Adverse childhood experience and asthma onset: a systematic review

Study characteristics

The online supplementary material associated with this article details the key contents of the studies included for review. There was a large disparity in sample size within the studies retained for review, ranging from 145 to 13 907. The studies investigated the association of a broad range of ACEs with asthma onset and, therefore, employed a broad variety of measurement tools. 11 of the 12 studies used prenatal recruitment and one study recruited asymptomatic children aged 5–9 years. The results of these studies are organised into categories and considered separately below. Studies are referred to by their first author and year as in the online supplementary material for ease of reference.

Parental mental health difficulties and asthma onset

Four studies examined the relationship between mental health difficulties among parents and the onset of asthma symptoms in their children. Calam et al. [16] found no significant effect of self-reported depression or anxiety (using the Hospital Anxiety and Depression Scale and the general health questionnaire). This study employed a relatively small sample size (n=411) and measurements were taken at a single time point when the child was 3 years of age. This study found a highly significant effect of behavioural difficulties preceding the onset of parent-reported wheeze (OR 8.95). Kozyrskyj et al. [17] used medical records to investigate the link between maternal depression and/or anxiety diagnosis and physician diagnosis of asthma in their children. At 7-year follow-up, the authors report a mildly significant effect (OR 1.25) with subsequent analyses revealing a dose–response relationship between intensity and severity of mental health difficulties and likelihood of asthma diagnosis. While this study was based on a large sample size (n=13 907), the analysis was not adjusted for maternal health behaviours such as smoking. Both Guxens et al. [18] and Cookson et al. [19] investigated the impact of maternal mental health difficulties in pregnancy, employing large sample sizes (n=4848 and n=5810, respectively). Guxens et al. [18] assessed maternal distress in pregnancy using the self-reported brief symptom inventory and reported an increased risk of parent-reported wheeze at 6-year follow-up (OR 1.6). Similarly, Cookson et al. [19] found an increased risk of physician-diagnosed asthma at 7.5-year follow-up associated with higher anxiety measured on the Crown-Crisp Experiential Index (OR 1.64). Guxens et al. [18] propose an intrauterine programming effect of prenatal stress exposure to explain this increased vulnerability and found no significant effect of paternal distress during the partner's pregnancy.

Parenting difficulties and asthma onset

Four studies examined the relationship between parenting difficulties and the onset of asthma symptoms in children. Three of these studies [15, 20, 21] relied on parental self-reporting via the perceived stress scale (PSS) and the parenting stress index (PSI). One study [22] employed a clinician-administered interview to assess parenting risk which considered a range of variables including emotional availability, behavioural regulation strategies, commitment to child care and psychiatric history. Three of the four studies used physician diagnosis of asthma at follow-up as the outcome variable, whereas Wright et al. [21] assessed asthma status via parental reporting of wheeze. Sample sizes varied between 145 [22] and 2497 [20] with follow-up intervals ranging from 14 months [21] to 8 years [22]. One study measured serum IgE levels in combination with physician diagnosis of asthma, finding a highly significant correlation [22].

A study by the National Institute of Child Health and Human Development Early Child Care Research Network [15] found no significant association between self-reported PSI 1 month after the child's birth and asthma onset in any stage of childhood, concluding that parental stress does not exert an effect on asthma onset. Shankardass et al. [20] separated their analyses into children who are exposed to either high or low levels of traffic-related air pollution (TRAP), derived by geo-coding addresses according to traffic density and locally measured CO₂ levels. Parental stress (derived via self-report of PSS) was categorised into four quartiles and the authors use the lowest quartile as a reference category in their analysis. They conclude that among those exposed to high TRAP, only those children whose parents report high levels of stress are at increased risk of developing physician-diagnosed asthma (HR 1.51). Children in the high stress, low TRAP group are not at an increased risk of asthma diagnosis leading the authors to suggest that chronic exposure to parental stress may lead to an immunological vulnerability and increased inflammatory response in children. Wright et al. [21] also employed the PSS as a proxy for parental stress in their investigation of a genetically predisposed community birth cohort. In adjusted analysis, they reported a significant effect size (relative risk 1.4), although this was amongst a considerably smaller sample than Shankardass et al. [20]. In this study, PSS was dichotomised simply as yes/no and parent-reported wheeze was employed as the outcome variable. Klinnert et al. [22] employed more objective measurement strategies with regards to parenting difficulties compared with other studies in this category; parenting risk scale (PRS) was assessed by clinician interview. Klinnert et al. [22] reports that PRS is significantly associated with physician-diagnosed asthma onset at 6–8-year follow-up (OR 2.07), with associated asthma biomarkers (serum IgE) supporting the association. This study did, however, have the smallest sample size in this category (n=145).

Exposure to violence and asthma onset

Two studies investigated the impact of exposure to violence on the development of asthma symptoms and one study investigated the combined effects of exposure to violence and poor housing conditions. Clougherty et al. [23] measured children's exposure to violence in questionnaires completed by parents or by the child if they were over 8 years-old. This measure was primarily concerned with neighbourhood violence exposure and domestic violence was included as a control variable in analysis. The study also investigated exposure to TRAP and susceptibility to asthma onset. The authors reported an increased likelihood of physician diagnosis of asthma at 3-year follow-up for children who experienced high TRAP and high exposure to violence (OR 1.63), although experiencing high exposure to violence alone did not significantly increase asthma risk. This investigation employed a relatively small sample size (n=413) and focussed on low SES neighbourhoods in one US city. Suglia et al. [14] employed a comparatively large sample (n=2013) and investigated maternal reports of inter-partner violence. Measures were taken at 12 and 36 months, with analysis revealing that chronic inter-partner violence exposure (occurring at both time points) was significantly associated with an increased risk of physician-diagnosed asthma at 3-year follow-up. Single time point inter-partner violence exposure was not significantly related to increased asthma risk. In further analysis, housing disarray (i.e. a crowded, cluttered or dirty home environment as assessed by an interviewer at 12 months after birth) was found to be significantly associated with physician-diagnosed asthma (OR 1.5). A combination of high inter-partner violence exposure and housing disarray resulted in a multiplicative effect (OR 4.6), leading the authors to suggest that asthma risk factors multiply rather than accumulate.

