Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Oct 10.
Published in final edited form as: Sci Total Environ. 2024 Jul 2;946:174475. doi: 10.1016/j.scitotenv.2024.174475

Childhood and adolescent residential and farm pesticide exposures and inflammatory bowel disease incidence in a U.S. cohort of women

Dazhe Chen 1, Jennifer MP Woo 1, Christine G Parks 1, Kaitlyn G Lawrence 1, Katie M O’Brien 1, Robert S Sandler 2,3, Dale P Sandler 1
PMCID: PMC11296211  NIHMSID: NIHMS2007497  PMID: 38964382

Abstract

Background and Aims:

There are few known risk factors for inflammatory bowel disease (IBD), an autoimmune disease characterized by chronic intestinal inflammation. Use of specific pesticides has been associated with higher incidence of IBD among pesticide applicators and their spouses, but no study has examined pesticide exposure in early life, a period where the human immune system undergoes rapid changes. We evaluated pesticide use during childhood and adolescence and incidence of IBD among US women enrolled in the Sister Study.

Methods:

Incident IBD diagnoses between enrollment (2003–2009) and 2021 were identified and validated with medication use and colectomy/colostomy surgery. We estimated hazard ratios (HR) and 95% confidence intervals (CI) for the relationship of childhood/adolescent residential and farm pesticide exposures with IBD incidence using Cox models, accounting for age, race and ethnicity, education, smoking, and birth year.

Results:

We identified 277 incident IBD cases among 48,382 eligible participants. IBD hazard was elevated among those whose childhood residence was regularly treated with pesticides, especially among those who ever personally applied pesticides (HR: 1.39, 95%CI: 0.65, 2.99). We observed a positive association between IBD and exposure to broadcast pesticide sprays before DDT was banned (>6 times vs. never HR: 1.56, 95%CI: 1.06, 2.31). Among participants who lived on a farm during childhood/adolescence for ≥1 year (N=9,162), IBD hazards were higher among those who were in crop fields during pesticide application (HR: 2.06, 95%CI: 0.94, 4.51) and who ever personally applied pesticides on crops (HR: 1.85, 95%CI: 0.81, 4.18) or livestock (HR: 2.58, 95%CI: 1.14, 5.83).

Conclusion:

Early-life pesticide exposure may be a novel risk factor for IBD. Practices that reduce pesticide exposure during early life may help reduce the burden of this disease.

Keywords: pesticides, occupational exposures, inflammatory bowel disease, autoimmune disease, epidemiology

1. Introduction

Inflammatory bowel disease (IBD) is an umbrella term for Crohn’s disease (CD) and ulcerative colitis (UC), which are autoimmune diseases characterized by chronic inflammation in the gastrointestinal tract (Hanauer, 2006). The burden of IBD on individual patients is substantial. As a debilitating condition with a relapsing course, IBD is associated with high morbidity (Argollo et al., 2019) and can significantly impair patients’ quality of life and level of functioning (Knowles et al., 2018). In addition, the extensive care required to manage this lifelong disease contributes to considerable healthcare resource utilization and costs (Lichtenstein et al., 2020). Although IBD was traditionally regarded as a disease of high-income nations, a recent rapid rise in incidence has been observed in newly industrialized nations (GBD 2017 Inflammatory Bowel Disease Collaborators, 2020), calling for research to better understand the etiology and identify modifiable risk factors for IBD to reduce its global burden.

Although understanding of the etiology of IBD remains elusive, studies suggest that the pathogenesis involves genetic susceptibility, external environment, and immune response (Piovani et al., 2019). Some research suggests that environmental exposures could contribute to autoimmunity in susceptible individuals by disrupting gut microbiota, an integral part of gut immunity that maintains intestinal barrier function and modulates gut-specific immune response (Kamada et al., 2013). Because the gut microbiome and the human immune system undergo substantial changes in the first few years of life, early life represents a sensitive window during which environmental factors could predispose people to autoimmune diseases (Rodríguez et al., 2015). Several early-life environmental factors (e.g., antibiotic use, breastfeeding, contact with livestock, urban living) have been linked to IBD in epidemiology studies, with disruption in gut microbiota (i.e., gut dysbiosis) as a putative mechanism (Agrawal et al., 2021; Cholapranee & Ananthakrishnan, 2016; Song et al., 2019).

In animal studies, perinatal exposure to specific pesticides (e.g., chlorpyrifos, glyphosate) have been shown to induce dysbiosis, inflammation, and histologic lesions in the gut (Fathi et al., 2020; Joly Condette et al., 2015; Mao et al., 2018). Several pesticides were associated with IBD incidence in a cohort of pesticide applicators and their spouses (Chen et al., 2024), but no study has investigated early-life pesticide exposure in relation to IBD. To learn more about the potential role of pesticides in IBD, we performed the first study of early-life residential and farm pesticide exposures and IBD in a large United States (U.S.) cohort. We expect the incidence of IBD to be higher among participants exposed to pesticides in early life, and the magnitude of effect may be larger among those who were directly exposed.

2. Materials and methods

2.1. Study population

The Sister Study is a prospective cohort of 50,884 women recruited in all 50 U.S. states and Puerto Rico from 2003 to 2009. Eligible women were 35–74 years of age and had a sister with a history of breast cancer but did not have breast cancer themselves at enrollment (Sandler et al., 2017). At enrollment, participants completed computer-assisted telephone interviews that ascertained information on socio-demographics, lifestyle, medical history, residential history, and various environmental and occupational exposures. Participants completed brief health updates annually and detailed follow-up questionnaires (DFUs) every 2–3 years (2008–2012, 2012–2014, 2014–2016, 2017–2019, 2020–2021) to update changes in health, including new IBD diagnosis. Follow-up response rates have been approximately 90% over time. Questionnaires are available online (https://www.sisterstudystars.org/). The institutional review board of the National Institutes of Health oversees the study, and all participants provided written informed consent.

We restricted our analysis to the 50,873 participants with available information on IBD diagnosis. By starting follow-up at study enrollment, we effectively excluded 1,016 IBD cases diagnosed prior to enrollment, leaving 49,857 participants at risk of being diagnosed with incident IBD. We excluded 601 participants who were missing any covariates adjusted for in the main analysis model and 248 participants with inconsistent reporting of IBD diagnosis across the DFUs. To further reduce potential misclassification in IBD associated with self-report, we excluded 626 participants who reported an IBD diagnosis that was not confirmed by use of an IBD medication or a colectomy/colostomy surgery. The analytical sample for the residential pesticide exposure analyses thus consisted of 48,382 participants. For analyses of farm pesticide exposures, we further restricted the sample to the 12,547 participants who completed a residential farm exposure module at enrollment and the 9,162 participants who reported spending ≥12 months in a farm residence before age 19 years. A participant selection flowchart is provided in supplemental materials (Figure S1).

