Skip to main content
NCBI home page
American Journal of Epidemiology logoLink to American Journal of Epidemiology
. 2024 Jul 16;194(1):44–55. doi: 10.1093/aje/kwae198

Preconception and first trimester exposure to pesticides and associations with stillbirth

Melissa A Furlong 1,✉, Kimberly C Paul 2, Kimberly L Parra 3, Alfred J Fournier 4, Peter C Ellsworth 5, Myles G Cockburn 6, Avelino F Arellano 7, Edward J Bedrick 8, Paloma I Beamer 9, Beate Ritz 10
PMCID: PMC12034837  PMID: 39013781

Abstract

Associations of pesticide exposures during preconception with stillbirth have not been well explored. We linked Arizona pesticide use records with birth certificates from 2006 to 2020 and estimated associations of living within 500 m of any pyrethroid, organophosphate (OP), or carbamate pesticide applications during a 90-day preconception window or the first trimester, with stillbirth. We considered a binary measure of exposure (any exposure), as well as log-pounds and log-acres applied within 500 m, in a negative control exposure framework with log-binomial regression. We included 1 237 750 births, 2290 stillbirths, and 27 pesticides. During preconception, any exposure to pesticides was associated with stillbirth, including cyfluthrin (risk ratio [RR] = 1.97; 95% CI, 1.17-3.32); zeta-cypermethrin (RR = 1.81; 95% CI, 1.20-2.74); OPs as a class (RR = 1.60; 95% CI, 1.16-2.19); malathion (RR = 2.02; 95% CI, 1.26-3.24); carbaryl (RR = 6.39; 95% CI, 2.07-19.74); and propamocarb hydrochloride (RR = 7.72; 95% CI, 1.10-54.20). During the first trimester, fenpropathrin (RR = 4.36; 95% CI, 1.09-17.50); permethrin (RR = 1.57; 95% CI, 1.02-2.42); OPs as a class (RR = 1.50; 95% CI, 1.11-2.01); acephate (RR = 2.31; 95% CI, 1.22-4.40); and formetanate hydrochloride (RR = 7.22; 95% CI, 1.03-50.58) were associated with stillbirth. Interpretations were consistent when using continuous measures of pounds or acres of exposure. Pesticide exposures during preconception and first trimester may be associated with stillbirth.

This article is part of a Special Collection on Environmental Epidemiology.

Keywords: pesticides, stillbirth, fetal death, organophosphate pesticides, pyrethroid pesticides, carbamate pesticides, environmental exposures, miscarriage

Introduction

The potential public health burden of pesticides is high because pesticides are designed for toxicity. Exposures to pesticides are ubiquitous, mostly through diet,1‑7 but also through household use, agricultural drift, occupational exposures, and para-occupational residue on shoes and clothes.8 The prenatal period is a particularly susceptible window of exposure because environmental exposures during this period affect lifelong health through fetal programming. Prenatal exposure to different classes of insecticides has been associated with several adverse birth and childhood outcomes.9‑16

Stillbirth, fetal death, and miscarriage present major physical and psychological health burdens for pregnant women.17 Stillbirth and miscarriages are types of fetal death, but stillbirths occur after gestational week 20 and miscarriages occur at or before week 20. Stillbirth occurs in approximately 4-5 births per 100018 in developed countries, and the rate is up to 3% of births in low- and middle-income countries.19 Stillbirth has been associated with some environmental exposures, including ambient temperature,20‑22 air pollution,23,24 and pesticides.25‑28 However, epidemiologic studies of pesticide exposures and stillbirth in the United States have not been published in the past 2 decades, a period that has seen dramatic changes in pesticide use compared with the 20th century.

The health effects of pesticides are usually studied with metabolite-based measures of exposure in prospective cohort studies. However, the rarity of stillbirth makes it difficult to study in the general population when using this approach, and biomarkers also are not typically available in a preconception setting. For organophosphate (OP) pesticides and pyrethroids, we generally also have used summary class measures, such as summed dialkylphosphates, which capture exposure to 75% of OP pesticides,29 or common metabolites like 3-phenoxybenzoic acid, which represents exposure to several pyrethroid pesticides but cannot identify the effects of a specific active ingredient. Finally, in a study of OP residues on fruits and vegetables, up to 60% of the OP residues were the metabolites, not the parent pesticide,30 so biomarkers reflect ingestion of both preformed metabolites and exposure to the parent ingredient.

One alternative to biomarker-based assessment includes the use of pesticide use registries (PURs). Although PURs do not capture the full range of pesticide exposures from personal behaviors (eg, diet), occupational use, tracked-in residue (hereafter referred to as track-in), or residential use, the ambient metric of pesticide exposure and the association with health outcomes is mostly unbiased by personal-level factors.31 Because some bias due to unmeasured confounding may remain from using these time-series data, we adopted a negative control exposure (NCE) framework, which has been successfully used in prior studies to account for residual bias.32‑37 The NCE is often operationalized by controlling for an exposure estimate that is a good proxy for the exposure but that cannot causally be associated with the outcome. One recommendation for such a variable in spatial and time-series data is to use exposures after the outcome has occurred. Here, we operationalize the NCE by controlling for exposure after the outcome (birth).

An additional advantage of using PURs instead of biomarkers is that it allows us to examine the preconception period as a susceptible window of exposure. This window is mostly unexplored in relation to pesticides, despite recent studies showing that other environmental exposures during this period may, indeed, affect fetal and child health.38‑46 Here, we used our unique study design to incorporate preconception pesticides into this study of stillbirth.

When PURs are combined with state-wide databases (eg, on reproductive outcomes), the result is enhanced power to study rare outcomes. Therefore, we evaluated associations of pyrethroids, OPs, and carbamate insecticides with stillbirth, using data from the Arizona Pregnant Women’s Environment and Reproductive Outcomes Study (Az-PEARS), a project that links Arizona’s PUR with birth certificates in the state of Arizona from 2006 to 2020. In Arizona, stillbirths are recorded on birth certificates.

Methods

Data were collected as part of the Az-PEARS, which links existing health data across multiple sources to enable population-wide epidemiologic analyses. Study protocols and data procedures were approved by the University of Arizona’s Institutional Review Board.

Birth certificates and study population

Arizona’s Department of Health Services provided geocoded birth certificates from 2006 to 2020. We restricted these to birth certificates that indicated maternal residential street address at birth in the state of Arizona that could be geocoded (eg, we excluded those that listed post office boxes or mile markers; n = 25 796), and we also excluded mothers older than 50 years (n = 1448), due to the probability of adoption or surrogacy. We extracted covariate, demographic, and outcome information from birth certificates, including maternal education, race/ethnicity, age, child sex, and fetal death/stillbirth. The state of Arizona requires that birth certificates be issued for stillbirths/fetal deaths for fetuses of more than 20-week gestations, and the birth certificates record these deaths. Hereafter, these outcomes are referred to as stillbirths.

Exposures

The state of Arizona requires that all commercial agricultural pesticide applications, including all aerial applications, be reported to the state. In addition, growers and applicators must report all soil-applied applications of pesticides on the Arizona Department of Environmental Quality’s groundwater protection list, as well as application of certain odiferous compounds.47 Arizona’s PUR, managed by the Arizona Department of Agriculture, reports information on active ingredient, acres, pounds applied, concentration, Public Land Survey Section (ie, location), and application method. The Public Land Survey Section grid corresponds to approximately a resolution of 1 square mile. To enhance this resolution, we linked reported pesticide applications to fields by linking the reported crop target on the PUR with crops from the US Department of Agriculture (USDA) CropScape rasters, a satellite-based rendering of crop identification, following methods previously used for the California PUR.48‑50 We matched targeted crops for the relevant chemical on the pesticide use report to the satellite-identified crop fields on the CropScape rasters, by year from 2006 to 2020. Because the USDA rasters are not available before 2008 in Arizona, we used the 2008 USDA crop rasters for 2006 and 2007 instead.

