(g/d), Mean(SE)	83.97 (0.77)	88.87 (1.13)	84.77 (1.09)	75.94 (1.25)	< 0.001
FI, n (%)					< 0.001
No	11,770 (91.82%)	4,095 (95.18%)	3,808 (92.10%)	3,867 (86.61%)
Yes	1,152 (8.18%)	232 (4.82%)	329 (7.90%)	591 (13.39%)
Body mass index(kg/m2), n(%)					< 0.001
normal weight	3,718 (31.43%)	2,262 (53.81%)	953 (21.37%)	503 (10.85%)
over weight	4,448 (33.77%)	1,426 (32.68%)	1,662 (40.10%)	1,360 (27.94%)
obese	4,756 (34.79%)	639 (13.51%)	1,522 (38.53%)	2,595 (61.21%)
Waist circumference(cm),Mean(SE)	98.13 (0.31)	87.02 (0.26)	100.25 (0.21)	111.74 (0.34)	< 0.001

WWI, weight-adjusted-waist index; FI, fecal incontinence; SE, standard error. data presented as unweighted numbers (weighted percentage) for categorical variables and mean (standard error) for continuous variables

Participants in the highest tertiles of WWI were older (mean [SE] = 55.59 [0.44]), with a higher percentage of female patients (male 38.26% vs. female 61.74%). They tended to have more self-reported medical comorbidities, be more likely to smoke, have a higher BMI, have lower family income, and have less physical activity and lower education levels.

The association between WWI and FI

Table 2 displays the findings of the weighted multivariable logistic regression analyses investigating the association between WWI and FI. A higher WWI was found to be linked to an elevated prevalence of FI, with an odds ratio of 1.36 (95% CI:1.20–1.55, P < 0.001), even after adjusting for potential confounders. When WWI was categorized into tertiles, the positive correlation remained significant in the highest WWI group compared to the lowest (adjusted OR = 1.63, 95% CI:1.31–2.02, P < 0.001) (Table 2). The smooth curve fitting analysis unveiled a linear relationship between WWI and FI (Fig. 2).

Table 2.

Weighted multivariate regression analysis of the associations between WWI and the risk of FI

Variable	Model 1		Model 2		Model 3
	OR (95%CI)	P value	OR (95%CI)	P value	OR (95%CI)	P value
WWI(continu-ous)	1.84(1.72–1.97)	< 0.001	1.47(1.33–1.62)	< 0.001	1.36(1.20–1.55)	< 0.001
WWI(tertile)
T1	1(Ref)		1(Ref)		1(Ref)
T2	1.72 (1.41–2.09)	< 0.001	1.35(1.09–1.66)	0.007	1.26(1.00–1.58)	0.046
T3	3.07(2.65–3.55)	< 0.001	1.93(1.61–2.33)	< 0.001	1.63 (1.31–2.02)	< 0.001
Trend test		< 0.001		< 0.001		< 0.001

WWI, weight-adjusted-waist index; FI, fecal incontinence

Model 1 was adjusted for none

Model 2 was adjusted for age, sex, race/ethnicity

Model 3 was adjusted for age, sex, marital status, race/ethnicity, education level, family income, smoking status, alcohol intake, physical activity, hypertension, diabetes, coronary heart disease, congestive heart failure, cancer, stroke, calorie consumption, protein consumption, carbohydrate consumption, fat consumption, and body mass index

Fig. 2.

Association between weight-adjusted-waist index and fecal incontinence. Note: WWI, weight-adjusted-waist index; FI, fecal incontinence; OR, odds ratio; CI, confidence interval. They were adjusted for age,sex,marital status, race/ethnicity, education level, family income, smoking status,alcohol intake, physical activity, hypertension, diabetes, coronary heart disease, congestive heart failure, cancer, stroke, calorie consumption, protein consumption, carbohydrate consumption, fat consumption, and body mass index. Only 99% of the data is shown

The ROC curve and DeLong’s test were utilized to assess and compare the AUC values for their predictive accuracy of FI among different obesity markers. Our study showed that WWI had a better predictive ability for FI than BMI and WC (AUC: 0.542 for BMI and 0.572 for WC, all P < 0.001), but insufficient for standalone diagnosis(Fig. 3). WWI demonstrated higher sensitivity, suggesting its potential utility in high-sensitivity screening scenarios (Supplementary Table S1).

