Since 1999, the number of overdose deaths involving opioids has increased dramatically. In 2017, 47,600 people died from drug overdose involving an opioid (Seth et al., 2018b), and the rate of deaths involving prescription opioids was five times higher than that in 1999 (Seth et al., 2018a). According to a nationally representative survey conducted in 2017, about 10.3 million adults aged 18 years or older misused prescription opioids, and 1.6 million had an opioid use disorder (OUD) (Center for Behavioral Health Statistics and Quality, 2018). The associated economic costs of prescription opioid misuse and overdose are substantial. The total estimated economic burden was $78.5 billion in 2013, which is attributed to healthcare and substance abuse treatment costs, lost productivity, and criminal justice costs (Florence et al., 2016). Collectively, these findings underscore the urgent need to address factors that contribute to prescription opioid misuse and overdose.

An area with promise to prevent prescription opioid misuse is the body of work focused on preventing and addressing Adverse Childhood Experiences (ACEs)-a collection of potentially, traumatic experiences that occur during the first 18 years of life. ACEs often include exposure to abuse (emotional, physical, sexual), neglect (emotional and physical) and household challenges (domestic violence, substance abuse, mental illness, parental separation or divorce and incarcerated family member) (Felitti et al., 1998; Merrick et al., 2018). While some degree of childhood adversity is a normal and essential part of human development, chronic or repeated exposure to ACEs can result in a toxic stress response that derails optimal development by producing changes in gene expression, brain function, immune function, and coping strategies adopted (Shonkoff and Garner, 2012), which may contribute to an individual’s risk for substance misuse, including prescription opioid misuse (Norman et al., 2011).

Previous studies demonstrate robust associations between ACEs and substance misuse (Dube et al., 2003; Orsi et al., 2018). Specifically, childhood emotional abuse, physical abuse, and sexual abuse are associated with high risk of using/misusing substances in later adolescence and adulthood. In studies that focused specifically on opioid misuse, the prevalence of childhood trauma was high among people with OUD (Conroy et al., 2009). ACEs were also associated with early onset for opioid misuse among patients with OUD (Stein et al., 2017). However, the number of studies focused on the association between ACEs and prescription opioid misuse and the pathways between these experiences is limited. One previous study documented the associations between ACEs and prescription opioid misuse (Austin and Shanahan, 2018), identifying ACEs as a risk factor that predict prescription opioid misuse. However, the pathways through which ACEs are associated with prescription opioid misuse remain unclear. Mediation analyses are needed to understand these pathways to uncover the underlying mechanisms which can inform more specific public health interventions based on different mechanisms.

Previous studies have shown the association of internalizing (depression, anxiety, and traumatic distress) and externalizing (aggression, delinquency, and hyperactivity) symptoms with both ACEs and prescription opioid misuse (Hunt et al., 2017; Martins et al., 2012; Young et al., 2012). However, few studies have explicitly tested whether internalizing and externalizing symptoms mediate the relationship between ACEs and opioid misuse. In addition, understanding this mediated association is important for designing or identifying existing interventions that address internalizing and externalizing symptoms for those exposed to ACEs to prevent them from misusing prescription opioids later.

Given these connections, this study focuses on internalizing and externalizing symptoms as potential mediators of the association between ACEs and prescription opioid misuse. Internalizing and externalizing symptoms may mediate the association between ACEs and prescription opioid misuse in different ways and may uncover important implications for interventions. For example, the potential pathway through which internalizing symptoms mediate the association between ACEs and prescription opioid misuse is based on the self-medication hypothesis that prescription opioids are used to alleviate distress and anxiety disorders (Garland et al., 2015; Martins et al., 2012). Only one study has examined this potential pathway using the National Longitudinal Study of Adolescent to Adult Health data, which found that distress and anxiety disorders did not mediate the association between ACEs and prescription opioid misuse (Austin and Shanahan, 2018). The authors thought one explanation for this finding was the inadequate measures for anxiety disorders that only captured symptoms in a short time frame (the past week). Externalizing symptoms may also mediate the association between ACEs and prescription opioid misuse based on the sensation seeking hypothesis that prescription opioids are used to meet the need for excitement (Boyd et al., 2006; Horvath et al., 2004; Young et al., 2012). Sensation seeking was defined as a trait of seeking varied, novel, complex and intense sensations and experiences (Zuckerman and Kuhlman, 2000) and was linked to externalizing symptoms (Joireman et al., 2003; Wilson and Scarpa, 2011). We therefore hypothesized that some individuals exposed to ACEs were likely to develop externalizing symptoms and misused prescription opioids for sensation seeking. Past studies demonstrate that externalizing symptoms, such as aggression, are associated with substance use/misuse (Colder et al., 2013). However, to our knowledge, no research has examined how externalizing symptoms mediate the association between ACEs and prescription opioid misuse.

Although previous studies have concluded that ACEs are risk factors for prescription opioid misuse, the pathways through which ACEs are associated with prescription opioid misuse are not clearly identified. This study aimed to address the following research gaps: 1) studying the associations between ACEs and prescription opioid misuse; 2) examining how internalizing symptoms and externalizing symptoms mediate this association by examining multiple prescription opioid misuse outcomes. In the present study, we hypothesize that 1) having a history of ACEs is associated with prescription opioid misuse in adulthood; 2) both internalizing and externalizing symptoms are positively associated with ACEs and prescription opioid misuse; 3) internalizing symptoms and externalizing symptoms mediate the association.

The present study examined the association between ACEs and prescription opioid misuse and whether internalizing symptoms and externalizing symptoms mediate this relationship. To our knowledge, this is the first study to examine the mediation effect of internalizing and externalizing symptoms on the association between ACEs and prescription opioid misuse.

