Cross-sectional association between adherence to a planetary health diet and self-reported infertility in U.S. women (NHANES 2013-2018).

This study has several important limitations that must be considered when interpreting the findings. First, and most critically, the cross-sectional study design precludes the establishment of temporal sequence or causality. We observed an association between higher adherence to the PHD and lower self-reported infertility, but we cannot determine whether the dietary pattern preceded and influenced fertility status or whether fertility problems or their treatments led to subsequent dietary changes. Cross-sectional studies are susceptible to reverse causality. For instance, women who encounter difficulties during the conception process may change their dietary habits to treat infertility, or make corresponding adjustments according to medical advice. This is a key consideration in cross-sectional reproductive epidemiological studies. Second, by definition, infertility is a condition of the couple, whereas this study focuses only on female infertility and uses a self-reported measure. This discrepancy between the clinical concept of infertility and the way it is operationalized in the study represents a potential source of bias. It is noteworthy that the potential exposure misclassification resulting from such errors would be biased toward the null and thus result in an underestimation of the association between PHDI and infertility. Third, It is also important to note that dietary data were derived from 2 nonconsecutive 24-hour recalls, which may not fully capture habitual long-term intake. This measurement error could have potentially weakened the true associations and reduced the accuracy of the PHDI classification. Fourth, dietary adherence was derived from 2 nonconsecutive 24-hour dietary recalls. While this method provides a quantitative snapshot, it may not represent an individual usual long-term dietary pattern, which is more relevant to chronic health outcomes. Day-to-day variability in food intake can lead to misclassification of an individual typical adherence to the PHD. Although we used the population ratio method, this approach is best suited for estimating group-level means rather than precise individual-level adherence. Finally, despite extensive adjustment for known confounders, residual confounding from unmeasured or imprecisely measured factors (e.g., genetic predisposition, detailed reproductive history, environmental exposures, or specific sexual behaviors) remains possible. The PHDI is a relatively new index. While this study presents its application, a critical discussion of potential challenges, such as scoring schemes, applicability to U.S. dietary patterns, and ceiling or floor effects, would be of added value. [ 10 , 12 , 14 , 15 ] Therefore, higher PHDI adherence is not necessarily associated with a reduction in self-reported infertility, as the dietary pattern may not meaningfully influence infertility risk in this context. Moreover, both exposure and outcome were subject to potential misclassification, which may affect the interpretation of the observed association. Collectively, these limitations: the cross-sectional design, the use of self-reported infertility, and the estimation of diet based on short-term recall, constrain any causal interpretation of the observed association between the PHDI and lower infertility prevalence. The findings should be considered hypothesis-generating and warrant prospective investigation rather than being interpreted as confirming a protective effect. Future longitudinal studies with repeated dietary assessments and clinically verified fertility outcomes are essential to clarify the potential role of sustainable dietary patterns in reproductive health. Despite these limitations, our study has notable strengths. It leverages a large, nationally representative sample with standardized data collection, enhancing the external validity of the findings. The use of the PHDI is innovative, as it incorporates both health and sustainability dimensions of diet; our study suggests this index can be informative in epidemiologic research beyond environmental or metabolic outcomes, extending to reproductive health. We adjusted for a broad array of covariates, including socioeconomic status and health behaviors, to isolate the independent association of diet quality. The consistency of the association in multiple sensitivity analyses (e.g., stratifications and weighted analyses) adds credibility to the robustness of the findings. Additionally, by examining both categorical and continuous PHDI, we demonstrated a clear trend that strengthens a potential dose-response interpretation.

