Adverse childhood experiences and sources of childhood resilience: a retrospective study of their combined relationships with child health and educational attendance

An increasing body of literature describes adverse childhood experiences (ACEs) and their impact on ill health later in adult life [1–3]. ACEs include suffering childhood abuse or neglect as well as environmental stressors such as living in a household affected by substance use or domestic violence. Chronic toxic stress resulting from ACEs can impact on the neurological, immunological and hormonal development of children [4, 5]. Repercussions of such impacts include substantive increases in risk of adopting anti-social and health-harming behaviours, accelerated development of chronic disease and premature death [1, 2]. Consequently, individuals with ≥4 ACEs in childhood (compared to those with none) are, as adults, more than twice as likely to smoke, nearly six times as likely to be problem alcohol users and over twice as likely to develop conditions such as cancer and heart disease [3]. These life course perspectives on ACEs expose their long-term health and financial costs to individuals and public services [6, 7]. Further, research is increasingly identifying more immediate impacts of ACEs on a wide range of health and social outcomes during childhood [8–10].

ACEs can result in direct physical damage to a child (e.g. fractures). However, children who have suffered or are suffering ACEs also experience higher levels of asthma [11] gastrointestinal conditions [12], headaches [13] and other somatic complaints [10] as well as lower overall levels of good health [8, 14]. Typically, increasing numbers of ACEs has a positive relationship with increasing health problems [10]. Developmental evidence supports ACEs as causal factors in such problems [4, 5]; although children with health difficulties are also at increased risk of ACEs [15]. Further, ACEs are associated with poorer childhood mental health, attendance at school, educational attainment and anti-social and violent behaviours [8, 16–18]. Such factors can be part of a life course that connects childhood adversity with long-term adult ill health.

Most children exposed to ACEs do not develop poor health outcomes. A range of factors may moderate the impact of ACEs on life course health, providing resilience to developmental harms and consequently, better outcomes despite a history of multiple ACEs [19]. Although many definitions are available [20], resilience typically describes the ability to adapt successfully to disturbances that threaten development of a positive life course or the ability to resume one following periods of adversity. Sources of resilience can include, but are not limited to, cultural engagement, community support, opportunity to control one’s personal circumstances and access to a trusted adult throughout childhood who can provide sanctuary from the chronic stress of ACEs [19, 21–23]. The concept of developing resilience in children as a moderator of ACE harms is widely advocated [19, 24, 25]. A range of interventions aim to enhance resilience through supporting parents; strengthening links with other family members, peers and schools; developing team working, decision-making abilities and confidence; and enhancing academic, athletic and other individual strengths [19, 26, 27]. However, relatively little empirical data is available on the extent to which resilience assets can counteract the impact of ACEs in childhood or which assets are most effective.

Using a national study on ACEs and resilience in Wales we examine if a history of ACEs is associated with increased levels of common childhood health conditions. We measure the relationships between ACEs and a history of childhood asthma, allergies, headaches and digestive disorders; each of which has previously been linked with childhood adversity [10–13] and is also a relatively common childhood condition in the study population. We examine the broader measures of overall poor childhood health status and poor school attendance. We test whether seven community resilience assets show protective relationships with each child health condition, overall health status and school attendance. Finally, we discuss which resilience assets have the potential to counter harms associated with chronic toxic stress.

We identified strong relationships between ACE counts and common childhood health conditions. For asthma, headaches and digestive conditions results show increases in prevalence with ACEs similar to those identified for childhood conditions in other studies; despite methodological differences in data collection [9–11]. While we found no ACE studies for allergies, previous work has described relationships between individual childhood stressors (e.g. physical abuse) and increased prevalence of allergic conditions [36]. Results here suggest ACEs are positively associated with childhood allergies; although the relative increase was not as great as for other conditions measured (Table 2). Common childhood conditions are a substantive component of pressures on primary care in the UK and other countries [37]. Consistent with an ACE aetiology, children presenting with such physical conditions often respond to interventions aimed at improving mental health rather than addressing their symptoms directly [38]. A larger mental health component in the development in children of conditions such as headaches and digestive conditions may in part underpin their stronger relationship with ACEs than asthma or allergies (Table 2) [3]. However, there has been little consideration of when or how ACE-related histories might be collected in general health care settings in order to better understand their contribution to common health conditions, reduce further childhood adversity or improve patient outcomes [39, 40].

Along with the physical conditions measured here, individuals with higher ACEs are also likely to suffer other (e.g. mental [17]) health and social problems. We used a broader measure of poor childhood health to capture wider impacts on health and also measured school absenteeism as a proxy for both health and social issues (e.g. ill health, school exclusion, truanting [41]). Proportions reporting poor childhood health increased more than three-fold from 0 to ≥4 ACE categories and approximately six-fold for high school absenteeism. Even in those with 2–3 ACEs (16.2% of people surveyed) levels of poor childhood health and high school absenteeism were more than doubled (vs. 0 ACEs, Table 1). Increases were maintained after correcting for potential demographic confounders (Table 2). Our results suggest schools might improve attendance by adopting approaches that tackle ACEs and their consequences. Some schools working with public health agencies have adopted ACE-informed approaches (i.e. staff understanding ACEs and adapting working practice to support children who have experienced them). This has increased attendance, reduced exclusions and improved educational attainment [42]. However, little has been done to adopt ACE-informed school models at scale.