2.2. Exposure assessment

Participants completed a residential history, which included questions on the residence where they lived the longest during childhood (<age 14 years). Questions were also asked about regular pesticide treatment for this childhood residence, the frequency of treatment (daily, weekly, monthly, every 2–3 months, once or twice a year), and whether the participants applied the pesticides themselves. Participants were also asked to report if they were ever in the fog or spray of pesticides or chased fogger trucks or airplanes, and if yes, if these experiences occurred before 1975 (around the time when DDT was banned) and the number of occurrences. Participants were asked if their childhood residence was ever used as a farm or orchard for any of the time they lived there, or if it was in proximity to (“within seeing, smelling, or hearing distance”) a farm or orchard. They were also asked if this residence was in an urban, suburban, small town, or rural area.

Participants who reported ever living on a farm for ≥12 months during their lifetime or whose current or longest childhood or adult residence was a farm were asked to complete a residential farm exposure module. Participants who reported spending ≥12 months in a farm residence before adulthood (<age 19 years) were asked to recall characteristics of the farm. Questions included whether pesticides were ever used on crops; whether participants personally mixed, loaded or applied pesticides, or cleaned the equipment used for pesticide mixing or application; and whether participants were ever present in the fields when pesticides were applied to crops. Participants were also asked if the farm raised livestock; if they had contact with the livestock; if the livestock or the building in which they were kept were treated with pesticides; and if they personally applied the pesticides to the livestock or the livestock housing.

2.3. Outcome assessment

IBD diagnoses were identified via self-report at enrollment, in each of the five DFUs, and at annual health updates. At enrollment, participants were asked if a doctor or other health professional had ever told them that they had a) Crohn’s disease and/or b) ulcerative colitis. Those who responded “yes” to either were asked to provide the age at the IBD diagnosis. In addition, those who reported an IBD diagnosis at enrollment were asked if they ever took medication and ever had a colostomy or colectomy surgery for this disease.

In the DFUs, participants were asked if they were ever diagnosed with CD and UC. Those who reported being diagnosed with IBD since the previous follow-up were asked to provide the age at diagnosis. Unlike the enrollment questionnaire, the DFUs did not ask about medication use or colostomy/colectomy surgery for IBD immediately following report of IBD diagnosis. Use of medications was ascertained for other medical diagnoses and as free-text responses to an open-ended question (“Do you currently take any prescription or non-prescription medications regularly or seasonally?”). In the DFUs and annual health updates, participants reported the surgeries that they underwent since the previous follow-up.

In the analysis, we examined incident IBD diagnosed between study enrollment and the end of follow-up (September 2021) confirmed by use of an IBD medication and/or a colostomy/colectomy surgery. Medications reported in questionnaires were classified using the Slone Drug Dictionary to identify medications used for IBD (Supplemental Text) (Kelley et al., 2003; Vafai et al., 2022). Reported surgeries were assigned Current Procedural Terminology codes for identification of colostomy/colectomy surgeries (Supplemental Text).

2.4. Covariates

All covariates were ascertained at enrollment and included: age, race and ethnicity (non-Hispanic White, non-Hispanic Black, Hispanic, other), birth year, highest household educational attainment when the participant was 13 years old (≤high school, some college or associate/technical degree, Bachelor’s degree, graduate school), highest education attained by the participant at study enrollment (≤high school, some college or associate/technical degree, Bachelor’s degree, graduate school), cigarette smoking (current, former, never), ever being breastfed (definitely/probably, probably not/definitely not), ever regular antibiotic use (≥3 times per week for 3 months in a row or longer), and ever hormonal birth control use.

2.5. Statistical modeling

We used multivariable Cox proportional hazards models with age as the timescale to assess the associations of early-life pesticide exposures with incidence of IBD after enrollment (Cox, 1972). The minimally sufficient adjustment set (age, race and ethnicity, highest household educational attainment at age 13, birth year) was identified based on a directed acyclic graph (Figure S2) (Greenland et al., 1999). Hazard ratios (HRs) and 95% confidence intervals (CIs) were adjusted for the minimally sufficient adjustment set. We also adjusted in models for smoking and participant’s highest educational attainment, which are risk factors for IBD, to improve the precision of effect estimates. For broadcast sprays, we examined exposures before 1975 because exposures after 1975 were likely to occur in adulthood only and because we were interested in exposure to DDT specifically, which was banned in the U.S. in the early-1970s. We tested the assumption of proportional hazards in the main analysis by including in the model interaction terms of age with exposure and each covariate. Findings from the main analyses were highlighted in the text when HRs were elevated (>1.20), and we characterized an association as “modest” when the HR was between 1.20 and 1.30.

In a sub-analysis, we evaluated associations between exposures and UC alone; there were too few cases to conduct separate analyses for CD. We performed a sensitivity analysis that additionally adjusted for ever being breastfed, regular antibiotic use, and hormonal birth control use, which have been associated with the incidence of IBD in some studies (Piovani et al., 2019). To account for continued adulthood pesticide exposures, we restricted to participants who did not live on a farm or orchard during adulthood. In sub-analyses, we restricted the study population to non-Hispanic White participants and separately, to participants who were not current smokers. We were underpowered to examine associations in other racial/ethnic groups or among ever and never smokers. In a sensitivity analysis of farm exposures, we included participants who did not complete the residential farm exposure module as the referent group for comparison with other exposure groups examined in the main analysis. In another analysis, we examined all incident IBD diagnoses, including diagnoses that were not confirmed by use of an IBD medication or a colostomy/colectomy surgery. We also performed an analysis of pesticide exposures and IBD cases that occurred between pesticide exposure (i.e., age 14 for residential exposures and age 19 for farm exposures) and the end of follow-up (September 2021). Because participants who developed IBD early may have had more difficulty recalling the age at diagnosis, we estimated odds ratios using multivariable logistic regression models, which was not sensitive to timing of event, in this sub-analysis. All analyses were performed using SAS, version 9.4 (SAS Institute Inc., Cary, NC, USA) using Sister Study Data Release 11.1.

