The National Health and Nutrition Examination Survey (NHANES) and the National Center for Health Statistics (NCHS) played essential roles in collecting data for this investigation. We acquired data from three consecutive 2-year NHANES cycles conducted between 1999 and 2006, utilizing a nationally representative stratified sample through interviews and physical examinations. The NCHS Ethics Review Committee granted ethical approval, and all subjects provided signed informed consent.

The original dataset included 5,557 female participants. After excluding individuals with missing endometriosis-related information and those outside the age range of 20 to 54 years, our final study population comprised 4,453 women. Among them, 337 were diagnosed with endometriosis, while 4,116 were not. Exclusions were made for 991 pregnant women and 113 individuals lacking information on dietary vitamin B6 intake. Figure 1 provides a detailed illustration of the inclusion and exclusion process.

We determined endometriosis based on participants’ responses to a specific question in the reproductive health questionnaire: “Has a doctor or other health professional ever diagnosed you with endometriosis?” We categorized those who answered affirmatively as patients. Professional interviewers assessed dietary vitamin B6 consumption during the NHANES dietary survey, a component of the “What We Eat in America” survey, conducted at the Mobile Examination Center (MEC) using a 24-h recall method. The NHANES computer-assisted dietary interview (CADI) system recorded participants’ food and beverage intake from the 24 h preceding the interview.

According to the study procedure, we randomly assigned participants to data collection sessions that occurred in the morning, afternoon, or evening. We determined dietary vitamin B6 and nutrient consumption using the US Department of Agriculture Survey Nutrients Database and the University of Texas Food Intake Analysis System. We excluded pharmaceuticals and dietary supplements from the nutritional calculations. We conducted two 24-h dietary recall interviews, followed by a phone interview 3 to 10 days later. We selected the first interview, conducted in person at the MEC, for analysis, as the 24-h recall method is the most commonly used in large-scale surveys (22).

In our study, we considered a wide range of covariates sourced from the literature (23–25), including Age, marital status, race/ethnicity, education level, family income, smoking status, physical activity, BMI, alcohol consumption, use of birth control pills, high blood pressure, diabetes, coronary heart disease, chronic bronchitis, caloric consumption, total fat intake, total cholesterol intake, and usage of nutritional supplements. We categorized race and ethnicity into Non-Hispanic White, non-Hispanic Black, Mexican American, and other races. We classified marital status as either living with a partner or living alone. We stratified educational attainment into three levels: fewer than 9, 9 to 12 years, and more than 12 years. We assessed family income using the Poverty Income Ratio (PIR) and categorized it into low, medium, and high income, based on ranges from 1.3 to 3.5, according to the US government’s Agriculture report (26). We dichotomized smoking status into smokers and never smokers (those who have smoked fewer than 100 cigarettes in total). Alcohol drinking status was determined by the survey question, “In any 1 year, have you had at least 12 drinks of any type of alcoholic beverage?” Participants who answered “yes” were defined as alcohol drinkers. We divided physical activity levels into three categories: unable to perform physical activity, moderate (defined as at least 10 min of movement within the previous 30 days resulting in light perspiration or a mild to moderate increase in respiration or heart rate), and vigorous (at least 10 min of activity within the last 30 days resulting in profuse sweating or an increased heart rate). The determination of previous disease (high blood pressure, diabetes, coronary heart disease and chronic bronchitis) was based on the inquiry in the questionnaire of whether the doctor had been informed of the condition in the past.

Prior to the MEC interview, participants completed a food recall questionnaire in order to get 24-h nutritional data, including macronutrient profiles and calorie intake. Additionally, details regarding medications, including dietary supplements consumed within the preceding month, were documented.

In this study, we conducted a secondary analysis of publicly available datasets and used descriptive statistics to characterize continuous variables (mean/SD or median/IQR) and proportions (%) for categorical variables. We assessed group differences using Kruskal-Wallis tests and one-way analyses of variance. We employed logistic regression models to explore the relationship between dietary vitamin B6 intake and endometriosis across three models. Model 1 adjusted for sociodemographic (age, race/ethnicity, education level, family income and marital status); Model 2 was adjusted for Model 1 plus BMI, smoking status, vigorous activity, moderate activity, alcohol consumption, birth control pills taken, high blood pressure, diabetes, coronary heart disease and chronic bronchitis; and Model 3 was adjusted for Model 2 plus calorie consumption, total fat consumption, total cholesterol consumption and dietary supplements taken. These models aimed to comprehensively address potential confounding factors and to enhance our understanding of the relationship between dietary vitamin B6 intake and endometriosis.

