Methods
The National Health and Nutrition Examination Survey (NHANES) is a nationally representative cross-sectional survey conducted through home interviews and mobile examination centers, aimed at assessing the health and nutritional status of the U.S. population. This survey utilized data from 29,400 participants over three cycles of NHANES, spanning from 2013 to 2018. After excluding individuals 45 years old ( n = 21,186), males ( n = 3,891), and participants with missing or incomplete BRI and infertility data ( n = 795), a total of 3,528 participants were included in the final analysis. Figure 1 displays a flowchart of the entire selection process. NHANES is approved by the Research Ethics Review Board of the National Center for Health Statistics, and all participants provided informed consent [ 16 ]. The data used in this study are de-identified and publicly available ( https://www.cdc.gov/nchs/nhanes/index.htm ).
Fig. 1 A flow diagram of eligible participant selection in the National Health and Nutrition
A flow diagram of eligible participant selection in the National Health and Nutrition
BRI is a novel body shape assessment index, calculated using participants’ height (cm) and waist circumference (cm) [ 17 ]. The specific calculation formula is as follows: \documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$\:BRI=364.2-365.5\times\:\sqrt{1-{\frac{\left(\frac{waist\:circumference}{2\pi\:}\right)}{{(0.5\times\:height)}^{2}}}^{2}}$$\end{document} .
According to previous research [ 18 , 19 ], infertility is defined as a reduction in an individual or their partner’s ability to conceive, characterized by the inability to become pregnant after one year or more of regular, unprotected intercourse. In this study, the assessment of infertility is derived from the NHANES Reproductive Health Questionnaire (RHQ074). Specifically, participants were asked the following question: “Have you ever tried to get pregnant for a year or longer without becoming pregnant?” Participants who answered “yes” were classified as infertile ( https://wwwn.cdc.gov/Nchs/Nhanes/2013-2014/RHQ_H.htm#RHQ074 ).
According to previous studies [ 18 , 19 ], the covariates in this research include age, race, marital status, education level, family poverty-to-income ratio (PIR), smoking, alcohol consumption, hypertension, diabetes, and hypercholesterolemia. For detailed information on these covariates, please refer to Table S1 .
In this study, all data were statistically analyzed using R (version 4.3.1). The data were weighted, with continuous variables presented as mean ± standard deviation, and p -values calculated using weighted linear regression models. Percentages for categorical variables (weighted N, %) and their p -values were calculated using weighted chi-square tests. The association between BRI and infertility was analyzed using multivariable logistic regression models, where BRI was categorized into quartiles. Trend tests and p -values for linear trends were calculated to determine the consistency of the relationship. Three models were constructed in this study: (1) an unadjusted crude model; (2) a model adjusted for age, race, education level, marital status, and family poverty-to-income ratio (PIR); and (3) a model further adjusted for smoking, alcohol consumption, hypertension, diabetes, and hypercholesterolemia. A smooth curve fitting was applied to further explore the potential linear relationship between BRI and infertility. Additionally, odds ratios (ORs) were calculated for every 1-unit increase in BRI, with subgroup analyses conducted based on age, race, marital status, education level, PIR, smoking, alcohol consumption, hypertension, diabetes, and hypercholesterolemia. Multiple imputations by chained equations (MICE) and repeated the main analyses. We used multiple imputations, based on 5 imputed data sets to account for missing baseline data [ 20 ]. Finally, to reduce selection bias and balance the distribution of covariates between the non-infertility and infertility groups, propensity score matching (PSM) was performed in a 1:1 ratio with a caliper width of 0.01 times the standard deviation of the logit of the propensity score. The significance was determined by p -values below 0.05.
Results
This study included 3,528 women aged 18 to 45, representing approximately 51,123,046 women of reproductive age in the United States. The prevalence of infertility was 11% (equivalent to 5,793,958 women), with a mean (SD) BRI value of 6.43 (3.08). Table 1 shows that women with infertility had a higher BRI compared to those without infertility (non-infertile: 5.24 (2.65), Infertile: 6.43 (3.08)). Significant differences were found between the infertile and non-infertile groups regarding age, cohabitation with a partner, smoking, hypertension, diabetes, and hyperlipidemia (all p < 0.05). The baseline after PSM is shown in Table S3 .
