This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Research Protocols, is properly cited. The complete bibliographic information, a link to the original publication on http://www.researchprotocols.org, as well as this copyright and license information must be included.
Adverse childhood experiences are prevalent robust risk factors for the development of substance use problems. However, less is known about the causal mechanisms that explain these relationships. While directly preventing adverse childhood experiences is ideal, it is not always possible. In such cases, the mechanisms themselves may be amenable to intervention, allowing for the effective prevention of problematic substance use among children exposed to adversity. Identifying such mechanisms is therefore a critical step for efforts aiming to reduce the high individual and societal burdens associated with substance use globally.
This study aims to systematically identify and synthesize evidence on the modifiable mediators and moderators of the relationship between adverse childhood experiences and substance use outcomes in young people (age 10-24 years).
A systematic review will be conducted using PubMed, MEDLINE, PsycINFO, Web of Science, and CINAHL databases to determine the modifiable mediators and moderators of the relationship between adverse childhood experiences and substance use in young people. Data from the review will be qualitatively synthesized, unless we identify a sufficient number of studies (at least five) that examine the same type of adversity (eg, physical or sexual abuse) and the same mediator/moderator, in which case a quantitative synthesis (meta-analysis) will be conducted. If a quantitative synthesis is warranted, standardized effect estimates of the indirect (mediated) effect between adverse childhood experiences and substance use outcomes will be combined using a random-effects meta-analysis. Mediators/moderators will be grouped according to a socioecological perspective, using the four levels of individual, interpersonal, community, and public policy/culture.
Electronic searches were completed in August 2019. A total of 4004 studies were included for screening after removing duplicates. After evaluating titles and abstracts against eligibility criteria, a further 3590 studies were excluded, leaving 415 studies for full-text screening. The results of the review are expected to be available by December 2020.
The mechanisms linking adverse childhood experiences and substance use outcomes in young people are vital targets for substance use prevention efforts. This review will provide evidence to inform the development of prevention strategies in order to interrupt the negative life trajectory that can begin with childhood adversity.
PROSPERO International Prospective Register of Systematic Reviews CRD42020148773; https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42020148773
DERR1-10.2196/22368
Children exposed to adversity are at greater risk of developing substance use problems later in life compared with children not exposed to such adversity [
Adolescence and emerging adulthood are critical periods for intervention and prevention of substance use problems, given the relatively young age of onset of substance use and development of substance use disorder symptoms [
However, effective prevention of substance use problems in young people with a history of adverse childhood experiences is lacking. This may be in part due to a lack of clarity around specific targets for prevention in this population. Mechanisms that mediate or moderate the relationship between adverse childhood experiences and substance use outcomes and are amenable to change reflect key targets for prevention. Existing research points to such mechanisms at the individual, interpersonal, and community levels of behavior. Specifically, early adversity has been linked to changes in inhibitory control and reward processing, which in turn predict vulnerability to substance use disorder [
The literature reviewed above provides some candidates for intervention in the pathway from childhood adversity to substance use outcomes. However, the existing literature on mechanisms involved in the relationship between adverse childhood experiences and substance use outcomes is limited in three important ways. First, studies often examine a single mediator or moderator, or a single type of adversity, despite evidence of multiple mechanisms that contribute to the relationship [
This study aims to address these gaps by systematically identifying and synthesizing evidence on the modifiable mediators and moderators of the relationship between adverse childhood experiences and substance use outcomes in young people. Specifically, through a systematic review of the literature, this study aims to determine what modifiable factors mediate or moderate the relationship between childhood adversity and substance use outcomes in young people (age 10-24 years).
This protocol adheres to the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) 2015 statement [
For this review, eligibility criteria are defined using population, intervention/exposure, comparator, outcome, and study characteristics (PICOS) [
Included studies must include human participants (no animal studies). Participants must have experienced childhood adversity between the ages of 0 and 18 years, and must have a measured substance use outcome between the ages of 10 and 24 years, corresponding to an inclusive definition of adolescence [
Studies must include a measure of childhood adversity, occurring between the ages of 0 and 18 years. Childhood adversity is defined here as experiences measured by the CDC-Kaiser Permanente Adverse Childhood Experiences Study [
No comparator/control group required for inclusion.
