The NHANES was a significant national survey project in the United States, conducted by the National Center for Health Statistics (NCHS) under the Centers for Disease Control and Prevention (CDC). This initiative focused on collecting and analyzing health and nutrition information from individuals across the country, encompassing various health domains, such as chronic diseases, mental health, environmental exposures, and biomarkers. To ensure national representation, NHANES employed a stratified, multistage probability sampling technique. The survey was conducted every 2 years and included participants of varying ages, genders, races, and geographic regions. The sample design accounted for the diversity within the United States, aiming to accurately reflect the health status of different demographic groups. The data collected by NHANES were widely utilized across multiple fields, including the formulation of public health policies, the implementation of health education programs, and scientific research. Researchers and public health officials leveraged this data to identify health issues, assess the effectiveness of health interventions, and develop targeted health policies for specific populations. Furthermore, every participant provided a signed informed consent form, and the study received approval from the National Center for Health Statistics Research Ethics Review Board. This study was an observational study that followed the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines to ensure the transparency and consistency of the study design and reporting. This research utilized data obtained from NHANES during the periods of 2001–2006 and 2011–2018. Participants had to be women aged 20 years and older with complete datasets. Additionally, ASM was obtained through dual-energy X-ray absorptiometry (DXA) scans, which have their own exclusion criteria. These criteria include being pregnant, reporting the use of radiographic contrast agents (such as barium) in the past week, and individuals self-reporting a weight greater than 450 pounds or a height over 6 feet 5 inches (limitations of the DXA equipment). Considering these inclusion and exclusion criteria, 19,009 participants were ultimately selected for this study (Figure 1).

ASMI was the exposure variable in this study. The calculation formula for ASMI was ASM (kg) divided by BMI (kg/m²). The total muscle mass of the arms and legs constitutes ASM. BMI values were directly obtained from the Body Measures section in the Examination Data of NHANES.

In this study, UUI served as the primary outcome variable, defined based on the following questionnaire: “Over the last 12 months, have you had any incidents of leaking urine or losing control, feeling a sudden urge to urinate and finding it hard to get to the bathroom quickly?” Individuals who responded “yes” were classified as having UUI, while those who responded “no” were classified as not having UUI.

Drawing from prior research, this study extracted various covariates from the NHANES database. Age was classified into three categories: 20–39 years, 40–59 years, and 60 years and older. Education was segmented into three tiers: individuals with education below high school, those who graduated high school, and those who pursued college or higher education. Race was classified into the following groups: Mexican American, Non-Hispanic Black, Non-Hispanic White, Other Hispanic, and Other Race. The family income to poverty ratio (PIR) was categorized as less than 1.3, between 1.3 and 3.4, and 3.5 or higher. Smoking status was classified into three categories: never smoked (fewer than 100 cigarettes in their lifetime and not currently smoking), former smokers (at least 100 cigarettes in their lifetime but had quit), and current smokers (at least 100 cigarettes in their lifetime and still smoking). Marital status is divided into the following groups: divorced, living with partner, married, never married, separated, and widowed. Diabetes or hypertension was considered present if diagnosed by a healthcare professional or if relevant medications were being used. Vigorous activity was defined as engaging in at least 10 min of intense activity that resulted in significant sweating or a substantial increase in breathing or heart rate. Moderate activity was defined as engaging in at least 10 min of moderate activity that resulted in only slight sweating or a slight to moderate increase in breathing or heart rate. If the participants engaged in vigorous or moderate activity, they answered yes; otherwise, they answered no. The assessment of stroke, congestive heart failure, and hysterectomy, relied on self-report in the Questionnaire, with the question “Has a doctor or other health professional ever informed you of having [respective condition]?” Previous pregnancies were divided into “Yes” and “No” groups. The study accounted for variables including urea nitrogen in the blood, uric acid, and creatinine in serum.

