PROTOCOL: A systematic review and meta-analysis of randomised controlled trials evaluating the impact of parenting programmes for parents of adolescents (10-18 years) on adolescent mental health outcomes, positive development and the parent-adolescent relationship.

Abstract

Mental health problems among adolescents are highly prevalent across the world, with international rates ranging from 10% to 20% (Collishaw, 2015; Kieling et al., 2011). In Australia, the 12-month prevalence of mental disorders has been estimated at 13.9% with the most common disorders being behavioural (e.g., attention-deficit/hyperactivity disorder and conduct) and anxiety disorders (Lawrence et al., 2016). Similar overall and disorder-specific prevalence rates have been reported in the United States (Ghandour et al., 2019), the UK and Europe (Collishaw, 2015), with data indicating stability in prevalence over time (Ghandour et al., 2019; Lawrence et al., 2016). Clinical diagnosis and treatment of mental health disorders in adolescents has also been shown to be increasing over recent decades (Collishaw, 2015) indicating significant burden on individuals, families, communities and health systems. Adolescent mental health problems have far-reaching and potentially long-lasting negative effects on the adolescents themselves, their families, and the wider community. Such adolescents are more likely to become socially and academically disengaged (McGue & Iacono, 2005; Smart et al., 2005; Smart, 2008), engage in criminal behaviour (Aebi et al., 2014; Bosick et al., 2015; Hughes et al., 2020), and have ongoing mental health, relationship and employment challenges in adulthood (Copeland et al., 2009, 2014; Hale et al., 2015; Naicker et al., 2013; Patton et al., 2014). Adolescents have also been demonstrated to engage in alcohol and substance use at relatively high levels (Guerin & White, 2020; Swendsen et al., 2012). Substance use and mental health problems such as depression have a bi-directional relationship with significant implications for adolescent wellbeing and transition to adulthood (Isaksson et al., 2020; Liu et al., 2018). Further, Sellers et al (Sellers et al., 2019) in their study of cross-time changes in outcomes following mental health problems during childhood using three UK longitudinal cohort studies and covering four decades of child mental health prevalence data, demonstrated that the experience of mental health issues during childhood has become more strongly associated with social, educational and mental health problems over time. Arguably, current rates of adolescent psychological, social, and educational issues indicate that many adolescents worldwide are “slipping through the cracks” and not being well-served by the delivery of prevention, early intervention or treatment programmes. Reducing the prevalence of problems among adolescents is only one part of addressing life-long risk and poor outcomes. It is equally as important to promote the social, psychological and self-regulatory skills and attributes that allow adolescents to become successful learners, build healthy relationships and to become healthy and productive contributors to their families and communities. Multiple studies and reviews of positive development have identified a range of psychological and emotional characteristics that promote positive mental health and social development in young people, including self-regulation and self-esteem, coping and persistence, responsibility and decision-making, problem solving, motivation and achievement, having a future orientation; and, connectedness to peers, family and community and to institutions such as educational facilities (Arnett, 2000; Ciocanel et al., 2017; Curran & Wexler, 2017; García-Poole et al., 2019; Lerner et al., 2009; O'Connell et al., 2009; Sardiñas et al., 2017). While the call for greater focus on health prevention and promotion has increased in recent years, there is still a relative lag in the application of this work to adolescence in comparison to adults (Van Allen et al., 2017). Much of the literature examining interventions for adolescents has focused on prevention and treatment programmes targeted directly to adolescents with an emphasis on reducing or treating mental health and social problems. While systematic reviews have indicated evidence for interventions targeting adolescents with anxiety (James et al., 2015), systematic reviews of others, such as depression (Cox et al., 2014), social skills training for adolescents with ADHD (Storebø et al., 2019) or school-based interventions for substance disorder (Carney et al., 2016) have shown mixed effects or inadequate ability to assess outcomes due to low quality studies. Prevention effects are also unclear. For example, Hetrick et al. (2016) Cochrane systematic review of the effects of cognitive behavioural therapy (CBT), third-wave CBT and interpersonal therapy for preventing depression found only modest effects for short-term positive effects with targeted populations (i.e., those with depression symptoms), and no effects for interventions delivered to universal populations. Exploration of the effects of these programmes is also complicated by low quality research designs. There is also a lack of good quality studies exploring the effects of interventions on adolescent positive development (Ciocanel et al., 2017). Thus, while directly targeting adolescents offers benefits for some conditions and at some population levels (treatment and targeted prevention), other