For this study, we examined the most commonly used insecticide classes over the past decades—OPs, pyrethroids, and carbamates—as well as all the specific pesticides that belonged to those classes. We limited pesticides to those ingredients that exposed at least 50 participants at the 500-m buffer during the entire study period. This buffer provides an optimal tradeoff for sensitivity and specificity of exposure.49 Mothers were defined as exposed if their residential address on the birth certificate was within 500 m of a given pesticide application during a specified trimester (preconception, trimester 0, or trimester 1). We limited to these trimesters because, in our data, the risk for stillbirth is highest in the second trimester (by definition, stillbirths in these data occur after 20 weeks of gestation). We evaluated 2 preconception periods that were 90 days long, to be consistent with the length of other trimesters, to allow for possible long-term effects of pesticide exposure and to account for possible errors around gestational dates. These periods included the window from 180 days to 91 days prior to conception (trimester 0, first window, which we indicate as T0.1), and the window from 90 days prior to conception, to conception (trimester 0, second window [T0.2]).

We considered a binary measure of exposure by trimester (eg, any exposure to a given pesticide during a specified trimester), as well as pounds of active ingredient and acres applied to in the 500 m buffer during the specified trimester. For acres and pounds, we replaced values of 0 with 0.1 × 10−15 divided by the square root of 2, which was approximately equal to the lowest detectable value divided by the square of 2, and logged the values. We also summed pounds and acres applied across the 90-day preconception window and the ~ 90 days of the first trimester, and we evaluated this total pounds and acres variable for this approximate 180--day exposure period, using the 180 days after the estimated due date as the NCE period for these models.

Statistical analysis and covariates

We describe demographics and key exposure characteristics by stillbirth status. We also describe correlations of pesticides across the first trimester and the preconception period, and correlations across included pesticides.

To evaluate the associations of individual pesticides with stillbirth, we used complete case analyses and log binomial regression. Because there may be residual confounding due to spatial correlations of events and exposures, we adopted an NCE framework33,34 using exposure in the 90 days after the estimated due date as the negative exposure. For each model’s NCE, we used the same ingredient for the specific pesticide or class as the exposure of interest, for the period after the due date. For example, when examining permethrin exposure in the preconception period, we controlled for permethrin exposure in the period after the due date. We additionally controlled for a priori selected covariates after a review of the literature, including maternal age, education, maternal race/ethnicity, child sex, season of conception, and year of conception.

We report both crude (unadjusted) and adjusted risk ratios (RRs). Because behaviors in the 90 days after the due date may be quite different from behaviors during the exposure window, we additionally evaluated the period from 90 days to 180 days after the estimated due date as an additional NCE window. We compared estimates from the 2 NCE models against a traditional model without the NCE variable. Additionally, because living in an agricultural region or in close proximity to crops may act as a confounder or modifier, due to higher probability of co-exposure to pesticides, or lifestyle variables with positive or negative effects on the outcome (eg, lower stress, more greenspace, rural living with less access to quality health care), we performed a sensitivity analysis restricting to births in agricultural regions only, and defined agricultural-region mothers as those who had been exposed during any trimester or preconception to any pesticide. We also performed sensitivity analyses evaluating modification of pesticide exposures by Medicaid status and by sex, and we set our α for interaction at .05. For these analyses, we evaluated interactions for the binary pesticide exposures during the first trimester and the 90-day window immediately preceding conception.

Results

Our study sample included 1 235 460 births, of which 2290 were stillborn. Overall, mothers were predominantly between ages 20 and 35 year, most had higher than a high school education, and most were non-Hispanic White or Hispanic (Table 1). Several characteristics differed by stillbirth; both older (>40 years) and younger (<20 years) mothers were more likely to experience stillbirth, and women with higher education were less likely to have a stillborn baby. Black women and Native American women were also at higher risk of stillbirth than non-Hispanic White or Hispanic women (Table 1).

Table 1.

Characteristics of Arizona Pregnant Women’s Environment and Reproductive Health Study (n = 1 237 750) by stillbirth status, 2006-2020.

Characteristic a No stillbirth (n = 1 235 460), No. (%) Stillbirth (n = 2,290), No. (%)
Maternal age, years
<20 85 822 (6.95) 215 (9.39)
20-35 999 672 (80.91) 1780 (77.73)
36-40 124 846 (10.11) 234 (10.22)
>40 25 120 (2.03) 61 (2.66)
Maternal education
Less than high school 717 358 (58.06) 1375 (60.04)
High school graduate 209 250 (16.94) 485 (21.18)
Some college 160 292 (12.97) 261 (11.40)
College or higher 148 560 (12.02) 169 (7.38)
Maternal race/ethnicity
Non-Hispanic White 548 750 (44.42) 808 (35.28)
Hispanic 532 219 (43.08) 1065 (46.51)
Black 66 599 (5.39) 247 (10.79)
Native American 57 598 (4.66) 110 (4.80)
Asian/Pacific Islander 27 692 (2.24) 57 (2.49)
Other/Unknown 2602 (0.21) 3 (0.13)
Pyrethroid exposure across pregnancy
None 1 163 452 (94.17) 2144 (93.62)
Any 72 007 (5.83) 146 (6.38)
Organophosphate exposure across pregnancy
None 1 182 958 (95.75) 2179 (95.15)
Any 52 501 (4.25) 111 (4.85)
Carbamate exposure across pregnancy
None 1 220 289 (98.77) 2261 (98.73)
Any 15 170 (1.23) 29 (1.27)
Open in a new tab

aMaternal age, education, and race/ethnicity are variables derived from birth certificate data from the Arizona Department of Health Services. Pyrethroids, organophosphate, and carbamate exposure are defined by whether or not mothers lived within 500 m of an agricultural application of pyrethroid, organophosphate, or carbamate pesticides during pregnancy or the 90-day preconception period, at the address recorded on the birth certificate.

During preconception, we observed several associations for pesticide exposures during the second preconception window (the 90 days immediately preceding conception) and the first trimester, but only 1 significant association for the first preconception window (the window from 180 days to 90 days prior to conception). For the binary metric of any applications in 500 m during the first preconception window, we report associations for the OP tribufos (RR = 2.66; 95% CI, 1.00-7.13), although this was based on only 4 exposed cases. During the second preconception window, we observed associations for the specific pyrethroids cyfluthrin (RR = 1.97; 95% CI, 1.17-3.32), zeta-cypermethrin (RR = 1.81; 95% CI, 1.20-2.74), bifenthrin (RR = 1.56; 95% CI, 0.97-2.49), and pyrethroids as a class (RR = 1.27; 95% CI, 0.95-1.71), although bifenthrin and pyrethroids as a class did not meet statistical significance (Table 2). We also observed elevated but nonsignificant associations for cypermethrin and fenpropathrin, which both had very few exposed cases (Table 2).

Table 2.

Crude and adjusted associationsa of a binary measure of insecticides with fetal death, by exposure window in Arizona Pregnant Women’s Environment and Reproductive Outcomes Study, n = 1 237 750.

