Shoaff JR, Coull B, Weuve J, et al. Association of exposure to endocrine-disrupting chemicals during adolescence with attention-deficit/hyperactivity disorder-related behaviors. JAMA Netw Open. 2020;3(8):e2015041.
To assess the link between adolescent exposure to specific endocrine-disrupting chemicals—including phthalates, parabens, phenols, and triclocarban—and behaviors related to attention-deficit/hyperactivity disorder (ADHD)
Cross-sectional study of adolescents from the prospective New Bedford Cohort, an ongoing birth cohort of mother-infant pairs with infants born between 1993 and 1998 including adult mothers (aged ≥18 years) who resided in the 4 towns neighboring the New Bedford Harbor Superfund Site during pregnancy. Researchers recruited subjects after delivery.
For this study, N=205 adolescents of the cohort, all of whom provided at least 1 spot urine sample during the time period of neurodevelopmental testing.
Among the 788 newborns enrolled from 1993 to 1998, 660 met inclusion criteria (residence in study area, biomarkers of chemical exposure, exclusion of catastrophic central nervous system injury or illness, and accurate contact information).
Study Parameters Assessed
Researchers performed neurodevelopmental testing when children were aged approximately 15 years (between 2008 to 2014).
Halfway through the 15-year follow-up, researchers added assessment of exposure to endocrine-disrupting chemicals (EDCs), such that they collected spot urine samples at 2 points: during the neurodevelopmental testing and approximately 1 week later (mean 7 days, range 1 to 35 days). Among the 252 adolescents invited to provide urine samples, 205 performed at least 1 spot urine test and were included in the analysis. From the 252 adolescents, 144 (57%) provided 2 samples. Researchers collected data from June 18, 2011, and June 10, 2014, and performed statistical analyses during the period from January 15, 2019, to December 31, 2019.
Baseline measurements included exposure assessment via urine concentrations of multiple EDCs according to pathway, including the sum of (1) anti-androgenic phthalates, (2) di(2-ethylhexyl)phthalate (DEHP) metabolites, (3) personal care product phthalates, (4) parabens, (5) bisphenols, and (6) dichlorophenols in addition to neurodevelopmental testing.
Researchers measured ADHD behaviors with validated behavioral checklists including the Behavior Assessment System for Children, Second Edition (BASC-2; parent-, teacher-, and self-reported) and the Conners Attention Deficit Scale (CADS; parent- and teacher-reported). Overall, researchers collected 14 indices of ADHD behavior.
ADHD alone has a large healthcare cost in the range of $40 billion to $70 billion annually, so finding more effective ways to mitigate EDC exposure, and elimination once exposed, is prudent in light of the precautionary principle.
Among the 205 adolescents, all had at least 1 outcome measure, 204 had parent- and self-completed checklists, and 173 had teacher-completed checklists.
Statistical analysis included repeated-measures analysis design utilizing multivariate modified Poisson models to assess adjusted relative risk of ADHD behavior with EDC exposure.
Investigators assessed covariates including the role of parental demographic factors such as race/ethnicity, education level, and income. When a child reached 15 years of age, the researchers updated demographic characteristics along with smoking habits and child medical history, including behavioral disorders and medication use. Researchers administered the Centers for Disease Control and Prevention (CDC) Youth Risk Behavior survey to the 15-year-olds to assess adolescent tobacco, cannabis, and alcohol use.
To measure prenatal levels of organochlorine exposure, researchers collected umbilical blood at birth for serum levels of polychlorinated biphenyls and dichlorodiphenyldichloroethylene. They assessed medical records for childhood lead levels. Adolescents completed diaries recording food consumption and use of personal care products for 24 hours prior to urine samples.
The CDC performed biomarker analysis in 2 batches. Batch 1 included 11 phthalate metabolites, and Batch 2 added 5 biomarkers of phthalates or phthalate substitutes.
Researchers also considered individual chemicals outside of the summary exposures, including: benzophenone-3, triclosan, triclocarban, mono-3-carboxypropyl phthalate (MCPP), monocarboxynonyl phthalate (MCNP), and cyclohexane-1,2-dicarboxylic acid monohydroxy isononyl ester (MHINCH), in addition to the individual chemicals covered in the summary measures.
Primary Outcome Measures
The primary outcomes were the association between ADHD behaviors, as assessed with validated measures, and urine spot measurements of EDCs, measured in aggregate by pathway and individually.
For the N=205 participants, the mean age was 15.3 ± 0.7 years, with 112 (55%) females and 125 nonHispanic White (61%) participants (racial information about the remaining participants was not provided).
There was a modest association between adolescent exposure to select EDCs, particularly phthalates, and greater risk of ADHD behavior problems, at levels typical of adolescents in the general US population.
Specifically, median urine concentrations were:
- 0.45 μmol/L of molar sum (Σ) antiandrogenic phthalates
- 0.13 μmol/L of ΣDEHP metabolites
- 0.49 μmol/L of Σpersonal care product phthalates
- 0.35 μmol/L of Σparabens
- 0.02 μmol/L of Σbisphenols
- 0.02 μmol/L of Σdichlorophenols
Further, n=82 participants (40%) demonstrated significant behavior problems, and n=39 (19%) were diagnosed with ADHD.