3. Results

During a median follow-up of 13.5 years (range: 0.1–17.9 years), we identified 903 incident IBD diagnoses, of which 277 were confirmed by medication use or colostomy/colectomy surgery and examined in analyses. Compared to participants without IBD, those with IBD were observed to be older, to have lower household educational attainment when they were 13 years old, and to have fewer years of school themselves at enrollment (Table 1). Participants with IBD were also more likely to have ever smoked at enrollment and at age 19 years old and to have reported ever regular antibiotic use. Compared to the residential exposure analysis population, analytical samples for the farm exposure analyses tended to be older, had lower childhood household and personal educational attainment, have been breastfed, and have never smoked at enrollment and before age 19 years (Table S1). We identified IBD subtype for 258 participants with IBD (201 UC and 57 CD). We classified 12 women who reported having both UC and CD in the same questionnaire(s) as having “indeterminate colitis”. IBD subtype was not reported for 7 cases, who were identified from free-text questionnaire responses in annual health updates or from next-of-kin reports.

Table 1.

Characteristics of Sister Study participants by incident inflammatory bowel disease case statusa

Characteristic Non-case
(n=48,105) 		Case
(n=277)
n (%) n (%)
Age at enrollment (years)
 < 45 6453 (13.4) 35 (12.6)
 46–55 16811 (35.0) 69 (24.9)
 56–65 16757 (34.8) 111 (40.1)
 ≥ 65 8084 (16.8) 62 (22.4)
Birth year
 ≤ 1940 6725 (14.0) 53 (19.1)
 1941–1950 16339 (34.0) 107 (38.6)
 1951–1960 17506 (36.4) 79 (28.5)
 ≥ 1961 7535 (15.7) 38 (13.7)
Race and ethnicity
 Non-Hispanic White 40351 (83.9) 241 (87.0)
 Non-Hispanic Black 4165 (8.7) 17 (6.1)
 Hispanic 2364 (4.9) 10 (3.6)
 Other Race/Multirace 1225 (2.6) 9 (3.3)
Highest household educational attainment at age 13 years
 ≤ High school diploma/GED 26037 (54.1) 160 (57.8)
 Some college/Associate/Technical 9081 (18.9) 57 (20.6)
 Bachelor’s 7926 (16.5) 31 (11.2)
 Graduate School 5061 (10.5) 29 (10.5)
Highest educational attainment at enrollment
 ≤ High school diploma/GED 7161 (14.9) 35 (12.6)
 Some college/Associate/Technical 16133 (33.5) 102 (36.8)
 Bachelor’s 13134 (27.3) 84 (30.3)
 Graduate School 11677 (24.3) 56 (20.2)
Smoking status at enrollment
 Never smoked 27215 (56.6) 135 (48.7)
 Former smoker 17031 (35.4) 116 (41.9)
 Current smoked 3859 (8.0) 26 (9.4)
Smoking before age 14 years
 No 46693 (97.1) 270 (97.8)
 Yes 1410 (2.9) 6 (2.2)
 Missing 2 1
Smoking before age 19 years
 No 34182 (71.1) 177 (64.1)
 Yes 13921 (28.9) 99 (35.9)
 Missing 2 1
Being breastfed
 Probably/definitely not 22365 (51.1) 129 (48.7)
 Probably/definitely yes 21440 (48.9) 136 (51.3)
 Missing 4300 12
Regular antibiotic use before enrollmentb
 No 41574 (89.5) 229 (85.5)
 Yes 4898 (10.5) 39 (14.6)
 Missing 1633 9
Ever hormonal birth control use before enrollment
 No 6990 (14.6) 40 (14.5)
 Yes 40860 (85.4) 236 (85.5)
 Missing 255 1
Open in a new tab

Abbreviations: GED, General Equivalency Diploma

a

Analytical sample for a complete case analysis excluding participants missing any covariates adjusted for in the main analysis (age at enrollment, birth year, race and ethnicity, highest household educational attainment at age 13 years, highest educational attainment at enrollment, smoking status at enrollment)

b

Regular antibiotic use was defined as using antibiotics at least three times per week for three months in a row or longer.

In fully-adjusted models, regular pesticide treatment at the long-term childhood residence was associated with modestly elevated hazards of IBD (HR=1.26, 95%CI: 0.94, 1.70) (Table 2). Further, compared to participants whose childhood residence was never treated with pesticides, the magnitude of effect estimate was larger among those who personally applied pesticides (HR=1.39, 95%CI: 0.65, 2.99) compared to those whose homes were treated with pesticides but who did not personally apply pesticides (HR=1.25, 95%CI: 0.92, 1.71). When we examined frequency of residential pesticide treatment, participants whose residence was treated weekly or daily had the highest hazards (HR=1.61, 95%CI: 0.66, 3.96). Compared to participants who were never exposed to broadcast pesticide sprays from trucks or airplanes before 1975, those reporting more-than-median occurrences of broadcast spray exposure had elevated IBD hazards (HR=1.56, 95%CI: 1.06, 2.31). We also found modestly higher hazards of IBD among participants reporting a long-term childhood residence near (HR=1.27, 95%CI: 0.90, 1.80) or used as a farm/orchard (HR=1.34, 95%CI: 0.97, 1.84). We did not find an association between childhood residence area (small town, suburban, urban, rural) and IBD hazards.

Table 2.

Associations between residential and farm exposures in childhood and adolescence and incident inflammatory bowel disease after study enrollment among Sister Study participants, 2003–2021