Active ingredient Exposure status No. alive at T0.1 Case T0.1, No. No. alive at T0.2 Case T0.2, No. No. alive at T1 Case T1, No. Model T0.1 RR (95% CI) T0.2 RR (95% CI) T1 RR (95% CI)
Pyrethroids and pyrethrins
Pyrethroids and pyrethrins as a class Unexposed 1 207 124 2226 1 206 448 2216 1 204 549 2218 Crude 1.22 (0.96-1.57) 1.39 (1.10-1.75) 1.26 (1.00-1.60)
Exposed 28 335 64 29 011 74 30 910 72 Adjusted 1.08 (0.81-1.42) 1.27 (0.95-1.71) 1.15 (0.88-1.50)
Beta-cyfluthrin Unexposed 1 230 145 2279 1 230 043 2276 1 229 500 2279 Crude 1.12 (0.62-2.02) 1.40 (0.83-2.36) 1.00 (0.55-1.80)
Exposed 5 315 11 5 417 14 5 960 11 Adjusted 1.08 (0.60-1.96) 1.30 (0.75-2.24) 0.96 (0.53-1.76)
Bifenthrin Unexposed 1 226 698 2269 1 226 654 2263 1 226 452 2269 Crude 1.30 (0.84-1.99) 1.66 (1.14-2.42) 1.26 (0.82-1.93)
Exposed 8762 21 8806 27 9008 21 Adjusted 1.07 (0.66-1.74) 1.56 (0.97-2.49) 1.07 (0.65-1.76)
Cyfluthrin Unexposed 1 230 448 2277 1 230 182 2271 1 229 960 2277 Crude 1.40 (0.81-2.41) 1.95 (1.24-3.06) 1.28 (0.74-2.20)
Exposed 5012 13 5278 19 5500 13 Adjusted 1.28 (0.73-2.27) 1.97 (1.17-3.32) 1.22 (0.68-2.20)
Cypermethrin Unexposed 1 234 740 2290 1 234 689 2287 1 234 695 2288 Crude NE 2.10 (0.68-6.49) 1.41 (0.35-5.63)
Exposed 720 0 771 3 765 2 Adjusted NE 2.47 (0.80-7.65) 1.58 (0.39-6.30)
Esfenvalerate Unexposed 1 231 949 2278 1 231 856 2280 1 231 822 2280 Crude 1.85 (1.05-3.25) 1.50 (0.81-2.79) 1.48 (0.80-2.76)
Exposed 3511 12 3604 10 3638 10 Adjusted 1.49 (0.82-2.72) 1.13 (0.55-2.32) 1.24 (0.64-2.40)
Fenpropathrin Unexposed 1 235 247 2289 1 235 244 2289 1 235 210 2288 Crude 2.53 (0.36-17.86) 2.49 (0.35-17.62) 4.29 (1.08-17.08)
Exposed 213 1 216 1 250 2 Adjusted 2.52 (0.35-17.83) 2.45 (0.34-17.84) 4.36 (1.09-17.50)
Gamma-cyhalothrin Unexposed 1 235 451 2290 1 235 446 2290 1 235 455 2290 Crude NE NE NE
Exposed 9 0 14 0 5 0 Adjusted NE NE NE
Lambda-cyhalothrin Unexposed 1 221 451 2253 1 220 906 2258 1 220 189 2250 Crude 1.43 (1.03-1.98) 1.19 (0.84-1.68) 1.42 (1.04-1.94)
Exposed 14 009 37 14 554 32 15 271 40 Adjusted 1.30 (0.92-1.84) 1.03 (0.69-1.55) 1.33 (0.95-1.85)
Permethrin Unexposed 1 227 013 2271 1 226 499 2270 1 226 321 2262 Crude 1.21 (0.77-1.91) 1.21 (0.78-1.87) 1.66 (1.14-2.41)
Exposed 8447 19 8961 20 9139 28 Adjusted 0.99 (0.60-1.64) 0.95 (0.54-1.67) 1.57 (1.02-2.42)
Pyrethrins Unexposed 1 233 861 2290 1 233 806 2289 1 233 725 2286 Crude NE 0.33 (0.05-2.30) 1.24 (0.47-3.31)
Exposed 1599 0 1654 1 1735 4 Adjusted NE 0.26 (0.03-2.02) 1.24 (0.44-3.48)
Zeta-cypermethrin Unexposed 1 224 234 2263 1 223 901 2253 1 223 444 2260 Crude 1.30 (0.89-1.90) 1.74 (1.26-2.40) 1.35 (0.94-1.94)
Exposed 11 226 27 11 559 37 12 016 30 Adjusted 1.14 (0.75-1.73) 1.81 (1.20-2.74) 1.25 (0.82-1.90)
Organophosphates
Organophosphates as a class Unexposed 1 217 496 2249 1 217 247 2239 1 216 248 2238 Crude 1.24 (0.91-1.68) 1.52 (1.15-2.01) 1.47 (1.12-1.93)
Exposed 17 963 41 18 212 51 19 211 52 Adjusted 1.19 (0.86-1.64) 1.60 (1.16-2.19) 1.50 (1.11-2.01)
Acephate Unexposed 1 233 028 2283 1 233 142 2283 1 233 059 2280 Crude 1.55 (0.74-3.26) 1.63 (0.78-3.42) 2.25 (1.21-4.18)
Exposed 2432 7 2318 7 2401 10 Adjusted 1.49 (0.70-3.19) 1.65 (0.75-3.65) 2.31 (1.22-4.40)
Bensulide Unexposed 1 229 141 2273 1 228 874 2273 1 229 084 2275 Crude 1.45 (0.90-2.34) 1.39 (0.87-2.25) 1.27 (0.77-2.11)
Exposed 6319 17 6586 17 6376 15 Adjusted 1.30 (0.79-2.16) 1.38 (0.76-2.50) 1.20 (0.69-2.10)
Chlorpyrifos Unexposed 1 232 123 2283 1 232 103 2283 1 231 735 2279 Crude 1.13 (0.54-2.38) 1.13 (0.54-2.36) 1.59 (0.88-2.88)
Exposed 3337 7 3357 7 3725 11 Adjusted 1.16 (0.55-2.44) 1.16 (0.54-2.49) 1.65 (0.90-3.01)
Diazinon Unexposed 1 234 592 2288 1 234 499 2288 1 234 523 2289 Crude 1.24 (0.31-4.96) 1.12 (0.28-4.49) 0.58 (0.08-4.09)
Exposed 868 2 961 2 937 1 Adjusted 1.13 (0.28-4.58) 0.98 (0.23-4.22) 0.55 (0.08-3.97)
Dimethoate Unexposed 1 231 933 2284 1 231 852 2278 1 231 457 2277 Crude 0.92 (0.41-2.04) 1.80 (1.02-3.16) 1.75 (1.02-3.02)
Exposed 3527 6 3608 12 4003 13 Adjusted 0.86 (0.38-1.92) 1.56 (0.86-2.84) 1.63 (0.94-2.84)
Ethephon Unexposed 1 234 578 2286 1 234 589 2287 1 234 641 2288 Crude 2.44 (0.92-6.50) 1.86 (0.60-5.75) 1.32 (0.33-5.26)
Exposed 882 4 871 3 819 2 Adjusted 2.14 (0.80-5.71) 1.06 (0.31-3.63) 1.13 (0.28-4.54)
Malathion Unexposed 1 230 011 2275 1 229 856 2272 1 229 034 2276 Crude 1.49 (0.90-2.47) 1.74 (1.09-2.76) 1.18 (0.70-1.99)
Exposed 5449 15 5604 18 6426 14 Adjusted 1.58 (0.95-2.62) 2.02 (1.26-3.24) 1.27 (0.75-2.14)
Oxydemeton-methyl Unexposed 1 235 454 2290 1 235 456 2290 1 235 454 2290 Crude NE NE NE
Exposed 6 0 4 0 6 0 Adjusted NE NE NE
Phorate Unexposed 1 235 264 2290 1 235 215 2290 1 235 164 2290 Crude NE NE NE
Exposed 196 0 245 0 296 0 Adjusted NE NE NE
Tribufos Unexposed 1 234 797 2286 1 234 824 2288 1 234 813 2287 Crude 3.25 (1.22-8.63) 1.69 (0.42-6.77) 2.50 (0.81-7.73)
Exposed 663 4 636 2 647 3 Adjusted 2.66 (1.00-7.13) 0.84 (0.19-3.65) 2.16 (0.69-6.71)
Carbamates
Carbamates as a class Unexposed 1 229 987 2278 1 229 600 2280 1 229 299 2279 Crude 1.18 (0.67-2.09) 0.92 (0.49-1.71) 0.96 (0.53-1.74)
Exposed 5472 12 5859 10 6160 11 Adjusted 1.26 (0.69-2.29) 1.04 (0.51-2.11) 1.08 (0.57-2.04)
Carbaryl Unexposed 1 235 144 2290 1 235 176 2287 1 235 113 2290 Crude NE 5.66 (1.83-17.45) NE
Exposed 316 0 284 3 347 0 Adjusted NE 6.39 (2.07-19.74) NE
Carbofuran Unexposed 1 235 433 2290 1 235 432 2290 1 235 437 2290 Crude NE NE NE
Exposed 27 0 28 0 23 0 Adjusted NE NE NE
Methomyl Unexposed 1 230 387 2279 1 229 961 2280 1 229 751 2280 Crude 1.17 (0.65-2.11) 0.98 (0.53-1.82) 0.94 (0.51-1.76)
Exposed 5073 11 5499 10 5709 10 Adjusted 1.21 (0.65-2.27) 1.09 (0.53-2.23) 1.03 (0.53-2.02)
Oxamyl Unexposed 1 235 340 2289 1 235 354 2290 1 235 340 2290 Crude 4.47 (0.63-31.48) NE NE
Exposed 120 1 106 0 120 0 Adjusted 4.87 (0.69-34.31) NE NE
Propamocarb hydrochloride Unexposed 1 235 380 2290 1 235 390 2289 1 235 362 2289 Crude NE 7.62 (1.09-53.34) 5.46 (0.78-38.41)
Exposed 80 0 70 1 98 1 Adjusted NE 7.72 (1.10-54.20) 5.80 (0.82-40.89)
Formetanate hydrochloride Unexposed 1 235 407 2290 1 235 400 2290 1 235 394 2289 Crude NE NE 8.07 (1.15-56.48)
Exposed 53 0 60 0 66 1 Adjusted NE NE 7.22 (1.03-50.58)
Open in a new tab

Abbreviations: NE, not estimable, due to zero exposed deaths; T0.1, the exposure window beginning 180 days before preconception and ending 91 days before conception; T0.2, the 90-day preconception window beginning 90 days preconception and ending the day before conception; T1, first trimester.

a These models are from binary exposure models (any exposure in preconception periods, any exposure in first trimester exposure). Associations were estimated with log binomial regression models in 1 237 750 births from 2006 to 2020 in Arizona. Adjusted models were adjusted for using negative control exposure models, including maternal race/ethnicity, child sex, maternal education, maternal age, conception year, season of conception, and exposure to the modeled pesticide in the 90 days after the estimated due date. Crude models are unadjusted for covariates. Models with statistically significant adjusted estimates (P < .05) are in bold for visualization purposes.