Infertility, defined as the inability to achieve pregnancy after 12 months of regular unprotected intercourse, affects a considerable proportion of couples worldwide. Recent estimates from the World Health Organization indicate that about one in every 6 people of reproductive age globally experiences infertility during their lifetime. In the United States, approximately 12 to 15% of couples are unable to conceive after one year of trying, reflecting a similar magnitude of the problem. Infertility is not only a medical condition but also imposes considerable psychological, social, and economic burdens on affected individuals and families. Identifying modifiable risk factors for infertility is therefore a public health priority. [ 1 , 2 ] Lifestyle and environmental factors, including diet, have been increasingly recognized as important influences on fertility. Previous studies have shown that healthier dietary patterns are associated with improved fertility outcomes. [ 3 ] For example, in a large prospective cohort of women without known infertility, [ 4 ] adherence to a “fertility diet” (characterized by higher unsaturated fat intake, plant protein, high-fat dairy, iron, and multivitamins, along with lower trans fat and low-fat dairy) was associated with a significantly lower risk of ovulatory disorder infertility (approximately 80% lower relative risk for those following 5 or more low-risk dietary/lifestyle factors). Conversely, diets with high inflammatory potential have been linked to a higher likelihood of infertility. An analysis of National Health and Nutrition Examination Survey (NHANES) 2013 to 2018 data reported that women consuming a more pro-inflammatory diet (as indicated by a high Dietary Inflammatory Index) had significantly greater odds of infertility compared to those with more anti-inflammatory diets. [ 5 ] Similarly, a recent study introduced a novel dietary index for gut microbiota (DI-GM) and found that women with the highest DI-GM scores (indicative of a microbiome-friendly, fiber-rich diet) had a markedly lower prevalence of infertility (adjusted odds ratio [OR] 0.64 for the highest vs lowest score group). [ 6 ] Diet may influence fertility through multiple mechanisms, including effects on reproductive hormonal balance, ovulatory function, body weight, and metabolism. [ 7 ] Excess intake of trans fats, red and processed meats, and added sugars can promote systemic inflammation and oxidative stress, which may impair ovulation and endometrial receptivity, whereas diets abundant in fruits, vegetables, whole grains, and plant protein provide antioxidants and anti-inflammatory nutrients that may be conducive to reproductive health. [ 8 ] Moreover, diet is a key environmental factor that shapes the gut microbiome; dietary patterns high in fiber and plant foods may enhance gut microbial diversity and estrogen metabolism, which may in turn influence fertility. [ 9 ] In addition to health considerations, the concept of a “sustainable healthy diet” has gained traction in recent years. The EAT-Lancet Commission in 2019 introduced the Planetary Health Diet (PHD)–a dietary framework that aims to promote human health while also safeguarding environmental resources (e.g., by reducing the environmental footprint of food production). [ 10 ] The PHD emphasizes higher consumption of plant-based foods such as whole grains, fruits, vegetables, legumes, nuts, and unsaturated oils, and lower consumption of animal-based foods (especially red meat), refined grains, added sugars, and saturated fats. To quantify adherence to this dietary model, Cacau et al [ 11 ] developed the plant-based diet index (PHDI), a composite score reflecting how closely an individual diet aligns with the EAT-Lancet recommendations. The PHDI consists of 16 components encompassing various food groups and nutrient ratios, with scoring designed such that higher PHDI values indicate greater conformity to the planetary health diet (maximum score 150). Early research on the PHDI has linked higher scores to better nutrient profiles and health outcomes in other contexts: for instance, higher PHDI has been associated with improved cardiometabolic indicators and lower risks of chronic diseases. [ 12 ] Notably, recent case–control studies found that greater adherence to the planetary health diet was associated with reduced odds of colorectal cancer and breast cancer, suggesting a potential protective role of this sustainable eating pattern. [ 13 ] However, the relationship between the PHDI and reproductive health outcomes, such as infertility, has not yet been examined. In this study, we aimed to evaluate whether adherence to the Planetary Health Diet (as measured by PHDI) is associated with the prevalence of infertility among women of reproductive age. We utilized data from NHANES 2013 to 2018, which provided a large, nationally representative sample and detailed dietary information. We hypothesized that higher PHDI scores, indicative of diets richer in plant-based foods and closer alignment with EAT-Lancet guidelines, would be associated with lower odds of reporting infertility, even after controlling for potential confounding factors such as age, body mass index (BMI), lifestyle behaviors, and comorbid conditions.

Conceptualization: Qiao Li. Data curation: Qiao Li. Formal analysis: Qiao Li. Funding acquisition: Qiao Li. Investigation: Qiao Li. Methodology: Qiao Li. Project administration: Qiao Li. Resources: Qiao Li. Supervision: Qiao Li. Validation: Qiao Li. Visualization: Qiao Li. Writing – original draft: Qiao Li. Writing – review & editing: Qiao Li.