Fig. 3.

ROC curve for FI. Note: WWI, weight-adjusted-waist index; FI, fecal incontinence; WC, waist circumference; BMI, body mass index; ROC, Receiver operating characteristic. ROC curve analysis and DeLong’test for comparison of the predictive power between WWI and BMI, WC for FI

Subgroup analysis

Subgroup analysis was employed to evaluate the stability of the association between WWI and FI across various demographic backgrounds. Subgroup analysis stratified by age, sex, BMI, and comorbidities revealed no significant interactions (all P > 0.05). This indicates that the association between WWI and FI is stable across these different populations (Fig. 4). Notably, when using 65 years as the age threshold, the interaction term reached statistical significance (P = 0.001; Supplementary Figure S1).

Fig. 4.

Subgroups analyses of the effect of WWI on FI. Note: WWI, weight-adjusted-waist index;FI, fecal incontinence; OR, odds ratio; CI, confidence interval. Except for the stratification component itself, each stratification factor was adjusted for all othervariables (age, sex, marital status, race/ethnicity, education level, family income, smoking status, alcohol intake, physical activity, hypertension, diabetes, coronary heart disease, congestive heart failure, cancer, stroke, calorie consumption, protein consumption, carbohydrate consumption, fat consumption, and body mass index.)

Sensitivity analysis

To account for the impact of advanced age, a sensitivity analysis was conducted by excluding participants aged 75 and older from the analysis, which confirmed the consistency of our findings (OR = 1.45, 95% CI:1.24–1.70; P < 0.001) (Supplementary Table S2). This indicates that the association between WWI and FI is stable across different age group Excluding incomplete sampling weights data (539), we conducted multiple imputations to address the missing covariates for the remaining 16,539 participants. Compared to the lowest tertile of WWI, the highest tertile showed a consistent positive association with FI (OR = 1.54, 95% CI:1.21–1.88; P < 0.001) (Supplementary Table S3).

Discussion

This study aimed to explore the correlation between the WWI and FI in American adults. Our study has identified linear relationships based on the 2005–2010 NHANES cross-sectional study involving 12,922 participants. WWI was independently associated with a 36% increase in the prevalence of FI. Our findings did not reveal any interactive effects between WWI and FI, indicating the stability of the conclusions across different subgroups and sensitivity analyses.

A variety of indicators have been employed to evaluate obesity, with a specific emphasis on the deleterious effects of abdominal visceral fat mass. Previous studies have endeavored to evaluate the association between obesity metrics, such as BMI and WC, and FI, yielding inconsistent and varied findings [8–10]. As a novel obesity metric, WWI isolates centripetal obesity independent of body weight and distinguishes fat/muscle components [11]. According to previous studies, WWI outperforms traditional indices (BMI, WC, waist-hip ratio,etc.) in predicting cardiovascular risk and non-accidental mortality, primarily because BMI ignores fat distribution while WC neglects body size variations [8, 12]. The integration of fat and muscle mass in WWI enhances its comprehensiveness and accuracy in evaluating obesity and associated health risks [12]. Crucially, the practicality of WWI lies in its reliance on simple, cost-effective anthropometric measurements (WC and weight), unlike complex imaging-based adiposity measures. This makes it feasible for routine clinical use, especially in resource-limited settings.

FI was characterized by the unintentional loss of mucus, liquid, or solid stool [29]. Diarrhea, anorectal sensorimotor dysfunctions including anal sphincter and/or levator ani muscle and/or puborectalis weakness, reduced rectal compliance, decreased or increased rectal sensation, and pelvic floor anatomical disturbances were the primary risk factors for FI [30].

Mechanistically, WWI has been implicated in the etiology of FI through diverse pathways. Specifically, central obesity as measured by WWI has been correlated with an elevated risk of diarrhea, a recognized risk factor for FI. Recent studies have demonstrated that obese individuals were more likely to experience bile acid malabsorption [30], accelerated colonic transit [31], increased mucosal permeability (linked to intestinal dysbiosis and elevated plasma endotoxins) [32, 33], and intestinal inflammation (characterized by altered interleukin-10 secretion) [33].