For the OPs, we observed associations with any exposure to OPs as a class (RR = 1.60; 95% CI, 1.16-2.19) and malathion (RR = 2.02; 95% CI, 1.26-3.24), along with elevated but nonsignificant associations for dimethoate (RR = 1.56; 995% CI, 0.86-2.84). The carbamates carbaryl (RR = 6.39; 95% CI, 2.07-19.74) and propamocarb hydrochloride (RR = 7.72; 95% CI, 1.10-54.20) were also associated during this window. During the first trimester, the pyrethroids fenpropathrin (RR = 4.36; 95% CI, 1.09-17.50) and permethrin (RR = 1.57; 95% CI, 1.02-2.42), OPs as a class (RR = 1.50; 95% CI, 1.11, 2.01), and the specific OPs acephate (RR = 2.31; 95% CI, 1.22-4.40) and dimethoate (RR 1.63; 95% CI, 0.94-2.84), and the carbamate formetanate hydrochloride (RR = 7.22; 95% CI, 1.03-50.58) were associated with stillbirth, although the finding for dimethoate was not statistically significant.

Interpretations were similar when evaluating associations with log acres and log pounds (Figure 1). Summing the exposures across the entire early pregnancy period (ie, the 90 days preconception and the first trimester) did not strengthen effects but appeared to be an average of the window-specific associations. In general, acres appeared to be slightly more sensitive than pounds for cyfluthrin, fenpropathrin, and tribufos, although the differences were mostly negligible.

Figure 1.

Figure 1

Open in a new tab

Adjusted risk ratios (RRs) of exposure to continuous measures of pesticides during preconception and first trimester with stillbirth in the Arizona Pregnant Women’s Environment and Reproduction Study (n = 1 237 750). Associations estimated with log binomial regression models in 1 237 750 births from 2006 to 2020 in Arizona, adjusted for maternal race/ethnicity, child sex, maternal education, maternal age, conception year, conception season, and exposure in the 90 days after the estimated due date (for the negative control exposure). Acres and pounds were logged. T0.1, the exposure window beginning 180 days before preconception and ending 91 days preconception; T0.2, the 90-day preconception window beginning 90 days preconception and ending the day before conception; T1, first trimester. RRs and CIs for some time windows and pesticides are missing because not all associations were estimable, due to zero exposed deaths.

Pesticides were not highly correlated across classes or active ingredients, with low to moderate correlations (Figure 2). Exposure to different pyrethroid active ingredients were moderately correlated; for instance, zeta-cypermethrin and cyfluthrin were correlated at 0.44 for the first trimester, and permethrin and zeta cypermethrin were correlated at 0.55 for the preconception period. However, correlations among specific OPs were very low. The class of OPs, as a whole, was moderately correlated with the class of pyrethroids as a whole, at 0.62 and 0.61, respectively, for the first trimester and the preconception period.

Figure 2.

Figure 2

Open in a new tab

Correlations of pesticides across exposure windows and active ingredients. Left panel shows correlations (Corr) for classes, and the right panel is for select specific insecticides. OP, organophosphate; Pyr, pyrethroid; T0, the preconception period (1-90 days before conception); T1, exposure during the first trimester; T4, the negative control exposure variable (0-89 days after the estimated due date).

Associations were generally not affected by the use of the NCE framework or the selection of the NCE window (Figure 3), although variability did affect significance for exposures with confidence limits that were close to the null.

Figure 3.

Figure 3

Open in a new tab

Associations of pesticide exposure by trimester, comparing negative control exposure windows. These models are from binary exposure models (any exposure in the preconception window 90 days before birth (T0.2) and any exposure in first trimester (T1). Associations are estimated with log binomial regression models in 1 237 750 births from 2006 to 2020 in Arizona, adjusted for maternal race/ethnicity, child sex, maternal education, maternal age, conception year, and conception season. “Traditional model” refers to the model with no negative control exposure (NCE). T4.1, time from the due date to 89 days later; T4.2, time beginning 90 days after the due date and ending 180 days after the due date. For visualization purposes, associations with very wide CIs are not shown.

Effect estimates were similar when restricting to agricultural regions only, although SEs and P values were greater, likely because of the decrease in sample size (Tabel S1).

In sensitivity analyses, we generally did not observe any interactions with sex or Medicaid status at birth, with 2 exceptions. There was an interaction between malathion in the second preconception window (90 days prior to conception) with Medicaid status. Malathion was positively associated with stillbirth among women who were not enrolled in Medicaid (RR = 2.99 [95% CI, 1.76-5.10]; P = 0.046 for interaction), and null for women who were enrolled in Medicaid (RR = 0.94; 95% CI, 0.35-2.55). First-trimester acephate effect estimates were stronger among female fetuses (RR = 4.03; 95% CI, 1.39-8.22) than male fetuses (RR = 0.85 [95% CI, 0.21-3.45]; P = 0.049 for interaction).

Discussion

In this study, we linked 15 years of data from a PUR enhanced with satellite-based crop linkage in Arizona with more than 1 million births registered on Arizona birth certificates, to examine associations of exposure to specific pesticides during early pregnancy and the preconception period with stillbirth. We report signals for exposure during the 90 days prior to conception or the first trimester for multiple pyrethroid pesticides (namely, cyfluthrin, fenpropathrin, permethrin, and zeta-cypermethrin), multiple OPs (ie, any OP, acephate, malathion, and dimethoate), and the carbamates carbaryl, propamocarb hydrochloride, and formetanate hydrochloride. A few other pesticides had elevated effect estimates, but these were not statistically significant, including for bifenthrin, cypermethrin, chlorpyrifos, and dimethoate. We generally did not observe associations for exposures during an earlier preconception window (3-6 months prior to conception), except for tribufos.

The literature on associations of these insecticides with stillbirth is sparse.51 In occupational studies of women and their partners who work in agriculture, pesticide exposures are associated with increased risk of stillbirth.52‑54 Other studies of specific pesticides or pesticide classes in humans report associations of pyrethroids with decreased fertility and suggestive associations for OPs and carbamates as a class,55 although another smaller study during the same period reported null associations.56 In a California study of fetal deaths in 1984, exposure to carbamates during the third month of gestation was associated with fetal death,26 which is consistent with our observations during early pregnancy for carbaryl, formetanate hydrochloride, and propamocarb hydrochloride, although these findings were only based on fewer than 3 exposed cases for each pesticide. In the same 1984 study, associations for pyrethroids and OPs as a class were null, although this period was prior to when pyrethroids were a commonly used class. A study of malathion in 1982 in California for mosquito control found no association between malathion and fetal anomalies.57 A more recent study in Brazil found that exposure to pesticides in general was associated with increased odds of stillbirth,27 although the researchers did not report on specific active ingredients. In a toxicology study, the pyrethroid cyfluthrin altered placental development,58 and disrupted placental development is hypothesized to be an important cause of stillbirths and miscarriages.59 Chlorpyrifos in mice has been associated with postimplantation loss and early neonatal death in mice and rats,60‑62 but not specifically stillbirth. We generally did not observe strong signals for chlorpyrifos, other than an elevated but not significant association in mothers who were exposed during the first trimester.