Table 1 Baseline characteristics of all participants were stratified by infertility, weighted Characteristic Overall, N = 51,123,046 (100%) Non-infertility, N = 45,329,088 (89%) Infertility, N = 5,793,958 (11%) P Value
No. of participants in the sample
3,528 3,164 364
-
Age (%)
34 19,952,775 (39%) 16,621,196 (37%) 3,331,579 (58%)
Race (%)
0.100 Non-Hispanic White 28,635,786 (56%) 25,017,130 (55%) 3,618,655 (62%) Other 9,510,598 (19%) 8,658,404 (19%) 852,193 (15%) Non-Hispanic Black 6,845,835 (13%) 6,139,901 (14%) 705,934 (12%) Mexican American 6,130,828 (12%) 5,513,652 (12%) 617,176 (11%)
Married/live with partner (%)
< 0.001
No 19,229,843 (40%) 17,919,894 (43%) 1,309,948 (23%) Yes 28,656,737 (60%) 24,205,957 (57%) 4,450,780 (77%)
Education level (%)
0.588 Below high school 5,493,446 (11%) 4,875,470 (12%) 617,976 (11%) High School or above 42,393,134 (89%) 37,250,382 (88%) 5,142,752 (89%)
PIR (%)
0.108 Not Poor 33,606,384 (70%) 29,431,242 (70%) 4,175,142 (75%) poor 14,072,815 (30%) 12,671,795 (30%) 1,401,020 (25%)
Smoking (%)
0.012
Never 35,406,036 (69%) 31,857,251 (70%) 3,548,786 (61%) Former 6,069,938 (12%) 5,201,022 (11%) 868,916 (15%) Current 9,647,072 (19%) 8,270,815 (18%) 1,376,256 (24%)
Drinking (%)
0.061 former 2,601,289 (5.3%) 2,094,740 (4.8%) 506,548 (9.0%) heavy 13,105,738 (26%) 11,561,989 (26%) 1,543,749 (28%) mild 12,825,776 (26%) 11,359,023 (26%) 1,466,753 (26%) moderate 13,559,212 (27%) 12,079,207 (28%) 1,480,005 (26%) never 7,429,995 (15%) 6,817,834 (16%) 612,161 (11%)
Hypertension (%)
< 0.001
No 43,682,966 (85%) 39,218,707 (87%) 4,464,259 (77%) Yes 7,440,080 (15%) 6,110,381 (13%) 1,329,699 (23%)
Diabetes (%)
< 0.001
No 46,014,179 (94%) 41,112,084 (94%) 4,902,095 (89%) Yes 2,984,448 (6.1%) 2,401,930 (5.5%) 582,519 (11%)
High cholesterol (%)
< 0.001
No 45,007,588 (88%) 40,305,200 (89%) 4,702,388 (81%) Yes 6,115,458 (12%) 5,023,888 (11%) 1,091,570 (19%)
BRI (mean (SD))
5.38 (2.73) 5.24 (2.65) 6.43 (3.08)
< 0.001
Mean (SD) for continuous variables: the P value was calculated by the weighted linear regression model Percentages (weighted N, %) for categorical variables: the P value was calculated by the weighted chi-square test Abbreviation: BRI, body roundness index; PIR, Ratio of family income to poverty
Baseline characteristics of all participants were stratified by infertility, weighted
Mean (SD) for continuous variables: the P value was calculated by the weighted linear regression model
Percentages (weighted N, %) for categorical variables: the P value was calculated by the weighted chi-square test
Abbreviation: BRI, body roundness index; PIR, Ratio of family income to poverty
As shown in Table 2 , the relationship between BRI and infertility was assessed using three models. In Model 3, after fully adjusting for covariates, each unit increase in BRI was associated with a 12% increase in the probability of infertility (OR: 1.12; 95% CI: 1.05, 1.19). The prevalence of infertility increased progressively with higher BRI quartiles (with Q1 as the reference). The corresponding results were: Q2 [Odds Ratio: 1.31; 95% CI: 0.83, 2.05], Q3 [Odds Ratio: 1.72; 95% CI: 0.98, 3.02], and Q4 [Odds Ratio: 1.60; 95% CI: 1.52, 4.44]. Additionally, there was a statistically significant trend of increasing infertility prevalence with higher BRI (P for trend < 0.001). Figure 2 further illustrates the significant positive linear relationship between BRI and infertility prevalence (overall P < 0.001; nonlinearity P = 0.468). It is important to note (Table S4 ) that logistic regression results did not show statistical significance after PSM, although the differences between confounders were balanced.