Studies must include a substance use outcome between the ages of 10 and 24 years. This includes alcohol, tobacco, psychoactive drugs, and nonmedical use of prescription drugs and any of the following:
Initiation and age of initiation of substance use.
Frequency of substance use.
Problem substance use or abuse, defined as the presence of any of the following: failure to fulfil major obligations at work, school, or home; recurrent use in situations in which it is physically hazardous; recurrent substance-related legal problems; and continued use despite persistent social or interpersonal problems.
Quantity of substance use, including single occasion risky drinking (binge drinking/heavy episodic drinking, defined here as at least four standard drinks on any one occasion).
Substance use disorder/dependence. The definition for this outcome will be according to the diagnostic criteria set out in the Diagnostic and Statistical Manual of Mental Disorders (DSM) version that was in use at the time the study outcome data were collected (either DSM-IV [
Studies must include a mediation/moderation analysis of at least one factor that is modifiable after birth. We consider a mediation analysis to be present if the authors test the indirect effect from the adverse childhood experience to the substance use outcome via a hypothesized mediator. In mediation analyses, we do not require studies to first demonstrate a significant direct effect from the adverse childhood experience to the substance use outcome. This is in recognition of consensus among mediation researchers that if the direct effect is presumed small or temporally distant to the outcome, it need not be significant for mediation to be established [
In addition, studies must be published in English, must be peer-reviewed, must employ a longitudinal study design, and must be original research. Full-text studies must be published from January 1, 1998, to August 14, 2019. The year 1998 was chosen as the historical cut-off point to include only studies published following the CDC Kaiser Permanente Adverse Childhood Experiences study [
Specific exclusion criteria were identified. Studies will be excluded if they meet any of the following conditions: evaluation of intervention or treatment outcomes only; assessment of scale formation only; reporting of the incidence of substance use only; and review of the literature. In addition, studies that presume physical neglect owing to poverty status or another income-related measure will be excluded if they do not examine another adverse childhood experience. This was due to the understanding that a family’s income does not necessarily reflect whether a child’s basic physical needs of food, shelter, adequate medical care, and clothing are met.
Electronic searches will be conducted in PubMed, MEDLINE, PsycINFO, Web of Science, and CINAHL from 1998 to August 2019. Two databases (MEDLINE and PubMed) will be searched without English language restriction to determine whether relevant studies published in other languages are being excluded. Searches will be rerun in 2020 prior to data analysis to identify any relevant studies published since the initial searches were conducted.
MEDLINE search strategy.
Number | Term |
1 | life change events/ |
2 | adverse childhood experiences/ or domestic violence/ or exp child abuse/ or physical abuse/ |
3 | ((childhood or adolescent) adj3 advers*).tw. |
4 | (child* or life or early) adj2 stress.tw. |
5 | bullying/ or cyberbullying/ |
6 | social isolation/ |
7 | ((family or parent*) adj3 (substance or alcohol* or drug or smok* or depression or illness or suicid* or jail or prison)).tw. |
8 | divorce/ or family conflict/ or family separation/ |
9 | (trauma* or maltreat* or assault* or violen* or molest* or neglect* or victim* or isolat* or reject* or mistreat* or poverty or depriv* or abus* or lonel*).tw. |
10 | 1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9 |
11 | resilience, psychological/ |
12 | adaptation, psychological/ |
13 | (adapt* or protect* or resilien* or mediat* or moderat*).tw. |
14 | protective factors/ |
15 | 11 or 12 or 13 or 14 |
16 | substance-related disorders/ or exp alcohol-related disorders/ or alcoholic intoxication/ or alcoholism/ or binge drinking/ or amphetamine-related disorders/ or cocaine-related disorders/ or drug overdose/ or inhalant abuse/ or marijuana abuse/ or exp opioid-related disorders/ or phencyclidine abuse/ or substance abuse, intravenous/ or substance abuse, oral/ or “tobacco use disorder”/ |
17 | ((substance or alcohol* or tobacco or drug or smok*) adj3 (misuse* or initiat* or abus* or problem or heavy or binge or disorder* or dependen* or frequen*)).ti,ab. |
18 | 16 or 17 |
19 | cohort studies/ or longitudinal studies/ or follow-up studies/ or prospective studies/ or retrospective studies/ or cohort.ti,ab. or longitudinal.ti,ab. or prospective.ti,ab. or retrospective.ti,ab. |
20 | 10 and 15 and 18 and 19 |
21 | child* or adolescen* or teen* or youth* or pediatr* or paediatr* or young or emerging or youth).tw |
22 | 20 and 21 |
23 | limit 22 to ((“all child (0 to 18 years)” or “young adult (19 to 24 years)”) and English) |
24 | limit 23 to yr=“1998 -current” |
If additional information is required from authors of studies identified by the review, the corresponding author of that study will be contacted by email to provide this information. After 3 weeks from the date of the first contact and one follow-up email, if no response has been received from the authors, we will deem them unreachable and proceed with our analysis. If any of the inclusion criteria cannot be confirmed, the study will be excluded.