Our findings suggest that the preconception period, and to some degree the first trimester, may be important windows for exposure to some insecticides. This may be due to biological effects on the female reproductive organs,63 changes in hormonal activity,64 or other unknown biological factors. Metabolites of permethrin and cypermethrin (which are shared with zeta-cypermethrin) act as endocrine disruptors and interact with cellular estrogen receptors.63 Such hormonal influences can affect women’s reproductive cycles and cycle lengths, and the overall quality of the uterine environment during pre-implantation.65,66 Although pyrethroids are rapidly excreted, with half-lives on the order of hours to days,67‑70 exposure can result in apoptotic and senescent cells,71 and environmentally induced senescent cells may persist in the human body for months, with long-term expression of senescent markers,72,73 even after the environmental exposure has ended.

Uterine and reproductive tract cells form the architecture for a successful pregnancy, and past exposure may thus continue to exert influence on pregnancy and placental health. Alternatively, associations during this exposure window may be due to possible confounding by effects on father’s sperm, which we were unable to account for. Because we assigned exposure by residence, exposure is likely shared with fathers who live in the same residence. A study of malathion and male reproductive parameters in mice showed toxic effects of malathion on sperm count, progressive motility, morphology, and viability; and testosterone.74 Paternal exposure to pyrethroids causes increased fetal loss among rats.75

Differences by exposure window may be due to miscalculation of exposure timing, particularly for fetuses with uncertain gestational ages, and exposure attributed to preconception may actually have happened in the first trimester. Differences between the 2 preconception windows were somewhat stark, with almost no associations for exposures in the first preconception window (beginning 180 days prior to conception and ending 91 days prior to conception). If causal, exposures closer to conception are driving the elevated risk, which is biologically consistent with the rapid metabolism and clearance of modern pesticides. There may also be some survival bias in our estimates, because we cannot account for early miscarriages. It is possible that pesticide exposures in the preconception period and the first trimester contribute to miscarriage, in addition to stillbirth, and that miscarriage and stillbirth may share biological mechanisms influenced by pesticides. Thus, our estimates are likely an underestimate of the true effect of pesticides on pregnancy loss.

Generally, we did not observe differences in RRs when restricting to agricultural regions, and we found minimal differences between the different NCE models, suggesting minimal residual confounding. We also did not observe substantial modification by sex or by Medicaid status.

Regardless, if causal, this exposure window presents unique problems for prevention and education, particularly because 50% of pregnancies are unplanned. From a policy perspective, buffer zones between fields sprayed with pesticides and residential and educational buildings could be required and/or expanded. Policies that promote integrated pest management may reduce risks from pests and pesticides through integration of biological and chemical controls.76‑78 For example, in Arizona cotton,79,80 10-fold reductions in use of OPs, pyrethroids, and carbamates were observed between 1995 and 2006, a period when integrated pest management was implemented and embraced. Behavioral modifications may also reduce exposures to agricultural pesticide use. For instance, changing air filters or increasing vacuuming frequency may reduce pesticide concentrations in household indoor air and dust.81,82

Other exposure pathways may be important for nonrural populations. For instance, malathion and some pyrethroids are regularly sprayed via fogging machines in urban and suburban areas in the United States and internationally to control mosquito populations that can carry West Nile virus, although mosquito control can be effectively achieved with better urban planning and landscape design.83 Permethrin is recommended by the World Health Organization for disinfecting aircraft in countries with endemic malaria and other insect-borne diseases. Permethrin is also the most commonly used ingredient in insecticide-treated clothing, and all US Department of Defense uniforms are required to be treated with permethrin. Soldiers who wear these uniforms have 30 to 50 times higher levels of pyrethroid metabolites than the general population.84,85 The Department of Defense should continue exploring alternative approaches to reduce vector-borne diseases.

Strengths and limitations

This study has several strengths. The study design allowed us to investigate associations on a population-wide basis without restricting to those actively participating in a research study, thus potentially including people who may be harder to reach in traditional birth cohort studies. The use of a pesticide registry instead of biomarkers allowed us to estimate exposures across multiple windows of exposure for more than 1 million people, as opposed to biomarker-based studies, which typically reflect exposure for a short period (eg, hours to days) and are typically only collected 1-3 times during pregnancy in small cohort studies. Biomarkers may also be specific to a particular pathological pathway, whereas the PUR method is not. We also were able to identify agricultural applications as the source for exposure, which may have implications for agricultural-specific policy (eg, implementation of buffers between fields and housing developments, limiting use of specific pesticides, lowering food tolerances), whereas biomarkers reflect exposure from multiple sources, which may be more difficult to regulate.

The study also had limitations. We were limited to OPs, pyrethroids, and carbamates that we could study with adequate statistical power; thus, we could not estimate some specific OP, pyrethroid, and carbamate effect estimates. We conducted several tests and might have experienced some Type 1 error, although findings should all be replicated in other populations and study designs regardless. We relied on birth certificates that may be less reliable than medical records, generating some outcome misclassification, and possibly exposure misclassification due to misreported addresses or women not staying at the address reported on the birth certificate. However, all stillbirths after 20 weeks’ gestation in Arizona are required to be issued a birth certificate and are presumed to be correctly recorded. Additionally, the tradeoff between precision and quantity of data is a commonly encountered issue in large-scale health databases. Because we used PURs, we also do not know if participants were exposed to pesticide drift; because there is likely some misreporting, the field-crop matching is likely inexact, and geological features (eg, hills, large buildings) may affect some people’s exposures relative to others at the same buffer. Some people who are defined as unexposed may also be exposed to some small number of grower-applied pesticides that are not in the PUR; that reporting is not required under the state statute. Although some women likely moved during pregnancy, movement is probably nondifferential for case and control participants, and prior studies have suggested that most moves are short distances and exposure estimates are minimally affected.86‑88 Still, we likely experienced residual nondifferential exposure misclassification and thus report effect estimates that are most likely biased toward the null.

It is also unclear why different specific pesticides within the class are associated with stillbirth, and not others. Although these pesticides vary in their chemical properties (eg, log of the n-octanol/water partition coefficient, Henry’s Law constant, vapor pressure, soil half-life), there is no apparent relationship between these characteristics and the strength of association with stillbirth. We did not account for mixtures, which may be important, and future research will use mixtures modeling techniques. Finally, our exposure measurements do not reflect wind, atmospheric conditions, or resuspension. Efforts are underway to develop atmospheric dispersion models, which will generate continuous measures of air concentrations and deposition, which may have more relevance for monitoring and regulating agricultural applications of pesticides.

Conclusions

Exposure during preconception or early pregnancy to certain OPs, pyrethroid, and carbamate pesticides may be associated with a higher risk of stillbirth.

Acknowledgments

Results from this study were originally presented at the International Society of Environmental Epidemiology. We thank the Arizona Department of Health Services and the University of Arizona for making this study possible.

Supplementary Material

Web_Material_kwae198
web_material_kwae198.docx (30.4KB, docx)

Contributor Information

Melissa A Furlong, Department of Community, Environment, and Policy, Environmental Health Sciences, University of Arizona, Tucson, Arizona, United States.

Kimberly C Paul, Department of Neurology, University of California at Los Angeles, Los Angeles, California, United States.

Kimberly L Parra, Department of Epidemiology & Biostatistics, University of Arizona College of Public Health, Tucson, Arizona, United States.

Alfred J Fournier, Department of Entomology, University of Arizona College of Agricultural and Life Sciences, Tucson, Arizona, United States.

Peter C Ellsworth, Department of Entomology, University of Arizona College of Agricultural and Life Sciences, Tucson, Arizona, United States.

Myles G Cockburn, Department of Community Medicine, University of Southern California, Los Angeles, California, United States.

Avelino F Arellano, Department of Hydrology and Atmospheric Sciences, University of Arizona College of Science, Tucson, Arizona, United States.

Edward J Bedrick, Department of Epidemiology & Biostatistics, University of Arizona College of Public Health, Tucson, Arizona, United States.

Paloma I Beamer, Department of Community, Environment, and Policy, Environmental Health Sciences, University of Arizona, Tucson, Arizona, United States.

Beate Ritz, Department of Epidemiology, University of California at Los Angeles, Los Angeles, California, United States.

Supplementary material

Supplementary material is available at the American Journal of Epidemiology online.