Ethnic minority status and asthma onset

One study reported effect sizes for ethnic minority status as a risk factor for physician diagnosed asthma. The study by the National Institute of Child Health and Human Development Early Child Care Research Network [15] reported a significantly increased risk of asthma onset (OR 2.77) for non-Caucasians in a population birth cohort (n=984). This finding was bracketed by broad confidence intervals (95% CI 1.39–5.49) and did not provide data concerning specific ethnic groups. This finding was independent of SES and other common asthma risk factors, suggesting that stresses associated with ethnic minority status may contribute to increased disease susceptibility.

Main findings

In all, 12 studies contained in 10 articles concerning the relationship between adverse childhood experience and asthma onset among a total of 31 524 individuals were evaluated in this systematic review. Four studies investigated parental mental health difficulties, four studies investigated parenting difficulties, two studies investigated exposure to violence, one study investigated ethnic minority status and one study investigated housing conditions. Within these studies, a wide range of measurement instruments were employed using different methodologies including self-report, medical records and interviews. 10 of the 12 studies reported statistically significant effects of early life stress on asthma onset with odds ratios ranging from borderline significance to exceeding double the baseline measure. Two studies produced significant effects only under certain conditions. Shankardass et al. [20] found that a combination of high PSS and high TRAP was necessary to produce an increased risk of disease susceptibility. Similarly, Clougherty et al. [23] found that exposure to violence is not independently associated with asthma onset risk, but that exposure to violence is associated with a greater susceptibility to TRAP reactivity. These findings suggest that ACE may not be sufficient to increase asthma risk in the absence of environmental exposures, indicating a diathesis-stress explanation for the ACE–asthma link.

Interpretation of findings in relation to previously published work

Considered collectively, the results of this review indicate that exposure to traumatic stress in childhood significantly increases the risk of subsequent asthma onset. This contrasts with the conclusions of a review by Tibosch et al. [11], which investigated parental psychological characteristics in relation to both asthma onset and exacerbation, finding no conclusive evidence of an elevated asthma onset risk. This contrast may be attributable to the broader range of ACEs included in this review. Furthermore, evidence has emerged of an interaction or “synergistic” [23] effect between ACE and physical environmental risk factors; when combined with environmental exposures, experiencing ACE has the potential to elevate sensitivity and pro-inflammatory responses, resulting in a significantly increased risk of developing asthma. This may be explained by the process of biological embedding of stress [4] or allostatic load theory [24] or body programming [25], a proposal that is supported by the observation that immune biomarkers are elevated in line with parental stress and that such elevated levels are strongly correlated with asthma diagnosis [22]. Studies concerned with prenatal stress exposure produced markedly similar results, warranting further investigation into the relationship between maternal distress and subsequent disease susceptibility. Such “perinatal programming” effects are reviewed by Wright [26], who reports that maternal exposure to chronic stress can stimulate placental secretion of corticotrophin-releasing hormone, subsequently activating the hypothalamic–pituitary–adrenocortical axis in the developing fetus and potentially resulting in dysregulation or decreased glucocorticoid sensitisation. This has been empirically linked with atopic disorders including asthma, eczema and food allergies [26].

Strengths and limitations of this study

The majority of studies were conducted in the UK, USA and Canada; subsequently, the extent to which the results can be generalised to the global population are limited. The potential influence of publication bias cannot be discounted and it is possible that an historical tendency to publish significant results may have skewed findings. Studies of statistical association do not provide unequivocal evidence of causality. ACE is associated with a variety of societal factors, including SES, and such factors (e.g. smoking) may have a causal role in asthma. Only two studies [18, 20] adjusted their results to the main well-known asthma risk factors. Finally, neither doctor-diagnosed nor self-diagnosed asthma provides an entirely accurate assessment of asthma prevalence in a population.

Implications for future research, policy and practice

This review has identified two key areas of concern: 1) that exposure to chronic stress in early life is associated with increased sensitivity to environmental asthma triggers, and 2) that maternal distress during pregnancy is associated with increased risk of asthma onset. Increasing evidence supporting the role of perinatal stress programming in asthma susceptibility suggests that an improvement in holistic antenatal support for expectant mothers would help to reduce population rates of asthma prevalence. Prioritisation of mothers with asthma or a family history of asthma/atopy would potentially have a significant effect. In addition to these principal findings, there is some evidence that exposure to multiple stressors can have a multiplicative effect on the risk of asthma onset; statistical analyses indicate that concurrent multiple stress exposure increases asthma risk to a greater extent than that expected by adding together the ORs for each separate stressor. This exponential effect warrants a focus on the identification of infants experiencing multiple stressors for health professionals working closely with families.

Conclusion

The results of this systematic review indicate that exposure to perinatal or early-life stress significantly increases the risk of asthma onset alongside elevating levels of asthma-relevant biomarkers. There is emerging evidence of a synergistic effect in which high stress combines with environmental exposures resulting in asthma onset. Exposure to multiple and chronic stressors presents a particular risk for receiving an asthma diagnosis and action should be taken from the stage of conception to identify and address dysfunctional family environments.