Exposure Total Cases Total N Age-adjusted Model Fully-adjusted Modela
HR (95% CI) HR (95% CI)
Residential Exposures before Age 14 b
Residence treated regularly with pesticides
 Never 175 33141 Referent Referent
 Ever 61 10318 1.19 (0.89, 1.59) 1.26 (0.94, 1.70)
  Frequency
   Treated every 2–3 months or less 44 7369 1.20 (0.86, 1.67) 1.27 (0.91, 1.77)
   Treated monthly 10 1767 1.18 (0.62, 2.23) 1.27 (0.67, 2.42)
   Treated weekly or daily 5 674 1.50 (0.62, 3.66) 1.61 (0.66, 3.96)
  Personally applied
   No 54 9146 1.18 (0.87, 1.61) 1.25 (0.92, 1.71)
   Yes 7 1149 1.26 (0.59, 2.69) 1.39 (0.65, 2.99)
Exposure to broadcast pesticide sprays before 1975
 Never 220 40187 Referent Referent
 No more than the median (≤6 times) 23 3857 1.10 (0.72, 1.69) 1.10 (0.72, 1.70)
 More than the median (>6 times) 29 3355 1.60 (1.09, 2.36) 1.56 (1.06, 2.31)
Residence near or used as a farm/orchard
 Residence not used as or near a farm/orchard 180 34692 Referent Referent
 Residence near a farm/orchard 39 5951 1.27 (0.90, 1.80) 1.27 (0.90, 1.80)
 Residence used as a farm/orchard 50 6771 1.38 (1.00, 1.89) 1.34 (0.97, 1.84)
Residential area
 Rural 71 11109 Referent Referent
 Small town 63 12340 0.80 (0.57, 1.12) 0.79 (0.56, 1.11)
 Suburban 68 13462 0.80 (0.57, 1.11) 0.83 (0.59, 1.17)
 Urban 70 10786 1.02 (0.74, 1.42) 1.02 (0.73, 1.43)
Farm Exposures before Age 19
Lived on a farm for 12 months or morec
 Never 21 3385 Referent Referent
 Ever 69 9162 1.21 (0.74, 1.97) 1.23 (0.75, 2.02)
Pesticide use on cropsd
 Never 15 2061 Referent Referent
 Ever 45 4951 1.23 (0.68, 2.23) 1.28 (0.70, 2.34)
  Personally applied or in fields during pesticide use to crops
   Used, not personally applied or in fields during pesticide use 19 2804 0.95 (0.48, 1.89) 1.01 (0.51, 2.03)
   Used, in fields during pesticide use but not personally applied 12 890 1.93 (0.89, 4.18) 2.06 (0.94, 4.51)
   Used, personally applied pesticides 10 841 1.69 (0.75, 3.79) 1.85 (0.81, 4.18)
Livestock contact and pesticide use on livestockd
 No livestock contact 17 2918 Referent Referent
 Contact, but no pesticides used 20 3223 1.06 (0.55, 2.02) 1.04 (0.54, 1.99)
 Contact, unsure if pesticides used 17 1375 2.14 (1.09, 4.18) 2.13 (1.08, 4.18)
 Contact, used pesticides but did not personally apply 6 1053 0.94 (0.37, 2.38) 0.90 (0.36, 2.30)
 Contact, personally applied pesticides 9 540 2.78 (1.24, 6.24) 2.58 (1.14, 5.83)
Open in a new tab

Abbreviations: HR, hazard ratio; CI, confidence interval

a

Cox proportional hazards models accounted for age, race and ethnicity, highest household educational attainment at age 13, highest educational attainment at enrollment, smoking status at enrollment, and birth year.

b

Analysis of 48,382 participants who did not have inflammatory bowel disease prior to study enrollment and had complete data in covariates adjusted for in analyses. Numbers do not always add to total because of missing values in exposures.

c

Analysis of 12,547 participants who additionally completed a residential farm exposure module. Numbers do not always add to total because of missing values in exposures.

d

Analysis of 9,162 participants who reported living on a farm/orchard before age 19 for 12 months or more. Numbers do not always add to total because of missing values in exposures.

Among participants who completed the residential farm exposure module, we found modestly higher hazards of IBD among those who lived on a farm for 12 months or more during childhood or adolescence (HR=1.23, 95%CI: 0.75, 2.02) (Table 2). Among participants who lived on a farm during childhood or adolescence for ≥12 months, we found a modest association between IBD and ever pesticide use on the crops grown on the farm (HR=1.28, 95%CI: 0.70, 2.34). We observed higher IBD hazards among participants who were in crop fields during pesticide application (HR: 2.06, 95%CI: 0.94, 4.51) and who ever personally applied pesticides on crops (HR: 1.85, 95%CI: 0.81, 4.18) compared to those who were never exposed. Compared to participants who did not have contact with livestock raised on the farm, those who personally applied pesticides to the livestock or the buildings where they were kept had higher IBD hazards (HR: 2.58, 95%CI: 1.14, 5.83). Associations were also positive among participants who were unsure if pesticides were used on livestock (HR: 2.13, 95%CI: 1.08, 4.18). We did not find a positive association among participants who had contact with livestock but were not exposed to pesticides or were exposed but did not personally apply pesticides. We did not find evidence of violation in the proportional hazards assumption in the main analysis (interaction p>0.05).

In analyses of UC alone, we found slightly attenuated associations in some analyses accompanied by wider confidence intervals, but interpretations remained similar (Table S2). Associations were similar in models additionally adjusted for regular antibiotic use, ever hormonal birth control use, and ever being breastfed (Table S3). Excluding participants who lived on a farm during adulthood (N=3,206, 6.6%), we found an attenuated association comparing those living on a farm during childhood or adolescence for ≥12 months vs. <12 months and stronger associations for personal application of pesticides on childhood residence and livestock/livestock housing; other associations were similar to those in the main analysis (Table S4). We observed similar results when we examined the associations among non-Hispanic White participants (Table S5) and participants who were not current smokers (Table S6), although confidence intervals were wider due to the smaller number of cases in these sub-analyses. In farm exposure analyses where we included participants who did not answer the residential farm exposure module as the referent group, we found somewhat stronger associations for those exposed to pesticides used on crops or livestock (Table S7).

In a sensitivity analysis where we examined all 903 incident IBD diagnoses, including the 626 diagnoses that were not confirmed by medication use or surgery, we observed similar associations accompanied by narrower confidence intervals (Table S8). In another analysis where we examined IBD diagnoses (validated by medication or surgery) between early-life exposure and the end of follow-up, we found somewhat attenuated associations (Table S9).

4. Discussion

In the first study of early-life pesticide exposures and incidence of IBD, we found evidence linking adult-onset IBD to several measures of early-life pesticide exposure, including pesticide treatment of residence, pesticide use on crops and livestock, and exposure to broadcast pesticide spray. In addition, we observed the highest IBD hazards among participants who personally applied pesticides. The magnitude of effect associated with personal application of pesticides was similar to that for exposure to antibiotics in infancy (odds ratio: 1.7), as estimated in a meta-analysis (Agrawal et al., 2021).

Increasing evidence suggests that the gut microbiome, which colonize the intestine at birth, plays a pivotal role in gut immunity and pathogenesis of multiple autoimmune diseases, including IBD (Zheng et al., 2020). A primary function of the gut microbiota is to protect the intestine against invasion by exogenous pathogens and overgrowth of indigenous pathobionts via several mechanisms, including enhancement of intestinal barrier function, direct competition for limited nutrients, and modulation of innate and adaptive immune responses (Kamada et al., 2013). Disruption in gut microbiome, either from environmental exposures or genetic defects, makes the host vulnerable to infections by pathogens and accumulation of colitogenic pathobionts (Belkaid & Hand, 2014). Dysbiosis can also result in impaired barrier function, which leads to translocation of microbes and food antigens into the mucosal layer, triggering the hyperactive inflammatory response seen in IBD patients (Belkaid & Hand, 2014). Because the gut microbiota undergoes rapid expansion and diversification in the first few years after birth, early life represents a critical period by which environmental factors could affect the maturation of the immune system and long-term health (Rodríguez et al., 2015).