Funding

This work was supported by the National Institutes of Health National Institute of Environmental Health Sciences (grants R00-ES028743 and P30 ES006694) and National Institute for Occupational Safety and Health, Western Center for Agricultural Health and Safety (grant U54OH007550 to the University of Arizona [account no. 4019490]).

Conflict of interest

P.C.E. has received research grants and contracts from pesticide registrants unrelated to the research reported here. The remaining authors declared no conflicts of interest.

Disclaimer

The National Institutes of Health (NIH) does not endorse or recommend any commercial products, processes, or services. The views and opinions of authors expressed on NIH websites do not necessarily state or reflect those of the US Government, and they may not be used for advertising or product endorsement purposes.

Data availability

The data used for this study are available upon request to the corresponding author.

References

  • 1. Oates  L, Cohen  M, Braun  L, et al.  Reduction in urinary organophosphate pesticide metabolites in adults after a week-long organic diet. Environ Res. 2014;132:105-111. 10.1016/j.envres.2014.03.021 [DOI] [PubMed] [Google Scholar]
  • 2. Curl  CL, Fenske  RA, Elgethun  K. Organophosphorus pesticide exposure of urban and suburban preschool children with organic and conventional diets. Environ Health Perspect. 2003;111(3):377-382. 10.1289/ehp.5754 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Lu  C, Toepel  K, Irish  R, et al.  Organic diets significantly lower children’s dietary exposure to organophosphorus pesticides. Environ Health Perspect. 2006;114(2):260-263. 10.1289/ehp.8418 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Hyland  C, Bradman  A, Gerona  R, et al.  Organic diet intervention significantly reduces urinary pesticide levels in US children and adults. Environ Res. 2019;171:568-575. 10.1016/j.envres.2019.01.024 [DOI] [PubMed] [Google Scholar]
  • 5. Bradman  A, Quirós-Alcalá  L, Castorina  R, et al.  Effect of organic diet intervention on pesticide exposures in young children living in low-income urban and agricultural communities. Environ Health Perspect. 2015;123(10):1086-1093. 10.1289/ehp.1408660 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Lu  C, Barr  DB, Pearson  MA, et al.  Dietary intake and its contribution to longitudinal organophosphorus pesticide exposure in urban/suburban children. Environ Health Perspect. 2008;116(4):537-542. 10.1289/ehp.10912 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Ye  M, Beach  J, Martin  JW, et al.  Associations between dietary factors and urinary concentrations of organophosphate and pyrethroid metabolites in a Canadian general population. Int J Hyg Environ Health. 2015;218(7):616-626. 10.1016/j.ijheh.2015.06.006 [DOI] [PubMed] [Google Scholar]
  • 8. Deziel  NC, Friesen  MC, Hoppin  JA, et al.  A review of nonoccupational pathways for pesticide exposure in women living in agricultural areas. Environ Health Perspect. 2015;123(6):515-524. 10.1289/ehp.1408273 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Wang  N, Huang  M, Guo  X, et al.  Urinary metabolites of organophosphate and pyrethroid pesticides and neurobehavioral effects in Chinese children. Environ Sci Technol. 2016;50(17):9627-9635. 10.1021/acs.est.6b01219 [DOI] [PubMed] [Google Scholar]
  • 10. Oulhote  Y, Bouchard  MF. Urinary metabolites of organophosphate and pyrethroid pesticides and behavioral problems in Canadian children. Environ Health Perspect (Online). 2013;121(11-12):1378-1384. 10.1289/ehp.1306667 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Engel  SM, Wetmur  J, Chen  J, et al.  Prenatal exposure to organophosphates, paraoxonase 1, and cognitive development in childhood. Environ Health Perspect. 2011;119(8):1182-1188. 10.1289/ehp.1003183 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Furlong  MA, Engel  SM, Barr  DB, et al.  Prenatal exposure to organophosphate pesticides and reciprocal social behavior in childhood. Environ Int. 2014;70:125-131. 10.1016/j.envint.2014.05.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Bouchard  MF, Chevrier  J, Harley  KG, et al.  Prenatal exposure to organophosphate pesticides and IQ in 7-year-old children. Environ Health Perspect. 2011;119(8):1189-1195. 10.1289/ehp.1003185 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Eskenazi  B, Huen  K, Marks  A, et al.  PON1 and neurodevelopment in children from the CHAMACOS study exposed to organophosphate pesticides in utero. Environ Health Perspect. 2010;118(12):1775-1781. 10.1289/ehp.1002234 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Yu  CJ, Du  JC, Chiou  HC, et al.  Increased risk of attention-deficit/hyperactivity disorder associated with exposure to organophosphate pesticide in Taiwanese children. Andrology. 2016;4(4):695-705. 10.1111/andr.12183 [DOI] [PubMed] [Google Scholar]
  • 16. Rauh  VA, Perera  FP, Horton  MK, et al.  Brain anomalies in children exposed prenatally to a common organophosphate pesticide. Proc Natl Acad Sci. 2012;109(20):7871-7876. 10.1073/pnas.1203396109 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Quenby  S, Gallos  ID, Dhillon-Smith  RK, et al.  Miscarriage matters: the epidemiological, physical, psychological, and economic costs of early pregnancy loss. The Lancet.  2021;397(10285):1658-1667. 10.1016/S0140-6736(21)00682-6 [DOI] [PubMed] [Google Scholar]
  • 18. Man  J, Hutchinson  J, Ashworth  M, et al.  Stillbirth and intrauterine fetal death: contemporary demographic features of >1000 cases from an urban population. Ultrasound Obstet Gynecol. 2016;48(5):591-595. 10.1002/uog.16021 [DOI] [PubMed] [Google Scholar]
  • 19. Stanton  C, Lawn  JE, Rahman  H, et al.  Stillbirth rates: delivering estimates in 190 countries. The Lancet. 2006;367(9521):1487-1494. 10.1016/S0140-6736(06)68586-3 [DOI] [PubMed] [Google Scholar]
  • 20. Ha  S, Liu  D, Zhu  Y, et al.  Ambient temperature and stillbirth: a multi-center retrospective cohort study. Environ Health Perspect. 2017;125(6):067011. 10.1289/EHP945 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Strand  LB, Barnett  AG, Tong  S. Maternal exposure to ambient temperature and the risks of preterm birth and stillbirth in Brisbane. Am J Epidemiol. 2012;175(2):99-107. 10.1093/aje/kwr404 [DOI] [PubMed] [Google Scholar]
  • 22. Basu  R, Sarovar  V, Malig  BJ. Association between high ambient temperature and risk of stillbirth in California. Am J Epidemiol. 2016;183(10):894-901. 10.1093/aje/kwv295 [DOI] [PubMed] [Google Scholar]
  • 23. Zhang  H, Zhang  X, Wang  Q, et al.  Ambient air pollution and stillbirth: an updated systematic review and meta-analysis of epidemiological studies. Environ Pollut. 2021;278:116752. 10.1016/j.envpol.2021.116752 [DOI] [PubMed] [Google Scholar]
  • 24. Bekkar  B, Pacheco  S, Basu  R, et al.  Association of air pollution and heat exposure with preterm birth, low birth weight, and stillbirth in the US: a systematic review. JAMA Netw Open .  2020;3(6):e208243. 10.1001/jamanetworkopen.2020.8243 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Bell  EM, Hertz-Picciotto  I, Beaumont  JJ. A case-control study of pesticides and fetal death due to congenital anomalies. Epidemiology. 2001;12(2):148-156. 10.1097/00001648-200103000-00005 [DOI] [PubMed] [Google Scholar]
  • 26. Bell  EM, Hertz-Picciotto  I, Beaumont  JJ. Case-cohort analysis of agricultural pesticide applications near maternal residence and selected causes of fetal death. Am J Epidemiol. 2001;154(8):702-710. 10.1093/aje/154.8.702 [DOI] [PubMed] [Google Scholar]