We conducted a cross-sectional analysis of data from NHANES 2013 to 2018. NHANES is an ongoing, nationally representative survey of the noninstitutionalized United States (U.S.) population, administered by the National Center for Health Statistics. It uses a complex, multistage probability sampling design to collect extensive health, nutrition, and demographic information through interviews, physical examinations, and laboratory tests. For the present study, we combined 3 consecutive NHANES 2-year cycles (2013–2014, 2015–2016, 2017–2018) to obtain a larger sample of women of childbearing age. We excluded women who were outside the 20 to 45 age range, as well as those with missing data on key variables (dietary data, infertility status, or covariates). We also excluded women who reported a history of hysterectomy or oophorectomy (surgical removal of uterus/ovaries), since they would not be at risk for natural conception. Of note, the timing of infertility occurrence corresponded to the year of dietary assessment. After applying inclusion and exclusion criteria, the final analytic sample consisted of N = 3105 women aged 20 to 45. Figure 1 illustrates the sample selection process. Flowchart of the participant selection process. NHANES = National Health and Nutrition Examination Survey, PHDI = plant-based diet index. Dietary intake for each participant was assessed via NHANES from 2 nonconsecutive 24-hour dietary recall interviews. The Planetary Health Diet Index (PHDI) was calculated for each participant as a summary measure of diet quality and sustainability alignment, based on the EAT-Lancet reference diet parameters. [ 10 ] We followed the approach described by Cacau et al [ 11 ] to construct the PHDI. In brief, the PHDI comprises 16 components representing major food groups or nutrients, which are categorized into four domains: Adequacy (components that should be consumed in adequate amounts: fruits, vegetables, legumes, nuts, and whole grains), optimal intake (foods to moderate: dairy, eggs, fish/seafood, vegetable oils, tubers), ratio components (ratios of certain vegetables to total vegetables, reflecting dietary diversity), and moderation (components to limit: red meat, poultry, animal fats, added sugars). Each component is scored on a scale (most on 0–10, or 0–5 for ratio components) based on the level of intake relative to the ideal range or limit defined by the planetary health diet. Scores for all components are summed to yield the total PHDI, which ranges from 0 (minimal adherence to the reference diet) to 150 (maximal adherence). A higher PHDI reflects a dietary pattern closer to the recommended balance of more plant-sourced and less animal-sourced foods. For analysis, we treated PHDI both as a continuous variable (per 1 standard deviation [SD] increase) and as a categorical variable by splitting the scores into tertiles (low, medium, high adherence). The mean PHDI in our sample was 43.0 (SD 14.3), indicating generally low alignment with the EAT-Lancet diet recommendations in this U.S. population. Dietary intake was assessed using 2 nonconsecutive 24-hour dietary recalls, collected by trained interviewers using the United States Department of Agriculture Automated Multiple-Pass Method. Day 1 data were collected in person at the Mobile Examination Center; day 2 data were collected via telephone interviews 3 to 10 days later, according to the methodology presented elsewhere. [ 12 ] The average of the 2 days was used to estimate the daily intake of foods and nutrients and total energy intake in kcal/day. Dietary data were then linked to the Food Patterns Equivalents Database, which categorizes foods into the 37 USDA Food Pattern Components, using a food composition table. [ 14 ] Food pattern equivalents were then converted into grams/day according to standardized conversion units. Adherence to the PHD was then assessed using the PHDI scoring system proposed by the EAT-Lancet Commission [ 10 ] and further adapted to the U.S. population [ 15 ] in order to reflect adherence to a nutritionally adequate and environmentally sustainable dietary pattern. The use of individual 24-hour dietary recalls as the basis for PHDI calculation may introduce measurement error. Notably, any resulting exposure misclassification would likely bias the association toward the null, potentially leading to an underestimation of the relationship between PHDI and infertility. The outcome of interest was self-reported infertility, defined using participants’ self-reports on the NHANES reproductive health questionnaire. Women were classified as having a history of infertility if they answered “Yes” to the question: “Have you ever attempted to become pregnant for at least a year without becoming pregnant?” This definition corresponds to the standard clinical definition of infertility (failure to conceive after 12 months of trying). Women who responded “No” to that question were considered not to have experienced infertility. We included both current and prior infertility experience in the outcome (i.e., a woman who had difficulty conceiving in the past but eventually became pregnant would still be counted as having a history of infertility). In additional sensitivity analysis, we also considered a more stringent definition requiring both a year of trying and having sought medical help for conception (data not shown), with similar results. Given that infertility is a clinical condition involving couples, it is unlikely to be identified solely through partner fertility assessment, which is typically conducted at birth or marriage. Therefore, this indicator is not considered a clinical diagnostic basis. [ 1 , 2 ] We identified several covariates a priori as potential confounders of the relationship between diet and infertility, based on literature and directed acyclic graph assumptions about common causes of diet quality and fertility. These included: age (in years, continuous), race/ethnicity (categorized as Non-Hispanic White, Non-Hispanic Black, Hispanic, or Other race), poverty-to-income ratio (PIR) as an indicator of socioeconomic status (categorized as 3.5, corresponding to low, middle, and high income relative to the federal poverty level), smoking status (ever smoker vs never smoker, defined by having smoked at least 100 cigarettes in life), alcohol use (classified as none/light vs moderate/heavy, using NHANES data on drinking frequency; here we considered “heavy alcohol use” as ≥12 drinks in the past year, consistent with NHANES categorization), BMI in kg/m 2 (continuous), secondhand smoke exposure (yes/no, defined by self-reported exposure to tobacco smoke at home or work), and hypertension (yes/no, defined by clinical measurement or self-reported diagnosis of high blood pressure or use of antihypertensive medication). These factors were chosen because they may influence both dietary habits and fertility. For example, older age, smoking, and obesity are known to impair fertility, and they might also correlate with poorer diet quality; thus, adjusting for these helps isolate the independent association of diet quality (PHDI) with infertility. All analyses accounted for the complex survey design of NHANES. We used appropriate sample weights for the combined 2013 to 2018 cycles to obtain nationally representative estimates, per NHANES analytic guidelines. Participant characteristics were summarized by infertility status using means or proportions, and comparisons were made using t -tests or chi-square tests as appropriate. PHDI was analyzed both as a continuous variable (per SD increase) and in tertiles to examine potential nonlinear associations. We used multivariable logistic regression to estimate the OR and 95% confidence interval (CI) for infertility associated with PHDI. Three models were constructed: unadjusted, adjusted for age and race/ethnicity, and fully adjusted for all covariates listed above (age, race/ethnicity, PIR, smoking, alcohol, BMI, secondhand smoke, hypertension, and total physical activity). Trend across PHDI tertiles was evaluated by assigning each tertile group its median PHDI value and modeling that as a continuous predictor in the logistic model, with the Wald test for linear trend. We also tested for effect modification by age group (<30 vs ≥30 years), BMI category (<25 vs ≥25 kg/m −2 ), and race/ethnicity by including interaction terms with PHDI in the models. Several sensitivity analyses were performed to assess the robustness of our findings. First, we ran the fully adjusted models using survey weights (to account for the NHANES sampling design) and using unweighted analyses; the point estimates were very similar, so we report the unweighted results for simplicity, as they are easier to interpret as ORs for the sample. Second, we conducted stratified analyses by categories of key covariates (age, BMI, smoking) to see if the association between PHDI and infertility held within subgroups. Third, we explored an alternative categorization of PHDI (quartiles instead of tertiles) and inclusion of additional covariates (physical activity and marital status) to check for residual confounding. None of these variations materially changed the results, so only the primary analysis is presented. All statistical tests were two-sided, and P  < .05 was considered statistically significant. Analyses were conducted using Stata 17.0 (StataCorp) and R 4.2 software. This study was exempt from institutional review board approval as it involved secondary analysis of de-identified public data; all NHANES participants provided informed consent, and NHANES protocols were approved by the NCHS Research Ethics Review Board.