In animal studies, numerous pesticides have been shown to induce gut dysbiosis and intestinal inflammation (Jin et al., 2015; Meng et al., 2022; Qiu et al., 2020; Tang et al., 2021; Tang et al., 2020), providing biological plausibility for the association between pesticide exposure and IBD. A few studies examined effects of perinatal or early-life pesticide exposures among exposed animals and found altered composition or abundance of microbiome and elevated markers of inflammation (Guardia-Escote et al., 2020; Jin et al., 2016; Jin et al., 2015; Mao et al., 2018; Qiu et al., 2020). Besides dysbiosis, specific pesticides have also been shown to dysregulate gut immune responses (Sanmarco et al., 2022) by affecting the viability and functions of immune cells (Lee & Choi, 2020) and modulating aryl hydrocarbon receptor and endocrine signaling (Mokarizadeh et al., 2015; Sanmarco et al., 2022), which may also contribute to the onset of autoimmunity.

To our knowledge, only two studies have evaluated the relationship between pesticide use and IBD; however, neither study examined exposures during early life. In an ecologic study of a California cohort, IBD incidence was not associated with total pounds of active pesticide ingredients used in production at the zip code level; however, the study was limited by a shorter follow-up and lack of individual-level data (Okafor et al., 2022). In contrast, a prospective cohort study of pesticide applicators and their spouses found positive associations of IBD with self-reported ever use of five organochlorine insecticides, three organophosphate insecticides, one fungicide, and five herbicides at study enrollment (Chen et al., 2024). Our study adds to the evidence for pesticide exposure as a contributor to IBD development, and we further showed childhood and adolescence as a potential window of susceptibility to pesticide exposures. Although we did not collect information on the specific pesticides used by participants, exposure to broadcast pesticide before 1975 could be considered a proxy for childhood/adolescent exposure to DDT, which was widely used in the U.S. in the 1950s-60s before it was banned in the early 1970s (Chen et al., 2020). Although we did not ascertain the age at exposure, 90.0% and 99.8% of participants who reported an exposure to broadcast sprays before 1975 were <19 years old in 1960 and 1950, respectively and were likely first exposed during childhood/adolescence. The effect estimates observed in our study for exposure to broadcast sprays before 1975 was similar to that associated with ever use of DDT in the agricultural cohort (HR=1.38).

Our findings are consistent with previous Sister Study analyses that linked early-life pesticide exposures to adult-onset systemic lupus erythematosus and rheumatoid arthritis (Parks et al., 2016, 2018). Similar to our study, these studies also found the highest hazards of disease among participants who personally used pesticides. Women who applied pesticides might have had a higher exposure burden from direct dermal contact with and inhalation of pesticide sprays compared to bystanders who were indirectly exposed. Although certain personal protective equipment (e.g., chemically resistant gloves and clothing, respirators) can reduce exposures during pesticide handling, only 8.7% and 5.0% of those who applied pesticides on crops and animals, respectively, reported using these protective items. We were underpowered to evaluate the effect of personal protective equipment use on IBD risk among pesticide users.

Previous studies have associated urban (vs. rural) living during childhood with a higher incidence of IBD (Song et al., 2019). Researchers attributed the observation to the “hygiene hypothesis”, where reduced microbial stimulation from urban residence and lifestyles during early life impairs the maturation of gut microbiota and immune system and increases the incidence of immune-mediated diseases (Song et al., 2019). In contrast, we did not find a difference in IBD incidence comparing those who lived in a rural area to those who lived in an urban area during childhood. We found slightly lower incidence among those who lived in small town or suburban areas. A comparison of characteristics among participants residing in different areas during childhood showed differences in distribution of age (youngest in suburban residents), race and ethnicity (lowest proportion of non-Hispanic White in urban residents), educational attainment (lowest in rural residents and highest in suburban residents), and smoking (lowest in rural residents) (Table S10). Differences in predictors of IBD, such as diet, infection, and access to healthcare (Piovani et al., 2019), that were not accounted for by existing covariates in the model might have explained the somewhat lower IBD incidence among small town and suburban residents and the lack of a positive association for urban vs. rural living.

In line with the hygiene hypothesis, early-life contact with livestock, which is associated with more opportunities for microbial exposures, has been shown to reduce the incidence of IBD in many epidemiologic studies (Cholapranee & Ananthakrishnan, 2016). In our study, contact with livestock without pesticide exposure was not associated with IBD in the main analysis and was associated with lower hazards of IBD in a sub-analysis of both confirmed and unconfirmed IBD diagnoses. We found elevated hazards of IBD among participants who personally applied pesticides to the livestock or their housing. This suggests that the adverse effects of pesticide exposure might have outweighed any benefit of contact with livestock during early life.

Our study has several strengths. It is the first to examine early-life pesticide exposure in relation to incidence of IBD. The prospective design allowed us to establish a clear temporal relationship between early-life exposure and incidence of IBD after enrollment. We examined several measures of early-life pesticide exposures, and associations were consistent across these exposure metrics. Lastly, detailed data on participant characteristics allowed us to account for several important potential confounders of the association (e.g., age, race and ethnicity, education, birth year) and predictors of IBD (e.g., cigarette smoking, antibiotic and hormonal birth control use, being breastfed, adulthood farm residence).

One limitation of the study is potential misclassification of the outcome, as we did not obtain medical records to confirm IBD diagnoses. Although a few studies have shown moderate to good agreement between self-reports and medical records in ascertaining IBD, these studies were conducted among less generalizable populations that either had a high prevalence of IBD (recruited from rosters of an organization dedicated to IBD education) (Randell et al., 2014) or were highly educated and more likely to accurately report medical diagnoses (O’Rourke et al., 2019). In our study, 277 out of 903 self-reported IBD diagnoses (30.7%) were confirmed by IBD medication use or colectomy/colostomy surgery. It is possible that some participants who reported a UC diagnosis had non-specific or other forms of colitis instead. On the other hand, the design of the DFU questionnaires (e.g., open-ended medication and surgery questions asked separately from IBD diagnoses) might have led to under-reporting in IBD medications or surgery. To reduce potential misclassification in IBD, we excluded participants with inconsistent reporting across the DFUs and further restricted the analysis to the 277 cases confirmed by medication or surgery. This resulted in wider confidence intervals in some analyses where the number of exposed cases is small. We found similar results in a sensitivity analysis of all IBD diagnoses, including the ones not confirmed by medication use or surgery, which suggests that misclassification in IBD (if any) is likely non-differential with respect to the exposures.