types and targets of intervention are also important to consider. The family context and parenting are two key targets for intervention. There is considerable evidence to suggest that harsh, ineffective parenting and/or poor-quality relationships between parents and children are important precursors for mental health problems in children and adolescents (Odgers et al., 2008; Yap et al., 2014), with difficulties in childhood and adolescence translating into more serious difficulties in adulthood. Adolescents who have poor quality relationships with their parents are more likely to become socially and academically disengaged, engage in criminal behaviour (Farrington et al., 2009), and have ongoing relationship and employment challenges in adulthood (Hale et al., 2015). Conversely, a close parent–adolescent relationship and parenting characterised by effective conflict management, clear communication of expectations and rules, and appropriate limit setting and monitoring, is associated with important social and academic competencies in adolescence, including academic engagement and achievement (Kelly et al., 2012), capacity to manage behaviour and emotions (Farley & Kim-Spoon, 2014), and better social and community connectedness (Smart et al., 2008). Furthermore, effective parenting has been shown to be a key factor protecting against a range of negative adolescent outcomes including truancy and other externalising behaviour difficulties (Wang et al., 2011), early sexual experience and alcohol and other drug use (Kelly et al., 2011). However, despite the known role of parenting and other family factors in child and adolescent mental health problems, there has been a lack of synthesis of the evidence for interventions that involve parenting for adolescents. Decades of carefully conducted trials have demonstrated the efficacy of parenting programmes based on social learning theory in reducing difficult child behaviours and increasing adaptive child behaviours, parent competence and wellbeing (e.g., Dretzke et al., 2009; Eyberg et al., 2008; van Aar et al., 2017) and preventing antisocial behaviour and delinquency (Piquero et al., 2016). Yet, in spite of the strength of the evidence for the continuing importance of parenting in adolescence, the parenting field has focused almost exclusively on preventing problems of childhood and adolescence by working with parents of preadolescent children (Baumel et al., 2016; Chu et al., 2012; Spencer et al., 2020; Thomas & Zimmer-Gembeck, 2007; Webster-Stratton & Taylor, 2001). This lack of attention directed towards developing programmes for parents of adolescents is likely due to the pre-eminence of early childhood as a critical intervention point, in conjunction with a belief that parental influence diminishes over time as adolescent behaviour becomes increasingly individually determined and peer influence strengthens (Kazdin, 2008). Given the significant and persistent challenges associated with mental health problems in adolescents, a clear understanding of the role and/or potential of parenting interventions with this age group is needed. Thus, it is essential to explore the evidence relating to the availability and effectiveness of parenting programmes that specifically target the adolescent years. This review will include any parenting intervention that aims to promote parenting practices and the parent–adolescent relationship during the developmental period from age 10–18 years. It is expected that included interventions will cover a range of theoretical paradigms and approaches, including but not limited to parenting interventions based on behavioural principles and social learning theory, attachment theory (including Interpersonal therapy), family systems theories, relational frame theory/acceptance and commitment therapy and nonviolent resistance theory. Thus, it is expected that programmes such as the Teen Triple P program (Ralph & Sanders, 2004), the Strengthening Families Program (Kumpfer & Magalhães, 2018) and the Nonviolent Resistance training model (Weinblatt & Omer, 2008) will be identified and considered for inclusion. Parenting interventions have relevance and applications across the continuum from health promotion to prevention, early intervention and the treatment of acute and chronic serious mental health disorders in young people. Parents are tasked with the ongoing responsibility of helping their child to develop the skills they need to be successful adults. Whether in the absence or presence of mental illness and other potential adverse childhood experiences, this includes providing a safe, warm and loving environment characterised by clear boundaries and effective supervision that enables the young person to grow, learn, fail and succeed. When mental health disorders are present, parents are also critical for helping their child to manage their symptoms and maintain their treatment plans (Burke, 2017). Therefore, to produce a comprehensive review of the extant evidence, we will use a relatively broad inclusion threshold for study inclusion. Specifically, the review will include studies where the intervention takes a prevention, early intervention or treatment focus (Select Committee on Mental Health, 2006). Preventive interventions are implemented prior to the initial onset of a disorder with the objective of preventing the development of the disorder by reducing risk factors and enhancing any protective factors associated with the targeted issue. Prevention interventions