Each 2-fold increase in the aggregate antiandrogenic phthalate concentrations was linked with a 1.34 (95% CI, 1.00-1.79) increase in the risk of ADHD-related behavior problems.
A 2-fold increase in the aggregate of dichlorophenols was associated with a 1.15 (95% CI, 1.01-1.32) greater risk.
Associations showed a stronger trend in male adolescents, particularly with regard to ΣDEHP metabolites (P=0.002), which demonstrated a 2-fold increase associated with a 1.62 (95% CI, 1.38-1.91) greater risk of ADHD-related behavior versus 1.06 (95% CI, −1.15-1.33) increased risk in female adolescents. However, comparisons of sex-specific differences were not possible given the sample size (Figure eTable 2 in the Supplement). Other sex differences were not statistically significant.
ADHD is a profound neurobehavioral disorder that affects between 2% and 7% of children, with another 5% experiencing difficulty with inattention, overactivity, and impulsivity and who are just below the threshold to meet full diagnostic criteria for ADHD.1 Symptoms often result in serious academic and social issues that may extend into adult life.2 ADHD has an uncertain yet multifactorial etiology that spans genomics, sociocultural influences, adverse childhood experiences, and a lifetime of external environmental exposures (sometimes referred to as the exposome3), with diet and environment probably playing key roles.4-6 On a population level, elimination diets and omega-3 supplementation demonstrate the most promising lifestyle approach.7
The goal of the present study was to determine if adolescent exposure to EDCs is detrimental to behavior considering the rapid brain development and neuroplasticity during this time of life. The answer from this study is that there is a modest association between EDCs and ADHD in adolescents, particularly with exposure to specific phthalates.
On the other hand, association is not causation. Did the exposure to the New Bedford Harbor Superfund Site and several ubiquitous EDCs cause ADHD? We don’t know. Was earlier exposure, such as during gestation or early childhood, associated with greater risk of ADHD later in adolescence? We don’t know, though previous studies suggest there may be a link.8,9 (On the other hand, the Environmental Protection Agency has published that there is no clear link between prenatal phthalate exposure and ADHD,10 but lack of proof does not disprove the hypothesis.) What is the most vulnerable time of exposure when mapped to neurodevelopmental outcomes like ADHD? So far, age 3 has emerged as important vis-à-vis phthalate exposure and cognitive development.11 Is there a sex or gender difference? We don’t know. Though male subjects showed a stronger association with ΣDEHP metabolites, there was insufficient power to assess sex differences more broadly.
Is this finding clinically relevant? When the European Union formed a steering committee to estimate the downstream risk of exposure to EDCs, it put the total bill at approximately $209 billion, assuming that EDCs cause IQ loss and associated intellectual disability, autism, ADHD, childhood obesity, adult obesity, adult diabetes, cryptorchidism, male infertility, and mortality associated with low testosterone.12 This new study offers limited evidence of the risk of geography-based risk factors and exposure to consumer products containing EDCs. ADHD alone has a large healthcare cost in the range of $40 billion to $70 billion annually, so finding more effective ways to mitigate EDC exposure, and elimination once exposed, is prudent in light of the precautionary principle.
Limitations of this study include the short half-lives of EDCs, which may lead to exposure misclassification and/or underestimation. Most participants (57%) provided 2 samples, which limits the effect of this potential bias. The authors point out another interesting area of reverse causation: that neurobehavioral problems like ADHD could bias exposure to EDCs, for example, by making an adolescent less likely to eat a healthy diet, more likely to misuse substances, and/or less likely to check labels before using a shampoo or fingernail polish. However, the adjustment did not alter results, suggesting that reverse causation is unlikely. While prenatal organochloride exposure was known from umbilical cord blood, EDC exposure is not known. Finally, given that the sample is derived from the small geographic region of New Bedford, it is unclear how generalizable the findings are for other populations.
The EDCs included in the study are common industrial chemicals present in many household goods. Phthalates are used in soft, flexible plastics and in polyvinyl chloride (PVC) products. They’re found in nail polish, shampoos, shower curtains, baby toys, vinyl flooring, car interiors, and medical devices such as IV bags. Studies on phthalates show detrimental effects on all ages including disruption of insulin, thyroid, and reproductive hormones, such as free testosterone and sex-hormone-binding globulin. In this study, 1 type of phthalate, called DEHP, doubled the risk of ADHD in male adolescents. DEHP has several adverse reproductive-based outcomes in males backed by robust evidence, including anogenital distance, semen parameters, and testosterone.13
In females, DEHP may block estradiol production in the ovaries, which reduces estradiol levels and may cause anovulation.14 One study showed that some phthalates may be associated with preterm birth.15
Sex-specific findings guide us toward future studies that may help us fund larger studies that demonstrate more significant effect sizes.
While the harmful effects of EDCs make most of us want to assume EDCs are guilty until proven innocent, there is a study from 2016 that may help and empower our patients. A 3-day, community-based youth-empowerment intervention aimed at educating adolescent females about ingredients to avoid in personal care products led to marked reduction in urinary EDCs, including some phthalate and paraben metabolites, as well as triclosan.16 Helping adolescents and other consumers to choose personal care products labeled to be free of phthalates, parabens, triclosan, and benzophenone-3 can reduce exposure to EDCs. Importantly, the study found that involving youth in the design and implementation was essential to successful recruitment, retention, compliance, and acceptance of the intervention.