In this study, all participants were divided into two groups: those with UUI and those without UUI. Continuous variables were represented by mean ± standard deviation (SD), while categorical variables were expressed as proportions (%) and counts (n). Weighted Student’s t-tests were used to analyze the differences in continuous variables between the two groups, while the association between categorical variables and the presence of urgency urinary incontinence was assessed using weighted chi-square tests. In this study, we employed two primary effect size indicators, Cohen’s d and Cramer’s V, to evaluate the relationships between different variables and the magnitude of group differences. Cohen’s d was used to measure the standardized effect size of the mean difference between two groups. Based on previous medical literature (22, 23), the effect sizes were categorized as: small (0.2 ≤ d < 0.5), medium (0.5 ≤ d < 0.8), and large (d ≥ 0.8). Cramer’s V was employed to assess the strength of association between categorical variables. In accordance with prior medical studies (24, 25), the association strengths were classified as: very strong (>0.25), strong (>0.15), moderate (>0.10), weak (>0.05), and very weak or no association (≤0.05). ASMI was analyzed not only as a continuous variable but also categorized into weighted quartiles. Multiple logistic regression models were used to estimate the relationship between ASMI and UUI, expressed as odds ratios (OR) and corresponding 95% confidence intervals (CI). Model 1 did not include any covariates, Model 2 incorporated covariates including age, race, PIR, marital status, citizenship, and education attainment. Model 3, building on Model 2, added smoking status, hysterectomy, stroke, vigorous activities, diabetes, hypertension, previous pregnancies, urea nitrogen in the blood, moderate activities, congestive heart failure, uric acid, and creatinine in serum. Restricted cubic splines (RCS) were a statistical method used to model nonlinear relationships by flexibly fitting data with different polynomial segments at multiple knots, while maintaining the smoothness and interpretability of the model. This method effectively captured complex nonlinear relationships between variables without easily leading to overfitting. Therefore, in this study, a restricted cubic spline (RCS) curve was employed to illustrate the relationship between ASMI and UUI, based on Model 3, which included all covariates. Subgroup analyses were conducted based on age, educational attainment, PIR, smoking status, vigorous activities, congestive heart failure, and hysterectomy to assess whether these factors influenced the relationship between ASMI and the risk of UUI. All analyses were performed using R software (version 4.2.3). Differences were considered statistically significant when the p-value was less than 0.05.

As shown in the flowchart (Figure 1), a total of 19,009 female adult participants aged 20 and above were enrolled in the final analysis of this study, including 13,049 individuals without UUI and 5,960 individuals with UUI. Table 1 presented the baseline characteristics of the two groups. The study found that, compared to the group without UUI, the group with UUI had a higher proportion of individuals over 60 years old, those who did not engage in vigorous or moderate activities, individuals with a history of stroke, those who had undergone a hysterectomy, as well as those with hypertension and diabetes, citizens, and those with previous pregnancies. In contrast, the proportions of individuals with a college education or higher, those with a PIR of 3.5 or greater, never smokers, and those who were never married were lower. Additionally, the UUI group had higher levels of appendicular skeletal muscle mass, but lower levels of ASMI. Compared to the group without UUI, the group with UUI exhibited higher appendicular skeletal muscle mass but a lower ASMI. The difference in ASMI between the two groups was statistically significant (p < 0.05), with a Cohen’s d of 0.29.

To verify the relationship between ASMI and UUI, we conducted logistic regression analysis using three models (Table 2). In Model 1, which did not adjust for any covariates, the risk of UUI decreased by 95% for each unit increase in ASMI (OR = 0.05; 95% CI: 0.02–0.11; p < 0.0001). A comparable correlation was noted in Model 2, which accounted for covariates including age, race, PIR, marital status, citizenship, and education attainment. In this model, the risk of UUI decreased by 81% for each unit increase in ASMI (OR = 0.19; 95% CI: 0.07–0.50; p = 0.001). Furthermore, Model 3, which includes all covariates, still showed a significant association between ASMI and the risk of UUI. As per the findings of Model 3, an increase of one unit in ASMI was linked to a 69% reduction in the likelihood of UUI (OR = 0.31; 95% CI: 0.12–0.82; p = 0.02). Relative to the first quartile of ASMI, being in the fourth quartile reduced the risk of UUI by 28% (OR = 0.72; 95% CI: 0.55–0.94; p = 0.02). RCS analysis revealed a nonlinear relationship between ASMI and the risk of UUI, showing a decreasing trend in UUI risk as ASMI increases (Figure 2; p for overall <0.001, p for nonlinearity = 0.02).

Based on stratified analysis considering age, PIR, marital status, citizenship, education attainment, smoking status, race, hysterectomy, stroke, vigorous activities, congestive heart failure, diabetes, hypertension, and previous pregnancies (Table 3). This indicates that identified baseline traits had no effect on altering this association.

As shown in Table 4, our study sample demonstrated a trend of decreasing ASMI with increasing age. According to the FNIH criteria for low ASMI (26), the prevalence of low ASMI in women aged 20–29 was 2.94%, while the prevalence of UUI was 13.28%. In contrast, among women aged 70 and older, the prevalence of low ASMI reached 24.80%, and the prevalence of UUI reached 44.70%. Both the prevalence of low ASMI and UUI exhibited an increasing trend with advancing age.