can be implemented at a universal (targeting the health of the whole population) or selected (targeting at-risk populations, e.g., targeting parenting programmes specifically to teenage parents) level. Early intervention programmes are implemented with individuals showing early indicators or symptoms of problems and in the context of mental health disorders, during a first episode. The aim of early intervention is to prevent the progression of issues into diagnosable disorders or other serious and long-lasting issues and to reduce the impact of any early indicators. Treatment interventions are designed to reduce, control or remediate symptoms and where possible underlying causes of issues. Further, they aim to reduce and manage the short- and long-term impacts of the consequences of mental health and associated concerns. We acknowledge that parenting programmes targeting adolescence are likely to overlap with programmes that predominantly target the adolescent directly and that contain psychoeducation components for parents. However, for the purposes of this review we specifically define parenting interventions as those that have a stated aim to improve parenting practices and/or the parent–adolescent relationship via direct engagement with parents. While interventions may include adolescent or broader family components the intervention synthesised in the final review must have a substantive parenting component that is of equal or greater dosage than any other programme component. To comprehensively synthesise the extant evaluation literature, this review will include studies where the intervention (manualised or nonmanualised) are delivered in any modality (e.g., face-to-face, group-based, individual, telehealth). Similarly, targets of the intervention may vary, providing that the parent/carer is defined as a primary rather than secondary target. Intervention targets may include: parent only, parent–adolescent dyads/triads, family-based interventions or multiple systems approaches (e.g., family and school). Adolescence is a developmental period that is characterised by significant psychological, social and physical changes that result in attainment of greater autonomy, identity and importance of relationships outside the family (Steinberg, 2017). There are a number of possible mechanisms by which parenting interventions might work in this context. Some of these are likely to be consistent with the mechanisms by which evidence-based parenting interventions work with younger children. This evidence suggests that parenting interventions using learning processes such as behavioural rehearsal (information provision, modelling, rehearsal of new skills and provision of feedback) to enhance parents' knowledge and skills result in large effects on parental behaviour and child outcomes in younger children (Wyatt Kaminski et al., 2008). These strategies increase parents understanding, confidence and capacity to implement strategies that enhance their communication and relationship with their children and as such it is likely that parenting interventions for parents of adolescents will work via similar mechanisms to encourage appropriate behaviour while assisting parents to manage more difficult issues such as risk taking. Research has for example shown that parenting interventions that target strategies such as the provision of emotional support, and the development of parenting skills that improve the relationship with the child in ways that support positive behaviour and offer strategies to deal with negative or challenging behaviours result in positive outcomes for children (Sanders et al., 2014), with the more limited evidence also demonstrating positive outcomes for adolescents (Medlow et al., 2016). Serious mental illness is common in children and adolescents worldwide and is associated with significant and wide-ranging functional impairments that continue into adulthood (Collishaw, 2015). Thus, prevention and treatment of mental health problems in children is a major international challenge. To effectively address this challenge we require more effective and evidence-based, contextually driven approaches to reducing symptoms and promoting factors associated with well-being (Burke, 2017). Parenting and the parent–adolescent relationship is a critical aspect of prevention, treatment and recovery for adolescents. In a prevention and early intervention context, parents have the task of supporting, teaching and managing their adolescents as they move towards defining their social identity and take on increased autonomy and responsibility. In treatment contexts, the parental role extends to assisting their adolescent to obtain support, to manage their symptoms and to promote activities and behaviours associated with wellbeing (Burke, 2017). This systematic review has direct relevance to policy within the authors' country of residence (Australia) and internationally. Child and adolescent health and/or mental health policies around the world (e.g., the WHO's Global Accelerated Action Plan for the Health of Adolescents and Global Strategy for Women's, Children's and Adolescent's Health (2016–2030); the UK's National Health Services' Long Term Plan; the U.S. Department of Health and Human Services Healthy People agenda) emphasise the importance of the adolescent period as a critical transition point in the prevention and treatment of mental health problems and the promotion of health and wellbeing, along with the importance of evidence-based