Another limitation is potential misclassification in exposures due to reliance on retrospective recall. In a validation study involving mothers of 1,802 Sister Study participants (D’Aloisio et al., unpublished), we found relatively high agreement between mothers’ and daughters’ self-reports, with sensitivity above 78% and specificity above 97% for farm exposures. This lends confidence to our self-reported exposure estimates. In addition, because incident IBD was assessed following participants’ reporting of exposures at enrollment, we do not expect the exposure misclassification to be differential. This may have biased effect estimates towards the null in analyses of binary exposures; the direction of bias is uncertain in analyses of polytomous exposures. While we were able to ascertain several measures of pesticide exposure, we lacked data on specific pesticides used in residences and farms to identify the individual pesticides that may contribute to the observed adverse associations. Future studies should investigate early-life exposure to specific pesticides in relation to IBD.

In the main analysis, we started follow-up at study enrollment and excluded IBD diagnoses before enrollment because participants with early IBD diagnoses might be less likely to enroll in the study. This also reduced concerns over temporality from recalling age at an early IBD diagnosis and early-life exposures. The incidence of IBD follows a bimodal age distribution, where the first peak occurs in the second or third decades of life, and the second smaller peak occurs in the sixth or seventh decade (Duricova et al., 2014; Hou et al., 2013). If participants with early-life pesticide exposures were more likely to develop IBD in young adulthood, our effect estimates might have underestimated the true effects. In a sensitivity analysis where we examined cases that occurred between exposures and end of follow-up, we found somewhat attenuated associations.

Although we accounted for several important potential confounders in our analysis, there could still be bias from unmeasured confounders. For instance, we did not collect and were thus unable to account for family history of IBD, which is a predictor of the outcome and may be related to residential and farm exposures if participants and their families resided in the same location.

Because the Sister Study consisted of women, our study findings may not be generalizable to men. While our study population was more racially and ethnically diverse than several other large U.S. cohorts, it still predominantly constituted non-Hispanic White participants, which limits generalizability of study findings to more racially and ethnically diverse populations. In a sub-analysis where we restricted to non-Hispanic White participants, we found similar associations; we had limited power to examine associations in other racial/ethnic groups.

Lastly, we conducted a sub-analysis for UC but had too few cases to examine CD alone. Although we expect early-life pesticide exposure would induce both UC and CD via the common dysbiosis and immune-dysregulating mechanisms, future studies with more cases can help ascertain the relationship of early-life pesticide exposure with both IBD subtypes.

In this study, we identified positive associations between incidence of IBD and early-life residential and farm pesticide exposures. We found the highest effect estimates among participants who personally applied pesticides, although those who were indirectly exposed (e.g., in crop field during pesticide spray) also had higher incidence of IBD. Together with a previous study that found multiple positive associations between individual pesticides and IBD in an agricultural cohort (Chen et al., 2024), our findings provide additional evidence for pesticide exposure as a contributor to IBD and further points to early life as a potential susceptible period for pesticide exposure. Practices that reduce pesticide exposure during early childhood and adolescence (e.g., avoid personal application of pesticides among children) may be especially helpful to reduce the burden of IBD.

Supplementary Material

1

Highlights.

  • The human immune system is susceptible to environmental exposures in early life.

  • We studied early-life pesticide use and inflammatory bowel disease (IBD) in women.

  • IBD was linked to residential pesticide use <age 14 and broadcast sprays.

  • Positive associations were found for pesticide use on crops and livestock <age 19.

  • Hazard ratios were highest among women who personally applied pesticides.

Funding:

This work was supported by the Intramural Research Program of the National Institutes of Health, National Institute of Environmental Health Sciences (Z01ES044005).

Abbreviations used in this paper:

CD

Crohn’s disease

CI

confidence interval

CPT

Current Procedural Terminology

DFU

detailed follow-up questionnaire

HR

hazard ratio

IBD

inflammatory bowel disease

UC

ulcerative colitis

U.S.

United States

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Declaration of competing interest: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data Availability:

Data are available upon request following the procedures described at the Sister Study website: https://sisterstudy.niehs.nih.gov/English/data-requests.htm.