Furthermore, we investigated potential modifiers of the association between dietary vitamin B6 intake and endometriosis by incorporating variables such as family income (low vs. medium or high), marital status (living with a partner vs. living alone), smoking status, and dietary supplements taken. We assessed heterogeneity among subgroups using multivariate logistic regression and examined interactions between subgroups and dietary vitamin B6 intake through likelihood ratio testing. To ensure the robustness of our findings, we conducted sensitivity analyses by excluding participants with extreme energy intake, defined as consuming less than 500 or more than 5,000 kcal per day. This meticulous approach aimed to assess the consistency and reliability of our results under different conditions, thereby enhancing the validity of the study outcomes.

We determined the sample size based on available data, without conducting a priori statistical power assessments. We performed statistical analyses using R 3.3.2 and Free Statistics Software 1.5 (27). We conducted a comprehensive descriptive study on all individuals. For hypothesis testing, we utilized a two-tailed analysis, considering a significance level of 0.05 for determining statistical significance. This widely accepted threshold ensures standard confidence levels for interpreting results and drawing meaningful conclusions from the analyses.

Table 1 presents the essential characteristics of the 4,453 participants in the study. Among the sample, 337 individuals (7. 57%) received a diagnosis of endometriosis, and 2,820 (63.92%) were classified as overweight. A total of 1741 participants (39.10%) reported being smokers, while 2099 (47.14%) acknowledged the use of dietary supplements. Regarding age distribution, 2,378 participants (53.40%) were below 40 years old, 1,424 (31.98%) were between 40 and 50 years old, and 651 (14.62%) were above 50 years old. Notably, dietary vitamin B6 intake appeared higher in individuals who used dietary supplements, engaged in moderate exercise, were aged under 40, had a BMI below 25, had an education level exceeding 12 years, did not smoke, and lived with a partner or had a higher income. These baseline characteristics provide an overview of the diversity within the study population and establish the foundation for subsequent analyses exploring the association between these factors and endometriosis.

Table 2 illustrates the relationships between dietary vitamin B6 intake and endometriosis. Univariate analysis revealed significant associations, indicating correlations between dietary vitamin B6 intake, age, race, family income, education level, use of birth control pills, high blood pressure, chronic bronchitis, and the use of dietary supplements with the presence of endometriosis. These findings underscore a complex interplay between dietary factors, demographic variables, and lifestyle choices influencing the occurrence of endometriosis. Subsequent multivariate analyses will further dissect these associations to elucidate the independent contributions of each factor to the risk of endometriosis.

Compared to individuals with lower vitamin B6 consumption (Q1 < 0.94 mg/day), the adjusted odds ratio (OR) values for dietary vitamin B6 intake and endometriosis in Q2 (0.95–1.39 mg/day), Q3 (1.40–1.99 mg/day), and Q4 (>1.99 mg/day) were 1.22 (95% CI: 0.88–1.69, p = 0.24), 1.22 (95% CI: 0.86–1.73, p = 0. 279), and 1.51 (95% CI: 1.01–2. 24, p = 0.04), respectively (Table 3).

In a thorough examination of various subgroups, stratified analyses were conducted to assess potential effect modifications on the relationship between dietary vitamin B6 intake and endometriosis (refer to Figure 2). Notably, no significant interactions were identified in any subgroups, whether stratified by age, BMI, marital status, family income, smoking status, education level, or dietary supplements taken. These results suggest that the observed association between dietary vitamin B6 intake and endometriosis remains consistent across diverse demographic and lifestyle factors, reinforcing the robustness of the findings.

Furthermore, after excluding individuals with extreme energy intake, the dataset comprised 4,375 individuals, and the association between dietary vitamin B6 intake and endometriosis remained robust. Compared to individuals with lower vitamin B6 consumption (<0.96 mg/day), and after adjustments for age, race, education level, smoking status, dietary supplements taken, vigorous activity, moderate activity, marital status, family income, and BMI, the adjusted odds ratio (OR) values for dietary vitamin B6 intake and endometriosis in Q2 (0.96–1.39 mg/day), Q3 (1.40–2.05 mg/day), and Q4 (>2.05 mg/day) were 1.26 (95% CI: 0.91–1.73, p = 0.16), 1.27 (95% CI: 0.92–1.75, p = 0.15), and 1.61 (95% CI: 1.14–2.26, p = 0.01), respectively (Supplementary Table S1).