The first hypothesis, that history of ACEs is associated with prescription opioid misuse in adulthood, was supported. Furthermore, among our prescription opioid misuse outcomes, ACEs had the strongest direct effect on early onset of prescription opioid misuse, which is consistent with previous studies on the association of ACEs with substance use/misuse initiation (Dube et al., 2003; Ompad et al., 2005; Stein et al., 2017). This suggests that ACEs play an important role in the early initiation of prescription opioid misuse. Although the odds ratios are not large, preventing ACEs may reduce the risk of early initiation of prescription opioid misuse and thus reduce the risk of future illicit opioid use, prescription opioid injection misuse (Jones et al., 2017), and adverse health outcomes, such as OUD and overdose. Policies, practices, and programs that change conditions for children and families can prevent ACEs from occurring in the first place and may simultaneously mitigate the impact of ACE exposure on later outcomes (e.g., prescription opioid misuse) (Fortson et al., 2016).

In line with the second and third hypotheses, the mediation analysis supports that lifetime internalizing and externalizing symptoms partially mediate the association between ACEs and prescription opioid misuse. Both internalizing and externalizing symptoms were positively associated with ACEs and were positively associated with each of the prescription opioid misuse outcomes. Our finding regarding the mediating role of internalizing symptoms supports the self-medication hypothesis (Colder et al., 2013; Garland et al., 2015; Martins et al., 2012). This hypothesis states that individuals who have experienced ACEs may misuse opioids to cope with internalizing symptoms, such as depression symptoms. Our findings regarding externalizing symptoms reveal another potential pathway between ACEs and prescription opioid misuse and supports the Behavioral Disinhibition Model, which suggests that individuals with behavioral disinhibition, an inability to self-regulate and to inhibit socially undesirable actions (i.e., aggression, delinquency), are often comorbid with substance use/misuse (Iacono et al., 2008; Young et al., 2012).

In general, the mediation effect of externalizing symptoms was larger than that of the internalizing symptoms. This finding suggests that among those individuals exposed to ACEs, externalizing symptoms may play a greater role in predicting prescription opioid misuse. This is consistent with previous findings that the association between externalizing symptoms and substance misuse is stronger than that for internalizing symptoms (Colder et al., 2013; King et al., 2004). Moreover, intervening with children who are exposed to ACEs to prevent externalizing symptoms, such as bullying or aggressive behaviors, may reduce their risk of misusing prescription opioids (David-Ferdon et al., 2016; Pophillat et al., 2016). For instance, a classroom-based program, Life Skills Training (LST), is effective at improving social skills and reducing risk behaviors including drug use and might also be effective at mitigating opioid misuse (David-Ferdon et al., 2016). A combination of LST and Strengthening Families Program (SFP), another evidence based prevention program that aims to improve children’s social skills, was implemented and demonstrated protective effects for prescription opioid misuse into early adulthood (Spoth et al., 2013).

This study also found that the mediation effects of internalizing symptoms and externalizing symptoms on opioids outcomes were heterogeneous. For example, internalizing symptoms played a relatively smaller role in mediating the early onset of prescription opioids, and a larger role in mediating both past-year and lifetime OUD. This suggests that interventions to prevent internalizing symptoms may have differential effects on different prescription opioid misuse outcomes. In contrast, externalizing symptoms had a much larger effect on early onset of prescription opioids than internalizing symptoms, which implies that efforts to prevent externalizing symptoms may be more effective in reducing the chance of early onset of prescription opioid misuse than internalizing symptoms. The heterogeneity implies that different approaches to preventing internalizing and externalizing symptoms may lead to different effects on prescription opioid misuse outcomes. Nevertheless, it remains important to prevent ACEs given the direct relationship between ACEs and opioid misuse.

There were limitations of the present study. First, we were not able to determine the temporal sequence of internalizing and externalizing symptoms, because respondents were asked whether those symptoms ever happened during their lifetime. It is possible that individuals may experience ACEs, misuse prescription opioids and internalizing and externalizing symptoms simultaneously. It is also possible that internalizing or externalizing symptoms developed after the onset of prescription opioid misuse for some individuals as a result of their misuse or life factors. Future longitudinal research to disentangle the temporal relationships among the ACEs, internalizing and externalizing symptoms and prescription opioid misuse are warranted. Second, given the complex statistical analyses, survey weights were not used in this study, and as such, our results may not yield national estimates, which limit generalizability. Third, NESARC-III was conducted in 2012–2013 and therefore our findings may not fully reflect the more recent opioid epidemic (e.g., rapid increase in overdose deaths involving synthetic opioids such as illicitly manufactured fentanyl since 2013). In addition, we used a conservative criterion (Dong et al., 2004; Roos et al., 2013) to establish ACE exposure. While this resulted in a lower ACE score, it may provide a truer sense of exposure to trauma. Despite these limitations, this study still provides important findings to inform the development of public health interventions.

Efforts to support safe, stable, and nurturing relationships and environments that can prevent ACEs may also provide an upstream approach to reducing risk for internalizing and externalizing symptoms and prescription opioid misuse (Centers for Disease Control and Prevention, 2014). In addition, interventions that prevent or treat mental health symptoms (e.g., trauma-focused cognitive behavioral therapy, multisystemic therapy) (Cary and McMillen, 2012; van der Stouwe et al., 2014) may reduce the risk of misusing prescription opioids for those already exposed to ACEs. Taken together, these findings highlight the importance of preventing and mitigating the impact of ACEs as a component of the public health strategy for addressing the opioid overdose epidemic.