  • 27. de  Siqueira  MT, Braga  C, Cabral-Filho  JE, et al.  Correlation between pesticide use in agriculture and adverse birth outcomes in Brazil: an ecological study. Bull Environ Contam Toxicol. 2010;84(6):647-651. 10.1007/s00128-010-0027-8 [DOI] [PubMed] [Google Scholar]
  • 28. Razi  S, Rezaeian  M, Dehkordi  FG, et al.  Exposure to pistachio pesticides and stillbirth: a case-control study. Epidemiol Health .  2016;(.38):e2016016. 10.4178/epih.e2016016 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Bravo  R, Caltabiano  LM, Weerasekera  G, et al.  Measurement of dialkyl phosphate metabolites of organophosphorus pesticides in human urine using lyophilization with gas chromatography-tandem mass spectrometry and isotope dilution quantification. J Expo Anal Environ Epidemiol. 2004;14(3):249-259. 10.1038/sj.jea.7500322 [DOI] [PubMed] [Google Scholar]
  • 30. Zhang  X, Driver  JH, Li  Y, et al.  Dialkylphosphates (DAPs) in fruits and vegetables may confound biomonitoring in organophosphorus insecticide exposure and risk assessment. J Agric Food Chem. 2008;56(22):10638-10645. 10.1021/jf8018084 [DOI] [PubMed] [Google Scholar]
  • 31. Weisskopf  MG, Webster  TF. Trade-offs of personal vs. more proxy exposure measures in environmental epidemiology. Epidemiology. 2017;28(5):635-643. 10.1097/EDE.0000000000000686 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Liew  Z, Kioumourtzoglou  M-A, Roberts  AL, et al.  Use of negative control exposure analysis to evaluate confounding: an example of acetaminophen exposure and attention-deficit/hyperactivity disorder in Nurses’ Health Study II. Am J Epidemiol. 2019;188(4):768-775. 10.1093/aje/kwy288 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Flanders  WD, Strickland  MJ, Klein  M. A new method for partial correction of residual confounding in time-series and other observational studies. Am J Epidemiol. 2017;185(10):941-949. 10.1093/aje/kwx013 [DOI] [PubMed] [Google Scholar]
  • 34. Flanders  WD, Klein  M, Darrow  LA, et al.  A method to detect residual confounding in spatial and other observational studies. Epidemiology. 2011;22(6):823-826. 10.1097/EDE.0b013e3182305dac [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Raz  R, Roberts  AL, Lyall  K, et al.  Autism spectrum disorder and particulate matter air pollution before, during, and after pregnancy: a nested case–control analysis within the Nurses’ Health Study II cohort. Environ Health Perspect. 2014;123(3):264-270. 10.1289/ehp.1408133 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Håberg  SE, Stigum  H, Nystad  W, et al.  Effects of pre-and postnatal exposure to parental smoking on early childhood respiratory health. Am J Epidemiol. 2007;166(6):679-686. 10.1093/aje/kwm134 [DOI] [PubMed] [Google Scholar]
  • 37. Zammit  S, Thomas  K, Thompson  A, et al.  Maternal tobacco, cannabis and alcohol use during pregnancy and risk of adolescent psychotic symptoms in offspring. Br J Psychiatry. 2009;195(4):294-300. 10.1192/bjp.bp.108.062471 [DOI] [PubMed] [Google Scholar]
  • 38. Graham  S, Levin  ML, Lilienfeld  AM, et al.  Preconception, intrauterine, and postnatal irradiation as related to leukemia. Natl Cancer Inst Monogr. 1966;19:347-371. [PubMed] [Google Scholar]
  • 39. Shu  X, Jin  F, Linet  M, et al.  Diagnostic X-ray and ultrasound exposure and risk of childhood cancer. Br J Cancer. 1994;70(3):531-536. 10.1038/bjc.1994.340 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. Kneale  G, Stewart  A. Pre-conception X-rays and childhood cancers. Br J Cancer. 1980;41(2):222-226. 10.1038/bjc.1980.33 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Choi  G, Wang  Y-B, Sundaram  R, et al.  Polybrominated diphenyl ethers and incident pregnancy loss: the LIFE Study. Environ Res. 2019;168:375-381. 10.1016/j.envres.2018.09.018 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Messerlian  C, Williams  PL, Ford  JB, et al.  The Environment and Reproductive Health (EARTH) Study: a prospective preconception cohort. Human Reproduction Open .  2018;2018(2):hoy001. 10.1093/hropen/hoy001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Robledo  CA, Mendola  P, Yeung  E, et al.  Preconception and early pregnancy air pollution exposures and risk of gestational diabetes mellitus. Environ Res. 2015;137:316-322. 10.1016/j.envres.2014.12.020 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44. Rammah  A, Whitworth  KW, Symanski  E. Particle air pollution and gestational diabetes mellitus in Houston. Texas Environ Res. 2020;190:109988. 10.1016/j.envres.2020.109988 [DOI] [PubMed] [Google Scholar]
  • 45. Najafi  ML, Zarei  M, Gohari  A, et al.  Preconception air pollution exposure and glucose tolerance in healthy pregnant women in a middle-income country. Environ Health. 2020;19(1):1-10. 10.1186/s12940-020-00682-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46. Segal  TR, Giudice  LC. Before the beginning: environmental exposures and reproductive and obstetrical outcomes. Fertil Steril. 2019;112(4):613-621. 10.1016/j.fertnstert.2019.08.001 [DOI] [PubMed] [Google Scholar]
  • 47. Arizona Department of Environmental Quality . Pesticide Groundwater Quality Protection 2022 Annual Report. Arizona Department of Environmental Quality; 2022.
  • 48. Rull  RP, Ritz  B, Shaw  GM. Neural tube defects and maternal residential proximity to agricultural pesticide applications. Am J Epidemiol. 2006;163(8):743-753. 10.1093/aje/kwj101 [DOI] [PubMed] [Google Scholar]
  • 49. Rull  RP, Ritz  B. Historical pesticide exposure in California using pesticide use reports and land-use surveys: an assessment of misclassification error and bias. Environ Health Perspect. 2003;111(13):1582-1589. 10.1289/ehp.6118 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50. Cockburn  M, Mills  P, Zhang  X, et al.  Prostate cancer and ambient pesticide exposure in agriculturally intensive areas in California. Am J Epidemiol. 2011;173(11):1280-1288. 10.1093/aje/kwr003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51. Shirangi  A, Nieuwenhuijsen  M, Vienneau  D, et al.  Living near agricultural pesticide applications and the risk of adverse reproductive outcomes: a review of the literature. Paediatr Perinat Epidemiol. 2011;25(2):172-191. 10.1111/j.1365-3016.2010.01165.x [DOI] [PubMed] [Google Scholar]
  • 52. Goulet  L, Thériault  G. Stillbirth and chemical exposure of pregnant workers. Scand J Work Environ Health. 1991;17(1):25-31. 10.5271/sjweh.1738 [DOI] [PubMed] [Google Scholar]
  • 53. Hanke  W, Jurewicz  J. The risk of adverse reproductive and developmental disorders due to occupational pesticide exposure: an overview of current epidemiological evidence. Int J Occup Med Environ Health. 2004;17(2):223-243. [PubMed] [Google Scholar]
  • 54. Pastore  LM, Hertz-Picciotto  I, Beaumont  JJ. Risk of stillbirth from occupational and residential exposures. Occup Environ Med. 1997;54(7):511-518. 10.1136/oem.54.7.511 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55. Sallmén  M, Liesivuori  J, Taskinen  H, et al.  Time to pregnancy among the wives of Finnish greenhouse workers. Scand J Work Environ Health. 2003;29(2):85-93. 10.5271/sjweh.709 [DOI] [PubMed] [Google Scholar]
  • 56. Curtis  KM, Savitz  DA, Weinberg  CR, et al.  The effect of pesticide exposure on time to pregnancy. Epidemiology. 1999;10(2):112-117. 10.1097/00001648-199903000-00005 [DOI] [PubMed] [Google Scholar]
  • 57. Thomas  DC, Petitti  DB, Goldhaber  M, et al.  Reproductive outcomes in relation to malathion spraying in the San Francisco Bay Area, 1981–1982. Epidemiology. 1992;3(1):32-39. 10.1097/00001648-199201000-00007 [DOI] [PubMed] [Google Scholar]