With the eradication of ACEs a long-term goal, developing individual and community assets with the potential to counteract the negative outcomes associated with ACE exposure is an immediate consideration. Developing resilience has been widely considered as the principal component of such assets; despite little empirical data on its benefits. Results here support greater access to childhood resilience assets being strongly related to better health and well-being (Table 3) with greater childhood resilience assets having a positive impact on each health and well-being measure (Table 4). However, which resilience assets made significant contributions varied by outcome measure. Thus, no resilience assets measured here significantly affected levels of asthma whilst different assets were most strongly associated with each childhood health condition (e.g. supportive friends with lower prevalence of headaches; Table 4). Moreover, poor self-reported childhood health and high school absenteeism were both significantly related to multiple resilience assets. For poor childhood health status reporting all significant assets (i.e. given opportunities, having supportive friends and a role model) was associated with approximately a two thirds reduction in adjusted prevalence of poor childhood health across all categories of ACE count (Fig. 1). Similar benefits in school attendance across all ACE count categories were associated with being given opportunities but also being treated fairly in childhood communities. These findings suggest that those resilience assets measured here provide a similar level of benefit to individuals irrespective of their exposure to ACEs. Thus, having good role models, providing networking opportunities and settings for friendship building as well as ensuring a sense of fairness and equity in how children feel they are treated may not just be enhancements to childhood but could be essential to health and educational attendance. With all individuals appearing to benefit from accessing resilience assets and those with most ACEs presenting with the worst outcomes, well-being strategies should consider how resilience assets can be supported universally but to a greater extent where childhood adversity is concentrated.

However, this study faced a number of limitations. In particular, data were retrospective and self-reported and consequently may be affected by recall bias or unwillingness to report past experiences. An analysis of individual ACEs reported by age identified prevalence of some ACEs reducing with increasing age (e.g. verbal abuse, parental separation, drug and alcohol abuse) while others showed no clear age-related trend (e.g. domestic violence, sexual violence; Additional file 1: Table S3). Some ACEs are known to have increased in recent years (e.g. parental separation). However, separation is only a proxy measure for potentially chronic stress in family environments and the absence of separation may, in some cases, increase or extend periods of familial conflict. The extent to which trends in others (e.g. verbal abuse) are real or a function of recall bias cannot be established here. Studies comparing retrospective and prospectively collected measures of childhood adversity find both correlate with poor health outcomes in later life, with retrospective and prospective measures also showing moderate correlations with each other [43]. While retrospective measures inevitably risk incomplete or inaccurate recollection, prospective studies typically rely on reporting by caregivers or professionals and may also miss cases. Moreover, as prospective studies identify current child abuse or neglect they may intentionally or unintentionally alter its impact on current or later health outcomes. However, whilst we cannot measure or correct for variability introduced by using retrospective ACE, resilience and outcome measures this remains a limitation of the study. Further, our resilience measures showed similar beneficial relationships across those with and without ACEs. We cannot tell if such benefits related to resilience impacting unmeasured adversities in those with few or no ACEs or if resilience assets are beneficial regardless of ACE exposure. The study population was nationally selected using a random, stratified methodology. The generalisability of results to populations outside of Wales or those based in settings not usually accessed through door to door surveys (e.g. prisons, care settings) also cannot be established without further studies. Unlike ACEs, presently there is no validated tool for measuring cumulative resilience retrospectively. Consequently, we used a combination of questions from more commonly used tools including those previously examining associations with ACEs. Further, while this study identified resilience assets associated with improvement in each health and educational outcome (except asthma), which assets were significant varied with outcome (Table 4). This study could not elucidate why different resilience assets appear to protect against different harms; although this is an important area for further study. Finally, further studies with more detailed longitudinal data are also required to examine causality including examining the potential for reverse causality where those with better health subsequently adopt more resilience building opportunities.

While ACEs can shape health across the life course, results here suggest their harmful impacts may manifest through increased common childhood health conditions as well as general poorer self-rated health. Such conditions are typically not life threatening but their long-term impact on physical and social development can be substantive. Thus, childhood ill health as well as anti-social behaviour impacts school attendance and consequently opportunities for educational attainment and better economic prospects throughout life. Integrated and trauma-informed (i.e. understanding, recognising and responding to the links between a history of trauma and current health and social problems) public services can provide support capable of preventing ACEs and potentially develop resilience assets that offset some of their harmful consequences [44]. Population surveys on ACEs can usefully inform national policy and the use of a combined ACE measure provides a framework for multi-agency engagement [45]. However, person level support for those at risk of, or exposed to, trauma require studies directly assessing the acceptability and effectiveness of interventions. Already parental support, early social skills development and pre-school enrichment interventions have all been associated with reduced risk of ACEs. For those who experience ACEs, trauma-informed public services are critical for understanding and addressing the underlying causes of health and behavioural problems seen in both educational and care settings. Currently, the eradication of ACEs remains beyond the scope of most communities. However, investment in assets that build resilience may counteract some of the harms disproportionately suffered by those with ACEs and results here suggest may also be of benefit to those with low or no ACEs. Many of the resilience assets examined in this study are features of thriving communities. Asset-based community development (ABCD) aims to enhance existing beneficial features within localities to enable residents to overcome the challenges they face [46]. An ABCD approach to resilience development would identify and invest in existing features that support cultural connectedness, friendship networks, community roles models and access to community support. The return on such investments could be substantive both in the short-term though improvements in child well-being and through the long-term gains from setting individuals on a healthier life course. While public policy and services consider how to effectively support such developments we should at the least ensure we do not disinvest from or dismantle community features that may be instinctively protecting some of the most vulnerable children.