Table 2 Adjusted odds ratios (ORs) of BRI and infertility, weighted BRI Model 1 [OR (95% CI)] p -value Model 2 [OR (95% CI)] p -value Model 3 [OR (95% CI)] p -value Continuous (Per 1 unit increase) 1.14 (1.09, 1.20) < 0.001 1.13 (1.07, 1.20) < 0.001 1.12 (1.05, 1.19) < 0.001 Quartile Q1 1 (ref.) 1 (ref.) 1 (ref.) Q2 1.48 (1.00, 2.19) 0.053 1.23 (0.79, 1.90) 0.300 1.31 (0.83, 2.05) 0.200 Q3 2.08 (1.32, 3.26) 0.002 1.71 (1.03, 2.83) 0.039 1.72 (0.98, 3.02) 0.057 Q4 3.14 (1.95, 5.06) < 0.001 2.62 (1.51, 4.52) 0.001 2.60 (1.52, 4.44) 0.001 P for trend < 0.001 0.001 0.001 Model 1: no covariates were adjusted Model 2: age, education level, marital, PIR, and race were adjusted Model 3: age, education level, marital, PIR, race, smoking, drinking, hypertension, diabetes, and high cholesterol were adjusted Abbreviation: BRI, body roundness index; PIR, Ratio of family income to poverty; ORs, odds ratios; CI, confidence interval
Adjusted odds ratios (ORs) of BRI and infertility, weighted
Model 1: no covariates were adjusted
Model 2: age, education level, marital, PIR, and race were adjusted
Model 3: age, education level, marital, PIR, race, smoking, drinking, hypertension, diabetes, and high cholesterol were adjusted
Abbreviation: BRI, body roundness index; PIR, Ratio of family income to poverty; ORs, odds ratios; CI, confidence interval
Fig. 2 The smooth curve fitting analysis of BRI and infertility. OR (solid lines) and 95% confidence levels (shaded areas) were adjusted for age, education level, marital, PIR, race, smoking, drinking, hypertension, diabetes, and high cholesterol were adjusted. Abbreviation: BRI, body roundness index; PIR, Ratio of family income to poverty; ORs, odds ratios; CI, confidence interval
The smooth curve fitting analysis of BRI and infertility. OR (solid lines) and 95% confidence levels (shaded areas) were adjusted for age, education level, marital, PIR, race, smoking, drinking, hypertension, diabetes, and high cholesterol were adjusted. Abbreviation: BRI, body roundness index; PIR, Ratio of family income to poverty; ORs, odds ratios; CI, confidence interval
As shown in Fig. 3 , a subgroup analysis was conducted based on age, race, marital status, education level, PIR, smoking, alcohol consumption, hypertension, diabetes, and hyperlipidemia. The results were like the main analysis, indicating no significant interaction effects (all p -values for interaction > 0.05).
Fig. 3 Subgroup analysis between BRI and infertility. ORs were calculated as each unit increased in BRI. Analyses were adjusted for age, education level, marital, PIR, race, smoking, drinking, hypertension, diabetes, and high cholesterol were adjusted. Abbreviation: BRI, body roundness index; PIR, Ratio of family income to poverty; ORs, odds ratios; CI, confidence interval
Subgroup analysis between BRI and infertility. ORs were calculated as each unit increased in BRI. Analyses were adjusted for age, education level, marital, PIR, race, smoking, drinking, hypertension, diabetes, and high cholesterol were adjusted. Abbreviation: BRI, body roundness index; PIR, Ratio of family income to poverty; ORs, odds ratios; CI, confidence interval
We compared the predictive ability of BRI and various body measurement indicators for infertility likelihood by calculating the area under the curve (AUC) (Fig. 4 ). In this analysis, BRI demonstrated a strong advantage over the other indicators (WWI, BMI, Weight) with an AUC of 0.618 (95% CI, 0.588–0.648).
Fig. 4 Receiver operating characteristic (ROC) curve analysis for infertility
Receiver operating characteristic (ROC) curve analysis for infertility
To ensure the robustness of the results, multiple imputation was performed for missing baseline data. The significant positive correlation between BRI and infertility was maintained (Table S2 ).
Discussion
This study demonstrates a positive correlation between BRI and infertility prevalence. After adjusting for various confounding factors, the positive association remained significant. Subgroup analyses showed that the relationship between BRI and infertility was consistent across different subgroups. Additionally, the results from smooth curve fitting and multiple imputation sensitivity analyses were similar, further supporting our conclusions. These findings suggest that BRI may be a useful predictor of infertility risk.
To our knowledge, this is the first study to investigate the association between BRI and infertility. The prevalence of obesity is increasing globally, and adipose tissue releases various bioactive molecules that affect reproductive health through multiple pathways [ 21 ]. The limitations of BMI as a measure of obesity are well-known; it does not account for differences in visceral fat distribution among individuals. While CT and MRI are standard methods for assessing visceral fat, they are expensive and time-consuming [ 22 ]. A meta-analysis indicated that BRI outperforms BMI, waist-hip ratio (WHR), body shape index (ABSI), and body adiposity index (BAI) in predicting metabolic syndrome [ 23 ]. Metabolic syndrome is a complex condition, with abdominal obesity and/or insulin resistance (IR) being increasingly recognized as its core components [ 24 ]. Studies assessing the metabolic and endocrine characteristics of obese women have observed that decreased secretion of FSH and LH coexists with hyperlipidemia and hyperinsulinemia, leading to the concept of “neurometabolic syndrome.” This underscores the profound impact of obesity on female reproductive function [ 25 ].
