Environmental risk factors, environmental protective factors and biomarkers of ADHD have been extensively studied in meta-analyses and systematic reviews. However, these studies are typically restricted to a single topic and results can be affected by bias (Ioannidis, 2005). As the boundaries are indistinct between risk factors, protective factors and biomarkers, the aim of this review was to identify and evaluate the evidence of environmental factors and peripheral biomarkers associated with a diagnosis of ADHD.
Screening, data extraction and methodological appraisal of the collated studies were performed by two independent investigators using PRISMA reporting guidelines. Studies were collected from systematic searches of PubMed®, MEDLINE®, EMBASE® and the Cochrane Database of Systematic Reviews from inception of the database until 31 October 2019.* Included meta-analyses were classified by the strength of the evidence of potential environmental risk factors, environmental protective factors and peripheral biomarkers for ADHD. The five classes were convincing (class I), highly suggestive (class II), suggestive (class III), weak (class IV) and not significant (NS).
There were 35 eligible articles, out of the original 1829 identified articles, which provided 63 unique meta-analyses (40 potential environmental factors and 23 peripheral biomarkers). There were five environmental risk factors graded as class I (convincing): pre-pregnancy obesity (defined as body mass index [BMI] ≥30 kg/m2, odds ratio [OR] 1.63, 95% confidence interval [CI] 1.49–1.77), childhood eczema (OR 1.31, 95% CI 1.2–1.44), hypertensive disorders during pregnancy (chronic hypertension, gestational hypertension and pre-eclampsia; OR 1.29, 95% CI 1.21–1.35), pre-eclampsia (de novo or superimposed chronic hypertension; OR 1.28, 95% CI 1.21–1.35) and maternal acetaminophen exposure during pregnancy (risk ratio 1.25, 95% CI 1.17–1.34). Class II (highly suggestive) environmental risk factors included maternal smoking during pregnancy (OR 1.6, CI 1.45–1.76), childhood asthma (OR 1.51, 95% CI 1.4–1.63) and pre-pregnancy overweight (defined as BMI 25–29.9 kg/m2, OR 1.28, 95% CI 1.21–1.35).
Class III (suggestive) evidence included some perinatal hypoxic conditions (5-minute Apgar score <7, OR 1.30, 95% CI 1.11–1.52; breech or transverse presentation, OR 1.14, 95% CI 1.06–1.22) and preterm birth (OR 1.84, 95% CI 1.36–2.49). The highest class for parenting environmental risk factors was Class IV (weak; low education level of father, OR 2.1, 95% CI 1.27–3.47; low education of mother, OR 1.91, 95% CI 1.2–3.03; single parent family, OR 1.28, 95% CI 1.08–1.52). Breast-feeding was determined to be the only statistically significant (class IV [weak]) environmental protective factor against ADHD (OR 0.70, 95% CI 0.53–0.93).
Furthermore, lower concentration of serum vitamin D in individuals with ADHD was the only biomarker graded as a high level of evidence (class II [highly suggestive]; weighted mean difference –6.93, 95% CI –9.34 to –4.51). Class III [suggestive] biomarkers included higher blood lead (OR 0.78, 95% CI 0.35–1.21) and lower blood magnesium (OR –0.55, 95% CI –0.82 to –0.28) in patients with ADHD.
There were several limitations of this systematic review. First, the identified associations do not imply causality due to the nature of observational studies. Second, the authors did not consider changes in classification for ADHD, the varieties of ADHD and they could not distinguish between specific symptoms for diagnosing ADHD. Third, the data did not allow for assessment of environmental factors or biomarkers of ADHD according to sex, intellectual disability and comorbid psychiatric disorders. Fourth, neurocognitive markers were not assessed, which the authors suggest may act as biomarkers for ADHD (Pievsky & McGrath, 2018). Fifth, the identified factors may not be independent of each other. Finally, the authors were only able to address the associations in the published meta-analyses and so some may have been missed.
Overall, this review revealed that within the 63 identified environmental factors and biomarkers, only pre-pregnancy obesity and overweight, maternal smoking and maternal acetaminophen exposure during pregnancy retained high level of evidence as risk factors for ADHD. However, the associations are possibly not causative and so the authors suggest that high-quality primary studies are required to confirm the results of this systematic review.
*Inclusion criteria included systematic reviews with meta-analyses of observational studies that examined associations of potential environmental risk factors, environmental protective factors or peripheral biomarkers with diagnosis of ADHD. Also, meta-analyses were included that used categorical ADHD diagnosis criteria according to Diagnostic and Statistical Manual of Mental Disorders, hyperkinetic disorder according to the International Classification of Diseases, or less rigorous criteria (e.g. self-reports). Exclusion criteria were articles with no sufficient data, non-human studies, primary studies, genetic studies and conference abstracts
Ioannidis JP. Why most published research findings are false. PLoS Med 2005; 2: e124.
Kim JH, Kim JY, Lee J, et al. Environmental risk factors, protective factors, and peripheral biomarkers for ADHD: an umbrella review. Lancet Psychiatry, 2020; 7: 955-970.
Pievsky MA, McGrath RE. The neurocognitive profile of attention-deficit/hyperactivity disorder: a review of meta-analyses. Arch Clin Neuropsychol 2018; 33: 143-157.