  • 58. Ni  W, Gao  H, Wu  B, et al.  Gestational exposure to cyfluthrin through endoplasmic reticulum (ER) stress–mediated PERK signaling pathway impairs placental development. Toxics. 2022;10(12):733. 10.3390/toxics10120733 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59. Burton  GJ, Jauniaux  E. Placental oxidative stress: from miscarriage to preeclampsia. J Soc Gynecol Investig. 2004;11(6):342-352. 10.1016/j.jsgi.2004.03.003 [DOI] [PubMed] [Google Scholar]
  • 60. Ambali  S, Abbas  S, Shittu  M, et al.  Effects of gestational exposure to chlorpyrifos on implantation and neonatal mice. J Cell Animal Biol. 2009;3(4):050-057. [Google Scholar]
  • 61. Iheanacho  J, Onyeka  P, Udujih  H, et al.  Adverse effects of sub-lethal doses of chlorpyrifos on birth outcome of female Wistar rats. J Applied Life Sci Int. 2020;23(8):1-6. 10.9734/jalsi/2020/v23i830177 [DOI] [Google Scholar]
  • 62. Maurissen  JP, Hoberman  AM, Garman  RH, et al.  Lack of selective developmental neurotoxicity in rat pups from dams treated by gavage with chlorpyrifos. Toxicol Sci. 2000;57(2):250-263. 10.1093/toxsci/57.2.250 [DOI] [PubMed] [Google Scholar]
  • 63. Marettova  E, Maretta  M, Legáth  J. Effect of pyrethroids on female genital system. Rev Animal Reproduct Sci. 2017;184:132-138. 10.1016/j.anireprosci.2017.07.007 [DOI] [PubMed] [Google Scholar]
  • 64. Bretveld  RW, Thomas  CM, Scheepers  PT, et al.  Pesticide exposure: the hormonal function of the female reproductive system disrupted?  Reprod Biol Endocrinol. 2006;4(1):1-14. 10.1186/1477-7827-4-30 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 65. Sullivan  DA, Wira  CR. Hormonal regulation of immunoglobulins in the rat uterus: uterine response to multiple estradiol treatments. Endocrinology. 1984;114(2):650-658. 10.1210/endo-114-2-650 [DOI] [PubMed] [Google Scholar]
  • 66. Paulson  RJ. Hormonal induction of endometrial receptivity. Fertil Steril. 2011;96(3):530-535. 10.1016/j.fertnstert.2011.07.1097 [DOI] [PubMed] [Google Scholar]
  • 67. Tomalik-Scharte  D, Lazar  A, Meins  J, et al.  Dermal absorption of permethrin following topical administration. Eur J Clin Pharmacol. 2005;61(5-6):399-404. 10.1007/s00228-005-0932-7 [DOI] [PubMed] [Google Scholar]
  • 68. Eadsforth  C, Baldwin  M. Human dose-excretion studies with the pyrethroid insecticide, cypermethrin. Xenobiotica. 1983;13(2):67-72. 10.3109/00498258309052238 [DOI] [PubMed] [Google Scholar]
  • 69. Woollen  B, Marsh  J, Laird  W, et al.  The metabolism of cypermethrin in man: differences in urinary metabolite profiles following oral and dermal administration. Xenobiotica. 1992;22(8):983-991. 10.3109/00498259209049904 [DOI] [PubMed] [Google Scholar]
  • 70. Ratelle  M, Côté  J, Bouchard  M. Toxicokinetics of permethrin biomarkers of exposure in orally exposed volunteers. Toxicol Lett. 2015;232(2):369-375. 10.1016/j.toxlet.2014.12.003 [DOI] [PubMed] [Google Scholar]
  • 71. Promthep  K, Nopparat  C, Mukda  S, et al.  Proteomic profiling reveals neuronal ion channel dysregulation and cellular responses to DNA damage-induced cell cycle arrest and senescence in human neuroblastoma SH-SY5Y cells exposed to cypermethrin. Neurotoxicology. 2022;93:71-83. 10.1016/j.neuro.2022.08.015 [DOI] [PubMed] [Google Scholar]
  • 72. Le  ON, Rodier  F, Fontaine  F, et al.  Ionizing radiation-induced long-term expression of senescence markers in mice is independent of p53 and immune status. Aging Cell. 2010;9(3):398-409. 10.1111/j.1474-9726.2010.00567.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 73. Chang  J, Wang  Y, Shao  L, et al.  Clearance of senescent cells by ABT263 rejuvenates aged hematopoietic stem cells in mice. Nat Med. 2016;22(1):78-83. 10.1038/nm.4010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 74. Jalili  C, Roshankhah  S, Salahshoor  MR, et al.  Resveratrol attenuates malathion induced damage in some reproductive parameters by decreasing oxidative stress and lipid peroxidation in male rats. J Fam Reproduc Health.  2019;13(2):70-79. 10.18502/jfrh.v13i2.1912 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 75. Farhat  A, Shafiee  M, Mohammadzadeh  A, et al.  Association of neonatal asphyxia with serum levels of heat shock protein 27 in a small sample of newborns. Acta Med Iran.  2019;57(5):303-307. 10.18502/acta.v57i5.1866 [DOI] [Google Scholar]
  • 76. Farrar  JJ, Ellsworth  PC, Sisco  R, et al.  Assessing compatibility of a pesticide in an IPM program. J Int Pest Manage .  2018;9(1):3. 10.1093/jipm/pmx032 [DOI] [Google Scholar]
  • 77. US Department of Agriculture . A National Road Map for Integrated Pest Management. Office of USDA’s Chief Economist: Office of Pest Management Policy; 2013. [Google Scholar]
  • 78. Vandervoet  TF, Ellsworth  PC, Carrière  Y, et al.  Quantifying conservation biological control for management of Bemisia tabaci (Hemiptera: Aleyrodidae) in cotton. J Econ Entomol. 2018;111(3):1056-1068. 10.1093/jee/toy049 [DOI] [PubMed] [Google Scholar]
  • 79. Anderson  JA, Ellsworth  PC, Faria  JC, et al.  Genetically engineered crops: importance of diversified integrated pest management for agricultural sustainability. Front Bioeng Biotechnol. 2019;7:24. 10.3389/fbioe.2019.00024 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 80. Naranjo  SE, Ellsworth  PC. Fifty years of the integrated control concept: moving the model and implementation forward in Arizona. Pest Manage Sci: Form Pest Sci. 2009;65(12):1267-1286. 10.1002/ps.1861 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 81. Lewis  RG, Roberts  J, Chuang  J, et al.  Measuring and reducing exposure to the pollutants in house dust. Am J Public Health. 1995;85(8_Pt_1):1168. 10.2105/AJPH.85.8_Pt_1.1168 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 82. Sugeng  AJ. From Field to Home: Assessing Air Infiltration and Soil Track-in Transport Pathways of Agricultural Pesticides into Farmworkers' Home and Identifying Risk Factors for Increased In-Home Pesticide Levels. The University of Arizona; 2016. [Google Scholar]
  • 83. Wong  GK, Jim  CY. Abundance of urban male mosquitoes by green infrastructure types: implications for landscape design and vector management. Landsc Ecol. 2018;33(3):475-489. 10.1007/s10980-018-0616-1 [DOI] [Google Scholar]
  • 84. Proctor  SP, Maule  AL, Heaton  KJ, et al.  Permethrin exposure from fabric-treated military uniforms under different wear-time scenarios. J Expo Sci Environ Epidemiol. 2014;24(6):572-578. 10.1038/jes.2013.65 [DOI] [PubMed] [Google Scholar]
  • 85. Rossbach  B, Appel  KE, Mross  KG, et al.  Uptake of permethrin from impregnated clothing. Toxicol Lett. 2010;192(1):50-55. 10.1016/j.toxlet.2009.06.863 [DOI] [PubMed] [Google Scholar]
  • 86. Chen  L, Bell  EM, Caton  AR, et al.  Residential mobility during pregnancy and the potential for ambient air pollution exposure misclassification. Environ Res. 2010;110(2):162-168. 10.1016/j.envres.2009.11.001 [DOI] [PubMed] [Google Scholar]
  • 87. Bell  ML, Belanger  K. Review of research on residential mobility during pregnancy: consequences for assessment of prenatal environmental exposures. J Expo Sci Environ Epidemiol. 2012;22(5):429-438. 10.1038/jes.2012.42 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 88. Warren  JL, Son  J-Y, Pereira  G, et al.  Investigating the impact of maternal residential mobility on identifying critical windows of susceptibility to ambient air pollution during pregnancy. Am J Epidemiol. 2017;187(5):992-1000. 10.1093/aje/kwx335 [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

Web_Material_kwae198
web_material_kwae198.docx (30.4KB, docx)

Data Availability Statement

The data used for this study are available upon request to the corresponding author.

Articles from American Journal of Epidemiology are provided here courtesy of Oxford University Press

RESOURCES