As shown in Table 1 , a total of 3105 women aged 20 to 45 years were included in this study. The mean age of participants was 32.7 years (SD 7.5). Among these women, 402 reported a history of infertility (attempting pregnancy for ≥1 year without success). Table 1 and Table S1 describe the baseline characteristics of the study participants (N = 3105) stratified by tertiles of the PHDI. Participants in the highest PHDI tertile were more likely to be older, have a lower BMI (28.46 vs 30.64 kg/m 2 , P  < .001), and have a higher socioeconomic status, as indicated by higher income and education levels (both P  < .001). Significant differences were also observed across ethnic groups ( P  < .001). Healthier lifestyle behaviors were strongly associated with higher PHDI scores. Those in the highest tertile had a lower prevalence of current smoking (11.01% vs 27.63%, P  < .001), higher levels of physical activity ( P  < .001), and lower exposure to secondhand smoke (35.09% vs 61.17%, P  < .001). Furthermore, a higher PHDI was associated with a more favorable cardiometabolic risk profile, including lower rates of hypertension (14.88% vs 19.52%, P  = .020) and hyperlipidemia (48.99% vs 55.85%, P  = .004). The prevalence of infertility was lower in the highest PHDI group, although this difference did not reach statistical significance (10.92% vs 13.82%, P  = .058). Baseline characteristics of study participants by tertiles of PHDI (N = 3105). PHDI = plant-based diet index. The association between PHDI and self-reported infertility was evaluated using multivariable logistic regression models with progressive adjustment for potential confounders (Table 2 ). In multivariate analyses, an inverse association was observed between PHDI and infertility. A one-standard-deviation increase in PHDI score corresponded to a 12% lower odds of infertility (OR 0.88, 95% CI 0.79–0.99). When analyzed by tertiles, a dose-response pattern was suggested: participants in the highest PHDI tertile had a 25% lower odds of infertility (OR 0.75, 95% CI 0.56–0.99) compared to those in the lowest tertile, with a trend across categories ( P for trend = .0486). These associations persisted after adjustment for a comprehensive set of demographic, lifestyle, and clinical confounders. The curve, illustrated in Figure 2 , reinforces the correlation between infertility risk and PHDI, suggesting an association between higher PHDI and a decreased prevalence of infertility. Association between PHDI and self-reported infertility. Model I: Adjusted for age, ethnicity, poverty income ratio; Model II: Fully adjusted for all covariates (age, ethnicity, poverty income ratio, marital status, smoking status, alcohol use, body mass index, secondhand smoke, hypertension, and total physical activity). Cl = confidence interval, OR = pdds ratio, PHDI = plant-based diet index, SD = standard deviation. The correlation between PHDI and self-reported infertility rate. Fully adjusted for all covariates (age, ethnicity, poverty income ratio, marital status, smoking status, alcohol use, body mass index, secondhand smoke, hypertension, and total physical activity). The blue line represents the 95% confidence interval. PHDI = plant-based diet index. Figure 3 illustrates an inverse dose–response relationship between PHDI and biological aging acceleration multi-assay-age acceleration (MAAA) across age tertiles. In both the lowest (panel A) and middle (panel B) age strata, participants in the highest PHDI quintile exhibited markedly attenuated MAAA values compared with those in the lowest quintile ( P for trend < .001). Notably, in the youngest tertile, the difference in mean MAAA between extreme PHDI quintiles reached −0.60 SD (95% CI: −0.73–−0.47). Stratified sensitivity analysis: linear association of PHDI with self-reported infertility by age group. Fully adjusted for all covariates (age, ethnicity, poverty income ratio, marital status, smoking status, alcohol use, body mass index, secondhand smoke, hypertension, and total physical activity). PHDI = plant-based diet index. Table 3 presents the results of a stratification analysis examining the association between the PHDI tertiles and self-reported infertility, without adjustment for covariates. Key interaction analyses identified age and marital status as effect modifiers (interaction P  = .043 and 0.003, respectively). Among women aged 20 to 32 years, high-PHDI was significantly associated with reduced infertility risk (OR = 0.46, 95% CI: 0.29–0.73, P  = .001). The association further varied by marital status: among never-married women, both middle and high-PHDI showed strong inverse associations with infertility (e.g., high-PHDI: OR = 0.37, 95% CI: 0.19–0.72, P  = .004). In contrast, among widowed, divorced, or separated women, middle PHDI was associated with significantly elevated odds of infertility (OR = 2.75, 95% CI: 1.25–6.03, P  = .012). Associations for high-PHDI were also observed in Mexican Americans (OR = 0.41, P  = .010), individuals with hypertension (OR = 0.54, P  = .030), and those living below the poverty threshold (OR = 0.51, P  = .044). An interaction was noted for hyperlipidemia status ( P for interaction =.022). These findings should be interpreted as exploratory, suggesting that the relationship between PHDI and infertility may be modified by age, marital status, hyperlipidemia, and BMI. In summary, adherence to a planetary health dietary pattern may be linked to reduced infertility risk in specific subgroups: particularly among younger, never-married women and certain racial or health-disadvantaged groups, though these associations are exploratory and appear to be significantly modified by age, marital status, and underlying health conditions. Stratified analysis of the association between PHDI and self-reported infertility. Outcome variable: Self-reported infertility; Exposed variable: PHDI tertile, Adjusted variables: None. CI = confidence interval, PHDI = planetary health diet index, OR = odds ratio.