Obesity has been shown to impact bowel continence mechanisms by elevating intra-abdominal pressure [6, 29]. WWI may contribute to FI through this mechanism, as it is associated with an increased deposition of body mass, potentially leading to higher intra-abdominal pressure [34]. As the intra-abdominal pressure rose, the anal sphincter weakened or deferred recruitment because of decreased sensory impulses from the pelvic floor [5]. In a representative Swedish cohort comprising 1001 individuals from the general population. Obesity has been significantly linked to an increased sense of urgency in stools [35].

Additionally, WWI-indexed obesity may lead to pelvic floor laxity, muscular atrophy, and changes in rectal sensitivity through longstanding pro-inflammatory states [36–38]. Obesity significantly contributes to oxidative stress due to the dysregulation of a variety of adipocytokines [39]. Elevated levels of oxidative stress have been demonstrated to potentially decrease the tone of the internal anal sphincter [40, 41].

Therefore, unlike BMI (which conflates muscle/fat) or WC (unadjusted for body size), WWI distinctly isolates visceral fat burden, which is a key driver of intra-abdominal pressure, pelvic floor dysfunction, and diarrhea. This positions WWI as a superior indicator of obesity-related FI pathophysiology.

However, several limitations should be acknowledged. First, the cross-sectional design precludes causal inference. Second, While subgroup analyses demonstrated robustness, we cannot exclude the potential influence of residual measured or unmeasured confounding on the study findings, such as pelvic floor disorders. Further mechanistic studies are needed to explore potential factors. Third, the modified WWI-FI relationship in older adults may reflect age-related body composition shifts, such as sarcopenia or neurodegenerative processes dominating FI pathogenesis. Future studies should integrate body composition assessments or bioimpedance analysis. Fourth, FI was assessed via self-reported questionnaire. While practical for large studies, this method is prone to recall bias and may miss cases compared to physician diagnosis or objective testing. Fifth, the measurement of WWI may vary due to differences in techniques or equipment, which could affect its reliability and consistency. Sixth, although the AUC for WWI (0.627) indicates only moderate discriminatory ability, it was statistically superior to both BMI and WC. This reinforces WWI’s relative advantage in capturing obesity-related pathophysiology relevant to FI. However, the modest AUC underscores that FI is multifactorial; this suggests that obesity metrics alone may not fully capture the etiology of FI, which involves neuromuscular, Pelvic floor anatomical, and behavioral factors. Future prospective or interventional studies should combine WWI with other biomarkers or functional tests to boost predictive accuracy.

Despite these limitations, we used data from a countrywide sample and account for sample weights, our findings were more typical of the general population in the US. We built multivariable logistic regression models with subgroup and sensitivity analysis to compensate for potential confounding factors.

Conclusions

Analysis of NHANES data reveals a significant relationship between WWI and the incidence of FI. WWI’s operational simplicity and focus on central adiposity support its use as a cost-effective screening adjunct in primary care, not a diagnostic tool. Future prospective or interventional studies should explore the underlying mechanisms and validate these findings in diverse populations.

Abbreviations

FI

Fecal incontinence

WWI

Weight-adjusted-waist Index

WHR

Waist-to-hip ratio

BMI

Body mass index

WC

Waist circumference

PIR

Poverty income ratio

ROC

Receiver operating characteristic

AUC

Area under curve

T

Tertile

OR

Odds ratio

CI

Confidence interval

Ref

Reference

NHANES

National health and nutrition examination survey

CDC

Centers for disease control and prevention

MECs

Mobile examination centers

FISI

Fecal incontinence severity index

BHQ

Bowel health questionnaire

Author contributions

Ying Zhang and Wenting Hu wrote the main manuscript text and Zhilian Zhou, Xiuming Wang, Lin chanchan prepared figures and Tables. All authors reviewed the manuscript.

Funding

This article was supported by the Sichuan Provincial Administration of Traditional Chinese Medicine Research Project Fund (2024MS589).

Availability of data and materials

All data used in this study is available in NHANES database at http://www.cdc.gov/ nchs/nhanes.htm.

Declarations

Ethics approval and consent to participate

This research analyzed de-identified information downloaded from the National Health and Nutrition Examination Survey public database. The National Center for Health Statistics Ethics Review Committee granted ethics approval. All methods were carried out in accordance with relevant guidelines and regulations (declaration of Helsinki). All individuals provided written informed consent before participating in the study.

Consent for publication

Not applicable.

Informed consent statement

All individuals provided written consent.

Conflict of interest

The authors declare no competing interests.