Systematic Review of Environmental and Psychosocial Risk Factors associated with Attention Deficit Hyperactivity Disorder, Oppositional Defiant Disorder, and Conduct Disorder in Children and Adolescents

The study population included children and adolescents under 18 years old with a diagnosis of Attention Deficit Hyperactivity Disorder with either comorbid Oppositional Defiant Disorder or Conduct Disorder or both. Studies set within both community and clinical populations are included. Population surveys, cross-sectional studies, cohort studies, and case-control studies were among the research designs covered. The review did not include any case studies, editorials, reviews, or opinions. We also excluded dissertations, conference abstracts and non-English articles. Included risk factors were Environmental and psychosocial risk factors associated with ADHD and ODD or CD in children and adolescents. Excluded are studies including genetic risk factors. Studies were included if published during or from 2015 until 2021.

Five electronic databases were searched for relevant articles in February 2021. The databases searched were Cochrane, Embase(ovid), Medline EBCSO, PubMed and PsychINFO.

The search strategy included combinations of 4 key blocks of terms (ADHD/HD, Risk factors, ODD and CD, observational studies) using medical subject headings (MESH terms) and text words.

Following the initial search and removal of duplications, the title and abstracts of the studies were screened by two independent reviewers using the eligibility criteria. Then the full texts of the remaining published articles were screened by the two reviewers. Subsequently the two reviewers met to resolve the conflicts by consensus.

Two reviewers independently extracted data from the included published articles to ensure concordance.

The following items were extracted from each publication: author name and year of publication, country of the study, the aim of the study, study design, characteristics of the population (age range, gender and the number of participants), setting, diagnostic method used for the risk factor and the outcome (ADHD, CD, ODD), results of the studies, relevant findings in discussion and conclusion, limitation and quality check of each study.

Environmental risk factors: in our study, we defined environmental risk factor as any non-genetic biological risk factor.

Psychosocial risk factors: we defined psychosocial risk factor as any factor related to socioeconomic disadvantage (e.g., low parental income, manual occupation, poor education, and not owning the family home) (15).

In our systematic review, the outcome variable was ADHD, ODD and CD. We defined DBD as the presence of any specific categories (such as ADHD, ODD and CD) in children.

The critical appraisal of the studies was performed using the Critical Appraisal Skills Programme (CASP) risk of bias tool. (Reference: Critical Appraisal Skills Programme (2022).We used CASP Case-Control and Cohort Study Checklist. [Online], available at: https://casp-uk.net/ Accessed: 12.03.2022. It was chosen because it is appropriate for use for the included studies. The CASP is an appraisal tool to systematically evaluate the reliability, applicability, and outcomes of published articles. The CASP tool consists of 11 questions for the case control studies and 12 questions for the cohort studies, and we are asked to record a “yes”, “no” or “can’t tell” to 9 out of the 11/12 questions. The authors agreed to score the answer “Yes” as (score= 1), and “No” as (score= 0), and “Can’t tell” as (score=0). The remaining three questions look at the precision of the result and how strong the association is between the exposure and the outcome (treatment effect), and the implication of the study for practice. The authors agreed to rate the precision of the result as High (score=1) or Low (score=0). Similarly, the treatment effect and the implication of the study for practice were rated as either High (score=1) or Low (score=0). A total score was calculated for each study, and studies were then rated as poor (total score less than 4 points), fair (total score between 5 and 7 points), or good (total score between 8 and 11/12 points), based on the scores obtained. Two independent assessors involved in the study completed the questionnaires. In case of disagreement, a third reviewer was contacted.

The characteristics of the studies included are summarised in Table 1.

Studies were carried out in a total of nine various countries across three continents. Seven of the studies used samples from the community, most of which were population-based cohort or cross-sectional research. Four of the studies utilised samples that were recruited, at least partly through a clinical setting. This review took in two case–control studies for its consideration. All research was published between 2015 and 2021. Sample sizes ranged from 112 to 74,565, with seven of the eleven studies including over 1000 participants. The ages of participants ranged from 3 to 17 years.

Different clinical measures were used to make the diagnosis of ADHD, ODD, and CD; data on the diagnosis was provided by parents, teachers, in some studies the children themselves, and also by researcher/clinician. The majority of studies used information from parents to reach a diagnosis of ADHD, OD, or CD. Two of the included studies used more than two informants to reach the diagnosis and two studies relied on parent report of clinical diagnosis.

Of the included studies, risk factors measured included perinatal maternal distress, prenatal maternal smoking, prenatal exposure to very severe maternal obesity, maternal PCOS and hirsutism, preterm birth and low birth weight, macrosomia, allergic diseases, socioeconomic status, energy drink and coffee consumption.

The primary aim of the included publications was to examine the association between the above-mentioned risk factors with ADHD, ODD and CD.

The quality rating was “fair” for four studies (16–19), and “good” for the remaining seven studies (20–26). The quality of the studies is summarised in table 2.

The results of the individual studies are summarised in table 2.

Eleven studies were included in this systematic review. Among these, ten studies involved environmental risk factors, and only one study involved socioeconomic risk factors as exposure measures. Of the ten studies highlighting the environmental risk factors, six studies reported perinatal risk factors and three reported atopic diseases as exposure measures. And one study involved exposure to energy and coffee drinks. We have found that the most common risk factors associated with ADHD, ODD and CD in Europe and North America were perinatal risk factors. In contrast, the risk factors of atopic diseases were more common in Asia.