In this cross-sectional study of U.S. women aged 20 to 45, we found that greater adherence to the PHD, as reflected by higher PHDI scores, was associated with lower odds of self-reported infertility. Women whose diets most closely aligned with the EAT-Lancet sustainable diet recommendations had about 25 to 30% lower prevalence of self-reported infertility compared to women with the least adherence, even after adjusting for age, socioeconomic and lifestyle factors, BMI, and health conditions. We also observed a pattern suggestive of a dose-response relationship: infertility prevalence decreased steadily across increasing PHDI tertiles, and each SD increase in PHDI was associated with an approximately 12% lower prevalence of infertility. To our knowledge, this is the first study to evaluate the relationship between the PHDI and female infertility. Although the overall adherence to the planetary health diet was low in this nationally representative sample of U.S. women, the PHDI scores exhibited meaningful variation across individuals. This variability enabled us to examine relative differences in diet quality and their association with self-reported infertility. The observed associations should be interpreted in the context of generally low adherence levels, and greater contrasts in dietary adherence may be necessary to detect stronger or more consistent associations. Our results are consistent with prior research showing associations between plant-forward or anti-inflammatory diets and favorable fertility outcomes. A number of epidemiological studies have assessed specific diet indices or patterns in relation to infertility and reproductive success. [ 16 , 17 ] For instance, Harvard’s “fertility diet” pattern, which shares several features with the planetary health diet, such as higher intakes of plant protein and unsaturated fats and lower intakes of red meat and trans fats, was associated with lower odds of self-reported infertility in the Nurses’ Health Study II cohort. [ 18 ] Adherence to a Mediterranean diet (another plant-rich dietary pattern emphasizing fruits, vegetables, legumes, fish, and olive oil) has been associated with improved outcomes in assisted reproductive technologies; a recent review reported that women with higher Mediterranean diet adherence had greater chances of pregnancy and live birth in some in vitro fertilization studies. [ 19 ] These diverse lines of evidence all point toward the conclusion that an overall healthful diet, particularly one rich in plant-based foods and low in heavily processed, high-sugar, or pro-inflammatory foods, is associated with better fertility. Our study extends this concept by focusing on the PHD, which not only encapsulates a high-quality diet for health but also integrates an environmental sustainability perspective. The finding that PHDI is inversely related to infertility aligns with the idea that dietary patterns beneficial for the planet (e.g., reducing excessive meat intake) may also be relevant to human reproductive health. Our stratified analysis results indicate that the inverse association between higher PHDI adherence and self-reported infertility is not uniform across all subgroups. The inverse association appeared particularly pronounced among younger women (aged 20–32), Mexican American women, and women who have never been married. For the majority of other demographic, socioeconomic, and health-related strata, while point estimates for the highest PHDI tertile were generally below 1.0, the associations did not reach statistical significance. These findings indicate potential effect modification by age, ethnicity, and marital status. However, these results are exploratory and must be interpreted with caution due to the reduced statistical power within strata, the cross-sectional design precluding causal inference, and the multiple comparisons performed. Potential mechanisms underlying the effect modification observed in younger, Mexican American, and never-married women may involve a combination of biological susceptibility and socio-behavioral context. The association in younger women (20–32 years) may reflect a greater biological responsiveness to dietary influences during peak reproductive years, where ovarian function is less compromised by age-related decline and modifiable factors like nutrition may exert a more detectable effect. For Mexican American women, the traditional diet rich in legumes, vegetables, and grains aligns closely with the PHD; thus, high adherence may represent a sustained, culturally embedded pattern that may correlate with long-term anti-inflammatory and metabolic profiles relevant to reproductive health. In this article, the population with infertility was not considered for potential variables such as the reproductive system, endocrine system and gynecological aspects, such as polycystic ovary syndrome, endometriosis, disorders of ovulation, the use of contraceptive measures, and the accessibility of medical resources. Additionally, regarding never-married women, the association may be influenced by selection bias and unmeasured confounding, such as higher health consciousness, differential fertility intentions, or a lower prevalence of conditions like endometriosis that are