References

  1. Agrawal M, Sabino J, Frias-Gomes C, Hillenbrand CM, Soudant C, Axelrad JE, Shah SC, Ribeiro-Mourão F, Lambin T, Peter I, Colombel JF, Narula N, & Torres J (2021). Early life exposures and the risk of inflammatory bowel disease: Systematic review and meta-analyses. EClinicalMedicine, 36, 100884. 10.1016/j.eclinm.2021.100884 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Argollo M, Gilardi D, Peyrin-Biroulet C, Chabot JF, Peyrin-Biroulet L, & Danese S (2019). Comorbidities in inflammatory bowel disease: a call for action. Lancet Gastroenterol Hepatol, 4(8), 643–654. 10.1016/s2468-1253(19)30173-6 [DOI] [PubMed] [Google Scholar]
  3. Belkaid Y, & Hand TW (2014). Role of the microbiota in immunity and inflammation. Cell, 157(1), 121–141. 10.1016/j.cell.2014.03.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Chen D, Grimsrud TK, Langseth H, Barr DB, Bassig BA, Blair A, Cantor KP, Gammon MD, Lan Q, Rothman N, & Engel LS (2020). Prediagnostic serum concentrations of organochlorine pesticides and non-Hodgkin lymphoma: A nested case-control study in the Norwegian Janus Serum Bank Cohort. Environ Res, 187, 109515. 10.1016/j.envres.2020.109515 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Chen D, Parks CG, Hofmann JN, Beane Freeman LE, & Sandler DP (2024). Pesticide use and inflammatory bowel disease in licensed pesticide applicators and spouses in the Agricultural Health Study. Environ Res, 249, 118464. 10.1016/j.envres.2024.118464 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Cholapranee A, & Ananthakrishnan AN (2016). Environmental Hygiene and Risk of Inflammatory Bowel Diseases: A Systematic Review and Meta-analysis. Inflamm Bowel Dis, 22(9), 2191–2199. 10.1097/mib.0000000000000852 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Cox DR (1972). REGRESSION MODELS AND LIFE-TABLES. JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY, 34(2), 187–187. http://unc.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwY2AwNtIz0EUrExJTjM1SzICN50TLFGAiMzJLTUwySk5OAmb7pJRU8LEFYU7GPt7mrk4mIUwMsDvHYENM4CqB9EEmxFgSbMgpoCgfPJCI3KvD2fOD3EBpbuRnag46xh1yqCioywRrdSEKfGPI9i5LXWAFbYS1DgPXV24CDHsRgzSQhSrgE6HykpEWcCOdBDnYAkCQgR_aUFZwhKRsIQam1DxhBi5Q2xhytLQIg2KQq3sQJOYVfP1dXH2CFRz9XBR8PN1cdUMcnXxcg0UZtN1cQ5w9dJEDKr4AcsBGPJqlRmIMvImg9f55JeB9gSkSDAqpxqmJhimmpmaWxibAFmFyYpppsmmqmaWFBVA4Lc1ckkGFGKMlGQyJUeYMPS8ddE5AiRRxRkszcIEEIMM1MgysacDiJFUW2ILJS5YDjyDIgVMRAPNw-iA [Google Scholar]
  8. Duricova D, Burisch J, Jess T, Gower-Rousseau C, & Lakatos PL (2014). Age-related differences in presentation and course of inflammatory bowel disease: an update on the population-based literature. J Crohns Colitis, 8(11), 1351–1361. 10.1016/j.crohns.2014.05.006 [DOI] [PubMed] [Google Scholar]
  9. Fathi MA, Han G, Kang R, Shen D, Shen J, & Li C (2020). Disruption of cytochrome P450 enzymes in the liver and small intestine in chicken embryos in ovo exposed to glyphosate. Environ Sci Pollut Res Int, 27(14), 16865–16875. 10.1007/s11356-020-08269-3 [DOI] [PubMed] [Google Scholar]
  10. GBD 2017 Inflammatory Bowel Disease Collaborators. (2020). The global, regional, and national burden of inflammatory bowel disease in 195 countries and territories, 1990–2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet Gastroenterol Hepatol, 5(1), 17–30. 10.1016/s2468-1253(19)30333-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Greenland S, Pearl J, & Robins JM (1999). Causal Diagrams for Epidemiologic Research. Epidemiology, 10(1), 37–48. 10.1097/00001648-199901000-00008 [DOI] [PubMed] [Google Scholar]
  12. Guardia-Escote L, Basaure P, Biosca-Brull J, Cabré M, Blanco J, Pérez-Fernández C, Sánchez-Santed F, Domingo JL, & Colomina MT (2020). APOE genotype and postnatal chlorpyrifos exposure modulate gut microbiota and cerebral short-chain fatty acids in preweaning mice. Food Chem Toxicol, 135, 110872. 10.1016/j.fct.2019.110872 [DOI] [PubMed] [Google Scholar]
  13. Hanauer SB (2006). Inflammatory bowel disease: epidemiology, pathogenesis, and therapeutic opportunities. Inflamm Bowel Dis, 12 Suppl 1, S3–9. 10.1097/01.mib.0000195385.19268.68 [DOI] [PubMed] [Google Scholar]
  14. Hou JK, Kramer JR, Richardson P, Mei M, & El-Serag HB (2013). The incidence and prevalence of inflammatory bowel disease among U.S. veterans: a national cohort study. Inflamm Bowel Dis, 19(5), 1059–1064. 10.1097/MIB.0b013e31828028ca [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Jin C, Zeng Z, Fu Z, & Jin Y (2016). Oral imazalil exposure induces gut microbiota dysbiosis and colonic inflammation in mice. Chemosphere, 160, 349–358. 10.1016/j.chemosphere.2016.06.105 [DOI] [PubMed] [Google Scholar]
  16. Jin Y, Liu Z, Peng T, & Fu Z (2015). The toxicity of chlorpyrifos on the early life stage of zebrafish: a survey on the endpoints at development, locomotor behavior, oxidative stress and immunotoxicity. Fish Shellfish Immunol, 43(2), 405–414. 10.1016/j.fsi.2015.01.010 [DOI] [PubMed] [Google Scholar]
  17. Joly Condette C, Bach V, Mayeur C, Gay-Quéheillard J, & Khorsi-Cauet H (2015). Chlorpyrifos Exposure During Perinatal Period Affects Intestinal Microbiota Associated With Delay of Maturation of Digestive Tract in Rats. J Pediatr Gastroenterol Nutr, 61(1), 30–40. 10.1097/mpg.0000000000000734 [DOI] [PubMed] [Google Scholar]
  18. Kamada N, Seo SU, Chen GY, & Núñez G (2013). Role of the gut microbiota in immunity and inflammatory disease. Nat Rev Immunol, 13(5), 321–335. 10.1038/nri3430 [DOI] [PubMed] [Google Scholar]
  19. Kelley K, Kelley T, Kaufman D, & Mitchell A (2003). The Slone Drug Dictionary: a research driven pharmacoepidemiology tool. Pharmacoepidemiol Drug Saf, 12(suppl 1), S168–S169. [Google Scholar]
  20. Knowles SR, Graff LA, Wilding H, Hewitt C, Keefer L, & Mikocka-Walus A (2018). Quality of Life in Inflammatory Bowel Disease: A Systematic Review and Meta-analyses-Part I. Inflamm Bowel Dis, 24(4), 742–751. 10.1093/ibd/izx100 [DOI] [PubMed] [Google Scholar]
  21. Lee GH, & Choi KC (2020). Adverse effects of pesticides on the functions of immune system. Comp Biochem Physiol C Toxicol Pharmacol, 235, 108789. 10.1016/j.cbpc.2020.108789 [DOI] [PubMed] [Google Scholar]