Among the four selected studies in North America, three were conducted in the USA and one in Canada. Three out of the four studies that were performed in North America examined the impact of the perinatal risk factor on ADHD, ODD and CD, while the remaining one selected article from this continent examined the effect of energy and coffee drinks as an environmental risk factor.

Moreover, three selected studies conducted in Europe investigated the impact of perinatal risk factors.

In Asia, four studies from this continent were included in this systematic review; three of them investigated the association between allergic diseases and ADHD, ODD and CD and one looked at the relationship between sociodemographic factors and ADHD and other DBDs.

Among selected studies, only one study reported the association between sociodemographic risk factors and ADHD, ODD and CD.

In Turkey, Senol V et al. (23) examined the relationships of ADHD and other DBDs with sociodemographic factors among school children in Kayseri, his results showed that compared to their peers, the prevalence of Attention Deficit Hyperactivity Disorder (ADHD) was significantly higher among boys and children whose mothers were homemakers, as well as those from low-income, and poorly educated families. CD was more prevalent among boys and children ages 13-15 whose parents had low incomes and separated. ODD was more prevalent in boys and those whose mothers were homemakers.

Although previous studies have investigated the risk of interaction between genetic and environmental risk factors (27–28). To the best of our knowledge, this review is the first to systematically evaluate evidence of an association between the most up to date wide range of independent environmental risk factors and psychosocial risk factors with ADHD, ODD and CD.

Studies from across three continents contributed to this review, and the conclusions drawn are relevant in many different countries.

One significant finding of the review is that six out of eleven studies explored perinatal risk factors as independent environmental risk factors associated with ADHD, ODD and CD and they were found to be more prevalent in North America and Europe. Whereas three of the included studies investigated the relationship between allergic diseases and ADHD, ODD and CD and these risk factors were more prevalent in Asia.

We found only one study which was conducted in Asia, looking at the relationship between sociodemographic factors and ADHD and other DBDs.

Two existing systematic reviews have examined the association between environmental perinatal risk factors and ADHD, ODD and CD (27, 29).

Both reviews highlighted that factors such as drinking, smoking, or stress during pregnancy are associated with ADHD, ODD and CD.

Our findings are in agreement with the aforementioned systematic reviews. However, Kalil et al. systematic review (27) examined the interaction between genetic and environmental risk factors, while ours didn’t focus on this interaction as this was beyond the scope of our review. Moreover, K. Latimer’s study (30) included an assessment of factors present antenatally and in the first four years of life (up to 4th birthday; 0–47 month), in comparison to our review, we included the most up to date perinatal risk factors in the literature, and we examined the association of these factors with DBDs presenting up to the age of 17.

Additionally, compared to previous reviews, our study found a new result of the association between maternal PCOS, hirsutism and DBDs. However, the link is positive but imprecise. Therefore, further research is needed to determine the etiological approach towards ADHD, ODD and maternal PCOS and Hirsutism (17).

This systematic review also reports interesting findings, which is the association between allergic diseases, such as atopic dermatitis, asthma, allergic rhinitis and ADHD, ODD and CD. Among reviewed papers, three of the included studies highlighted this association (20,24,26). Existing literature seems to be consistent with our finding, Miyazaki C et al. (31) conducted a meta-analysis showing that children with ADHD are more likely to have asthma, allergic rhinitis, atopic dermatitis, and allergic conjunctivitis than their counterparts. Nonetheless, our review not only highlighted evidence of the association between allergic disease and ADHD (20). It also showed evidence of association with other DBDs like ODD/CD. Indeed, further studies should compare the association between these illnesses and ADHD with comorbid ODD/CD.

Another interesting finding of our study suggests that allergic risk factors are more prevalent in Asia; this could possibly be due to Asian populations being more genetically or environmentally prone to allergic diseases. This would be a fruitful area for further research in order to establish whether there is an association between the Asian population and allergic diseases, as this can play a major role in the early diagnosis and intervention of DBDs (31).

Our systematic review also highlighted energy drink consumption as an environmental risk factor for an increase in ADHD and CD symptoms over time. It would be prudent to investigate how energy drink consumption predicts an increase in ADHD and CD symptoms over time, as this could lead to the development of interventions for high-risk youth (18). However, this evidence comes from a single study and replication is required.

This study has some limitations. First, the diagnosis of ADHD and behavioural disorders were parent reported in a number of studies, so this may be influenced by recall bias. Second, due to the fact that the evidence consists of an observational cohort, cross-sectional, and case-control studies, it is impossible to draw conclusions regarding temporality and whether the observed association constitute a causal relationship; in addition, another reason we can not clarify the exact nature of the relationship is the existence of various confounding familial and psychosocial variables. Therefore, future studies should aim to reduce this weakness by evaluating a wide range of family, social and psychological variables that may confound findings. Furthermore, certain methodological limitations inherent to this review must be acknowledged. As in most systematic reviews, there is a risk of reporting bias since only published studies in identifiable sources and articles in the English language were included. Also, some papers that have strongly influenced the field have been excluded because they focus on genetic risk factors. We also discovered that there was just one study that examined psychosocial risk variables, this may be a result of the difficulty in maintaining families facing adversity in longitudinal studies and the difficulties involved in conducting research into complicated psychosocial risk factors.