both diet-sensitive and associated with partnership status; alternatively, never-married status may correlate with later childbearing plans where preconception health behaviors, including diet, are more deliberately optimized. There are several biological mechanisms that could speculatively relate to the observed association between a higher PHDI and lower infertility prevalence, though these pathways remain hypothesis-generating and cannot be directly tested with the current data. First, diets abundant in fruits, vegetables, nuts, and whole grains provide ample vitamins (such as folate), antioxidants, and phytochemicals that may be associated with improved oocyte quality and ovulatory function by reducing oxidative stress in the reproductive tract. It has been suggested that antioxidant intake may be associated with improved ovarian parameters and embryo quality in some studies. [ 20 ] Second, diets characterized by high consumption of red and processed meats, refined carbohydrates, and sugary foods have been proposed to contribute to systemic inflammation and insulin resistance, which are known to impair ovarian function and fertility. [ 21 , 22 ] A cross-sectional study using the same NHANES 2013 to 2018 data found that women with more pro-inflammatory diets (as indicated by a high Dietary Inflammatory Index score) were more likely to experience infertility. [ 8 ] A key feature of the planetary health diet is limited red and processed meat and moderate dairy intake; it has been hypothesized that high consumption of red/processed meats may be associated with endometriosis and ovulatory infertility, potentially via increased inflammation or exposure to endocrine-disrupting chemicals (such as persistent organic pollutants) that can accumulate in animal fat. [ 23 – 25 ] By limiting these foods, a high-PHDI diet could conceivably be associated with lower exposure to such factors. One plausible explanation is that a plant-focused diet tends to have a lower glycemic load and healthier fatty acid profile, which may be associated with improved insulin sensitivity. [ 26 ] Improved insulin sensitivity is particularly relevant for conditions like polycystic ovary syndrome, a common cause of anovulatory infertility, since dietary management that reduces insulin resistance has been suggested to be associated with more regular ovulation. [ 27 ] Third, diet quality is strongly linked to body weight and composition. [ 28 ] Obesity is a well-established risk factor for infertility, due to hormonal imbalances (e.g., hyperestrogenism, insulin resistance) and other effects on ovulation and implantation. [ 29 ] In our analysis, we adjusted for BMI, which means the PHDI–infertility association was independent of baseline adiposity; however, it remains possible that part of the association between a high-PHDI diet and fertility may still be mediated by improved weight management. Consistent with this hypothesis, in the DI-GM study, the authors estimated that BMI mediated roughly 6% of the diet–infertility relationship. [ 30 , 31 ] The generally healthier body weight profile among women with high-PHDI in our study raises the possibility that maintaining a plant-rich diet may be associated with lower obesity prevalence, which in turn is linked to lower infertility prevalence. Finally, an additional speculative mechanism involves the gut microbiome. An analysis of NHANES data (2013–2020) introduced a DI-GM and found that women in the highest DI-GM category (e.g., those eating in a way that promotes gut microbial diversity, with high fiber and probiotic food intake) had a lower prevalence of infertility (OR = 0.64) compared to those in the lowest category. It has been proposed that diets high in fiber and plant foods help maintain a diverse gut microbiota, which can influence systemic inflammation and estrogen metabolism. [ 6 ] The gut microbiota produces enzymes (such as β-glucuronidase) that regulate the recirculation of estrogen in the body. [ 32 ] If a diet poor in fiber leads to gut dysbiosis, estrogen metabolism may be altered, resulting in suboptimal hormone levels for reproduction. [ 33 ] Thus, one could hypothesize that a high-PHDI diet may correspond to a favorable microbiome that could be involved in hormonal balance conducive to fertility. These mechanistic considerations, while beyond the scope of our dataset to directly assess, are intended to be hypothesis-generating and suggest plausible pathways through which adherence to a planetary health diet might relate to fertility. Further studies are warranted to explore these potential mechanisms.

In this nationally representative sample of U.S. women, greater adherence to the planetary health diet, as reflected by higher PHDI scores, was associated with lower prevalence of self-reported infertility. Given the cross-sectional design, causality cannot be established; however, the results underscore diet quality as a potentially important factor in self-reported infertility. Prospective cohort studies and intervention studies with repeated dietary assessments and clinically validated fertility outcomes are needed to further evaluate these associations.

The authors thank all the participants and researchers who contributed and collected data.