1. Introduction
Mental health and substance use disorders are the 6th leading contributors to the global burden of disease in young people below the age of 24 [1], measured as disability-adjusted life years (DALYs). Alcohol is the most widely used psychoactive substance in adolescent populations [2]. Here, alcohol use will be used to refer to any use, experimentation and irregular to frequent heavy use, which may reach clinical levels for abuse or dependence [3,4]. The co-occurrence of alcohol use and mental health problems are associated with poor school performance and dropout [5,6], legal problems [5], suicidal ideation [7,8], poorer treatment outcomes [9] and poorer health outcomes compared to those young people with either alcohol use or mental health problems separately [10,11]. Further co-occurring alcohol use and mental health problems in young people can lead to longitudinal effects into adulthood [12]. Consequently, adolescence provides an optimal opportunity to intervene with the potential to impact the entire lifespan [13].
Mental health problems, including internalizing problems (emotional problems) and externalizing problems (behavioral problems), frequently co-occur with alcohol use in young people [14,15,16,17]. A systematic review reported that up to 60% of young people aged 14 to 18 years who engage in alcohol and other substances also have co-occurring mental health problems (internalizing or externalizing problems) [6]. More recently, an England-based survey in 2017 found that rates of alcohol use and frequency of alcohol use were higher in those young people with clinical levels of mental health problems compared to those without [17]. Specifically, 36% of young people aged 11–16 with a formally diagnosed mental disorder had tried alcohol, compared to 22.7% without a mental health disorder [17]. Similarly, 31.7% of young people with a mental health disorder were more likely to drink monthly, in contrast to 19.4% of those without a mental disorder [17]. However, these estimates may be an underrepresentation as many people experience co-occurring problems without meeting the threshold for a diagnosis by a health and care professional [18]. With mental health and alcohol use problems presenting on a continuum, sub-threshold levels can still lead to detrimental outcomes [19]. Examining sub-threshold levels, Lewinsohn and colleagues reported that 33% of young people with conduct problems and 27.8% of young people with depression had co-occurring alcohol problems [18]. Studies such as Lewinsohn et al., which report subthreshold prevalence estimates for young people, are few [20].
One theoretical model delineating the possible cause of co-occurring mental health problems and alcohol use is the common factor model. This model suggests that risk and protective factors are not problem/disorder specific but rather that alcohol and mental health problems may be a result of common underlying factors, including factors relating to the family [13]. Please see Figure 1. Whilst studies suggest that heritability plays an important role in co-occurring alcohol use and mental health problems [21,22], external factors have also been found to be key. Studies have identified protective factors, including emotional closeness, bonding with family, carers rewarding good behavior and family cohesion [23], opportunities and rewards for prosocial involvement and attachment to be associated with subthreshold and clinical levels of co-occurring alcohol use and mental health problems. They have also identified a range of familial risk factors to be associated with co-occurring alcohol use and mental health problems. These include family conflict, family history of antisocial behavior and substance use [20,24], parental attitudes favorable toward drug use and antisocial behavior [24], poor family management, and low levels of familial support [25]. Targeting these risk and/or protective factors offers an opportunity for the prevention and early treatment of a broad range of outcomes, including both alcohol use and internalizing and externalizing symptoms [13]. Increasingly, the need for the prevention of alcohol use and mental health problems, alongside treatment, is recognized [13,16,26]. Prevention aims to delay the onset or initiation and reduce levels of symptoms before they reach a diagnostic threshold [27]. As such, this review encompasses both.
2. Materials and Methods
The review protocol was preregistered on Prospero—CRD42016039147
2.1. Eligibility Criteria
Studies were deemed eligible for inclusion if they:
- Targeted young people aged 12–17 of any ethnicity or gender. Twelve years was selected as the lower cut-off as it is the common age of onset for both alcohol use and mental health problems [36,37]. Seventeen was selected as the upper age limit as family does not always remain a key influential context beyond this age. This is partly due to no longer having legislated age-related restrictions in the UK and other European countries [38]. Therefore, alcohol use may be less dependent on the family’s influence. Trials that had a broader age range were included if the mean age of participants fell between 12–17 years.
- Reported on a family-involved psycho-social intervention in which a young person and caregiver were included, either separately or together, in at least one session. A broad definition of family was employed to include parents, carers, grandparents, aunts, uncles and siblings. All levels of prevention and treatment were included to ensure a more thorough evaluation and to enable comparisons between these three levels of family-involved psycho-social interventions. These levels include: ‘universal prevention’ targets the entire population irrespective of risk [39]; ‘targeted prevention’ consists of ‘selective’ interventions [39]; targeting individuals at risk and ‘indicated’ interventions [39]; individuals with pre-existing symptoms or pre-clinical diagnoses, with the aim of reducing alcohol use and mental health problems before it reaches a diagnostic threshold [39] and ‘treatment’ is aimed at individuals with a diagnosis [13]. Levels of prevention can be considered to be on a continuum, with the levels merging into one another rather than occurring as distinct alternatives [40].
- Reported on both the primary outcomes: alcohol consumption (including frequency of drinking, binge drinking defined as drinking five or more drinks on any one occasion, regular or problem drinking) and common adolescent mental health problems including (a) internalizing problems such as anxiety and depression as well as (b) externalizing problems such as conduct problems and ADHD symptoms. Outcome measures could either report on the specific mental health problems or the overall internalizing or externalizing symptom score. Secondary outcomes included other substances and family functioning.
- Had a robust evaluation design, specifically randomized controlled trials (RCTs), controlled trials, randomized trials (RTs) and quasi-experimental trials. Trials that included active controls (such as a different variant of the same intervention or a different kind of therapy) were defined as RTs, and those employing inactive controls (such as no treatment, waitlist control and standard care) were defined as RCTs in this review [41].
Trials were excluded if they were limited to young people with experience of trauma, such as sexual assault, domestic violence and abuse; or specific care needs, e.g., autistic spectrum disorder, learning difficulties or cancer; or with unique environmental circumstances, including refugee, war-torn/disaster zone, military families and homelessness. Furthermore, trials were excluded if they did not report on a measure of alcohol use either separately or within a composite measure (alcohol and other drugs together).
2.2. Search Strategy
The following databases were searched from inception to January 2023 without language, year or publication status restrictions: MEDLINE (OVID), PsycINFO (OVID), Web of Science (EBSCO), The Cochrane Central Register of Controlled Trials (OVID), CINAHL (EBSCO), ASSIA (Proquest) and Embase (OVID). The search strategy included a combination of medical subject headings/thesaurus headings, appropriate keywords and free text terms applying boolean, proximity and truncation operators. This approach was supplemented with a search of the grey literature and relevant journals, e.g., Journal of Adolescent Health, Journal of Youth and Adolescence and Journal of Child and Family Studies. Here, combinations of the keywords developed in the search strategy were used. Citations and references of included trials were also screened. The search strategy is available as a Supplementary File.
2.3. Study Selection, Risk of Bias Assessment and Extraction
Two researchers independently screened all titles and abstracts, followed by full-text review of eligible trials against pre-specified inclusion/exclusion criteria. Please see Figure 2. Two researchers also data-extracted and appraised the methodological quality of the trials using the Cochrane Collaboration’s risk of bias tool, which assesses selection, performance, detection, attrition and reporting bias [42]. Trials were not excluded based on the quality appraisal; rather, it informed critical evaluations of the conclusions of included trials. A third researcher resolved disagreements arising at any stage.