Studies identified in the databases will be exported to Covidence [
Researcher 1 will screen 100% of the titles and abstracts for inclusion in the review. A second and third researcher will screen 5% of the titles and abstracts, and proportionate agreement will be calculated through the systematic review software [
Full-text studies will be obtained for studies deemed eligible for inclusion. Two researchers will read all full-text studies. Discrepancies in the scores of the two researchers will be resolved through consultation, and if required, a third researcher will be included.
From the studies included in the final selection, researcher 1 will independently extract author information, publication year, study characteristics (sample size, age of participants, gender, length of follow-up, and location), substance use outcomes (age of initiation, any use, frequency of use, problem use, heavy use or binge drinking, abuse, and disorder/dependence), characteristics of child adversity (type, age of exposure, and duration of exposure where available), mediators examined, and moderators examined. A summary of the findings for each mediator/moderator will be extracted, including statistical significance and the effect size of the mediated and/or moderated effect.
Study quality will be assessed using the Joanna Briggs Institute Critical Appraisal Checklist for Studies Reporting Prevalence Data [
We will synthesize mediators and moderators. These will be grouped according to the socioecological model [
The GRADE (Grading of Recommendations, Assessment, Development, and Evaluations) approach will be used to assess the strength of the evidence overall [
This study does not include human or animal participants and thus does not require approval from ethical review boards.
Electronic searches were conducted in August 2019. After removing duplicates, 4005 studies were included for screening. On evaluating titles and abstracts against eligibility criteria, a further 3590 studies were excluded, leaving 415 studies for full-text screening. The results of the review are expected to be available by December 2020 (
Progress and timeline (adapted from the PROSPERO International Prospective Register of Systematic Reviews).
Month, Year | Aug-Dec, 2019 | Jan-Mar, 2020 | Apr-Jun, 2020 | Jul-Sep, 2020 | Oct-Dec, 2020 |
Preliminary searches | ✓ | ||||
Piloting of the study selection process | ✓ | ||||
Formal screening of search results against eligibility criteria | ✓ | ✓ | ✓ | ||
Data extraction | ✓ | ||||
Risk of bias assessment | ✓ | ||||
Data analysis | ✓ |
Substance use and mental disorders are the leading causes of disease burden in young people and are associated with 3.4 million years of health lost [
Database search terms.
The authors would like to thank Jessica Hughes, Academic Liaison Librarian, for her advice regarding the search strategy. This research was supported by the National Health and Medical Research Council (NHMRC) Centre of Research Excellence in Prevention and Early Intervention in Mental Illness and Substance Use (PREMISE; APP1134909). LRG is supported by an Australian Government Research Training Program Scholarship. The funders have no input in the conduct of this review.
LRG conceived the idea of this study, with input from EVK and NCN. ELB and KMK provided advice on technical aspects of the planned review. All authors contributed to the final manuscript.
None declared.