The mechanisms underlying the relationship between BRI and infertility are multifaceted and complex. Firstly, obesity, particularly visceral obesity, can affect female reproduction through direct mechanisms that damage the luteal phase and indirectly influence the hypothalamic-pituitary-ovarian (HPO) axis, causing neuroendocrine changes [ 26 ]. Obesity may impair the HPO axis, with hyperlipidemia and hyperinsulinemia in obese women leading to insensitivity to hypothalamic GnRH secretion [ 25 ]. Additionally, a study simulating hyperinsulinemia and hyperlipidemia in non-obese women showed that elevated insulin and lipid levels can acutely suppress LH and FSH, providing a possible mechanism for the relatively hypogonadotropic hypogonadism observed in obesity [ 27 ]. Another cross-sectional study indicated that obesity can reduce LH pulse amplitude and significantly decrease FSH secretion [ 28 ]. Secondly, systemic oxidative stress is positively correlated with visceral fat accumulation [ 29 ]. Ovarian adipose tissue induces oxidative stress either through the catalytic activity of NADPH oxidase or dysfunctional mitochondrial oxidative phosphorylation, which can damage oocytes through various pathways [ 30 ]. Thirdly, obesity is associated with low-grade inflammation, predominantly occurring in visceral fat deposits. Elevated lipoprotein lipase (LPL) and higher free fatty acid (FFA) uptake in visceral fat may lead to inflammation due to nutritional overload in the microcirculation of visceral adipose tissue [ 31 ]. Chronic inflammation can impact reproduction by damaging folliculogenesis, altering blood coagulability, and impairing endometrial receptivity through oxidative stress [ 32 ].
Numerous studies have shown the impact of obesity on female reproduction, and weight loss interventions have been demonstrated to benefit reproductive outcomes. A retrospective cohort study of 14,213 patients indicated that cumulative live birth rates (CLBRs) decrease with increasing BMI, while weight loss is beneficial for improving overall CLBRs [ 33 ]. Conversely, a randomized controlled trial involving 379 patients found that preconception-intensive lifestyle interventions did not improve fertility or reproductive outcomes [ 34 ], possibly due to differences in sample size. Previous research on obesity and reproductive health has predominantly used the World Health Organization (WHO) Body Mass Index (BMI) classification, which accounts for 50–70% of the variance in fat mass among non-pregnant women [ 13 ]. Recently, the focus has shifted towards more detailed management of obesity, particularly abdominal or central obesity. Various indices have been established to estimate central or abdominal obesity, including neck circumference (NC), waist-hip ratio (WHR), lipid accumulation product (LAP), visceral adiposity index (VAI), and Chinese visceral adiposity index (CVAI) [ 35 , 36 ]. Studies have linked visceral obesity with cancer [ 37 ], diabetes [ 38 ], and cardiovascular diseases [ 39 ]. As the first study to explore the relationship between BRI and infertility, our findings indicate a linear relationship between BRI and infertility prevalence. BRI, which better accounts for visceral fat distribution compared to BMI, may offer new insights into the management and treatment of women with infertility.
To date, this study is the first to investigate the relationship between BRI and infertility using the NHANES database. The large sample size is a significant advantage, and the findings remain robust after adjusting for numerous confounding factors, conducting subgroup analyses, and performing multiple imputation analyses. However, the study has several limitations. Firstly, as a cross-sectional study, it cannot establish a causal relationship between BRI and infertility, which necessitates further prospective studies. Secondly, although many confounding factors were adjusted for, the limitations of the NHANES database prevented the inclusion of some potential infertility confounders, such as anatomical causes of infertility, which may have a weaker association with BRI. In addition, PSM and adjusted logistic regression analysis have their own advantages. PSM improves the comparability between groups, but it may also affect the statistical power due to the reduction of sample size. We believe that the results of both methods provide important information for understanding the potential relationship between BRI and infertility. Furthermore, metabolism-related disorders such as polycystic ovary syndrome (PCOS) and endometriosis may have a significant impact on the development of infertility [ 40 , 41 ]. Due to limitations of the NHANES database, we were unable to obtain detailed information on polycystic ovary syndrome and endometriosis, and therefore could not directly control for these potential confounders in our analyses. Underrepresentation or imbalance of polycystic ovary syndrome and endometriosis in the study group may therefore have some impact on the interpretation of the results, and we need to explore these factors further in future studies. Finally, this study failed to differentiate between specific causes of infertility, which may affect the generalizability of this study’s findings. Future studies should consider stratifying the different causes of infertility to better understand the impact of obesity on different infertility types.