  22. Lichtenstein GR, Shahabi A, Seabury SA, Lakdawalla DN, Espinosa OD, Green S, Brauer M, & Baldassano RN (2020). Lifetime Economic Burden of Crohn’s Disease and Ulcerative Colitis by Age at Diagnosis. Clin Gastroenterol Hepatol, 18(4), 889–897.e810. 10.1016/j.cgh.2019.07.022 [DOI] [PubMed] [Google Scholar]
  23. Mao Q, Manservisi F, Panzacchi S, Mandrioli D, Menghetti I, Vornoli A, Bua L, Falcioni L, Lesseur C, Chen J, Belpoggi F, & Hu J (2018). The Ramazzini Institute 13-week pilot study on glyphosate and Roundup administered at human-equivalent dose to Sprague Dawley rats: effects on the microbiome. Environ Health, 17(1), 50. 10.1186/s12940-018-0394-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Meng Z, Sun W, Liu W, Wang Y, Jia M, Tian S, Chen X, Zhu W, & Zhou Z (2022). A common fungicide tebuconazole promotes colitis in mice via regulating gut microbiota. Environ Pollut, 292(Pt B), 118477. 10.1016/j.envpol.2021.118477 [DOI] [PubMed] [Google Scholar]
  25. Mokarizadeh A, Faryabi MR, Rezvanfar MA, & Abdollahi M (2015). A comprehensive review of pesticides and the immune dysregulation: mechanisms, evidence and consequences. Toxicol Mech Methods, 25(4), 258–278. 10.3109/15376516.2015.1020182 [DOI] [PubMed] [Google Scholar]
  26. O’Rourke JA, Ravichandran C, Howe YJ, Mullett JE, Keary CJ, Golas SB, Hureau AR, McCormick M, Chung J, Rose NR, & McDougle CJ (2019). Accuracy of self-reported history of autoimmune disease: A pilot study. PLoS One, 14(5), e0216526. 10.1371/journal.pone.0216526 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Okafor PN, Dahlen A, Youssef M, Olayode A, Sonu I, Neshatian L, Nguyen L, & Charu V (2022). Environmental Pollutants Are Associated With Irritable Bowel Syndrome in a Commercially Insured Cohort of California Residents. Clin Gastroenterol Hepatol. 10.1016/j.cgh.2022.09.025 [DOI] [PubMed] [Google Scholar]
  28. Parks CG, D’Aloisio AA, & Sandler DP (2016). Early Life Factors Associated with Adult-Onset Systemic Lupus Erythematosus in Women. Front Immunol, 7, 103. 10.3389/fimmu.2016.00103 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Parks CG, D’Aloisio AA, & Sandler DP (2018). Childhood Residential and Agricultural Pesticide Exposures in Relation to Adult-Onset Rheumatoid Arthritis in Women. Am J Epidemiol, 187(2), 214–223. 10.1093/aje/kwx224 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Piovani D, Danese S, Peyrin-Biroulet L, Nikolopoulos GK, Lytras T, & Bonovas S (2019). Environmental Risk Factors for Inflammatory Bowel Diseases: An Umbrella Review of Meta-analyses. Gastroenterology, 157(3), 647–659.e644. 10.1053/j.gastro.2019.04.016 [DOI] [PubMed] [Google Scholar]
  31. Qiu S, Fu H, Zhou R, Yang Z, Bai G, & Shi B (2020). Toxic effects of glyphosate on intestinal morphology, antioxidant capacity and barrier function in weaned piglets. Ecotoxicol Environ Saf, 187, 109846. 10.1016/j.ecoenv.2019.109846 [DOI] [PubMed] [Google Scholar]
  32. Randell RL, Long MD, Cook SF, Wrennall CE, Chen W, Martin CF, Anton K, Sandler RS, & Kappelman MD (2014). Validation of an internet-based cohort of inflammatory bowel disease (CCFA partners). Inflamm Bowel Dis, 20(3), 541–544. 10.1097/01.Mib.0000441348.32570.34 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Rodríguez JM, Murphy K, Stanton C, Ross RP, Kober OI, Juge N, Avershina E, Rudi K, Narbad A, Jenmalm MC, Marchesi JR, & Collado MC (2015). The composition of the gut microbiota throughout life, with an emphasis on early life. Microb Ecol Health Dis, 26, 26050. 10.3402/mehd.v26.26050 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Sandler DP, Hodgson ME, Deming-Halverson SL, Juras PS, D’Aloisio AA, Suarez LM, Kleeberger CA, Shore DL, DeRoo LA, Taylor JA, & Weinberg CR (2017). The Sister Study Cohort: Baseline Methods and Participant Characteristics. Environ Health Perspect, 125(12), 127003. 10.1289/ehp1923 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Sanmarco LM, Chao CC, Wang YC, Kenison JE, Li Z, Rone JM, Rejano-Gordillo CM, Polonio CM, Gutierrez-Vazquez C, Piester G, Plasencia A, Li L, Giovannoni F, Lee HG, Faust Akl C, Wheeler MA, Mascanfroni I, Jaronen M, Alsuwailm M, … Quintana FJ (2022). Identification of environmental factors that promote intestinal inflammation. Nature, 611(7937), 801–809. 10.1038/s41586-022-05308-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Song C, Yang J, Ye W, Zhang Y, Tang C, Li X, Zhou X, & Xie Y (2019). Urban-rural environmental exposure during childhood and subsequent risk of inflammatory bowel disease: a meta-analysis. Expert Rev Gastroenterol Hepatol, 13(6), 591–602. 10.1080/17474124.2018.1511425 [DOI] [PubMed] [Google Scholar]
  37. Tang J, Wang W, Jiang Y, & Chu W (2021). Diazinon exposure produces histological damage, oxidative stress, immune disorders and gut microbiota dysbiosis in crucian carp (Carassius auratus gibelio). Environ Pollut, 269, 116129. 10.1016/j.envpol.2020.116129 [DOI] [PubMed] [Google Scholar]
  38. Tang Q, Tang J, Ren X, & Li C (2020). Glyphosate exposure induces inflammatory responses in the small intestine and alters gut microbial composition in rats. Environ Pollut, 261, 114129. 10.1016/j.envpol.2020.114129 [DOI] [PubMed] [Google Scholar]
  39. Vafai Y, Yeung EH, Sundaram R, Smarr MM, Gerlanc N, Grobman WA, Skupski D, Chien EK, Hinkle SN, Newman RB, Wing DA, Ranzini AC, Sciscione A, Grewal J, Zhang C, & Grantz KL (2022). Racial/Ethnic Differences in Prenatal Supplement and Medication Use in Low-Risk Pregnant Women. Am J Perinatol, 39(6), 623–632. 10.1055/s-0040-1717097 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Zheng D, Liwinski T, & Elinav E (2020). Interaction between microbiota and immunity in health and disease. Cell Res, 30(6), 492–506. 10.1038/s41422-020-0332-7 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS2007497-supplement-1.docx (462.3KB, docx)

Data Availability Statement

Data are available upon request following the procedures described at the Sister Study website: https://sisterstudy.niehs.nih.gov/English/data-requests.htm.

RESOURCES