interventions and the influential role. For example, this review will provide evidence-based information for the action of “enhancing and promoting resources and mechanisms to support parenting in the middle years and adolescence,” which was identified as a neglected and less-resourced area but still an influential period of transition where protective parenting has a strong impact on outcomes; and, for “tackling mental health and risky behaviours, such as building social and emotional coping skills in the middle years and during the transitions into adolescence and early adulthood” within the recently released Australian Government's National Action Plan for the Health of Children and Young People (2020–2030). The findings from this review will this support action within identified priority areas for governments across the world. Despite the importance of parents and parenting in the protection and risk for the development of mental health problems in children and adolescents (Wang et al., 2011) as well as the large evidence base for parenting interventions with younger children, little attention has been given to synthesising the evidence for these interventions for the adolescent years. For example, a meta-analysis evaluating the effectiveness of child and adolescent psychotherapies identified only 13 studies in which parent or family focused interventions were evaluated (Weisz et al., 2013). This study aimed to determine if evidence-based psychotherapies (EBPs), produced better outcomes that usual care in youth psychotherapy. EBPs were broadly defined as “any treatment listed in at least 1 of the published reviews systematically identifying EBPs for youths based on the level of empirical support” (Weisz et al., 2013, p. 751) and covering age range of 3–18 years. The theoretical approach and target (e.g., parent or adolescent) for the intervention were not specified. Further, while interventions that included a parenting component were included, this was not a specific focus and outcomes from parent-included interventions were only briefly examined. The proposed study extends and updates this work in several ways: (1) it specifically targets the role of parenting interventions for the adolescent developmental period; and (2) broadens the search strategy to include a larger range of databases and grey literature with an updated search period. A systematic review of available programmes for parents of adolescents and their impact on adolescent mental health outcomes is therefore needed to assess not only availability of evidence-based parenting interventions for this cohort across prevention, early intervention and treatment contexts, but also the efficacy of parenting interventions for this cohort and their potential to contribute to the reduction of mental illness and promotion of skills associated with positive development in adolescents. There are several systematic reviews available that touch on parenting interventions for parents of adolescents. However, these have tended to be of narrow focus, such as targeting adolescent substance abuse (e.g., Allen et al., 2016; Kuntsche & Kuntsche, 2016); externalising mental health disorders (e.g., conduct disorder, antisocial behaviour; McCart et al., 2006; Medlow et al., 2016; Woolfenden et al., 2009) or internalising mental health problems (e.g., anxiety disorders, depression; Das et al., 2016; Gillham et al., 2000), or they are outdated or target a wider age range without a specific focus on adolescence. Others have targeted adolescents with existing diagnoses or clinical symptoms, meaning that the preventative role of parenting programmes is largely unexplored (e.g., Woolfenden et al., 2001). One other review of family-based interventions for child and adolescent mental health disorders by Kaslow et al. (2012) was a narrative rather than a systematic review. In addition, existing reviews have taken a problem focus and tended not to report outcomes related to positive development and the promotion of adolescent skills and competencies. Thus, to fully identify the role and scope of parenting interventions in addressing adolescent mental health problems from prevention to treatment, in-depth and systematic exploration of the evaluation literature is required to understand the effectiveness of parenting interventions on adolescent mental health outcomes, positive development and the parent–adolescent relationship. The primary objective of this review is to answer the following research question: do parenting programmes for parents of adolescents impact adolescent mental health outcomes, positive development and the parent–adolescent relationship? We will achieve this by systematically searching for and synthesising the extant evaluation evidence that meets the inclusion criteria for this review. The secondary objective of this review is answer the research question: does the impact of parenting programmes for parents of adolescents on adolescent mental health outcomes, positive development, and the parent–adolescent relationship vary by: (a) diagnosis; (b) sociodemographic risk status (e.g., family structure, household income); (c) type of caregiver (e.g., biological parent, foster carer, kinship caregiver); (d) geographical location of the study; (e) intervention setting and modality; (f) type of outcome measurement modality (e.g., observation, self-report); (g) type of intervention model (prevention, early intervention, treatment, theoretical framework); (h) participant age (parent and/or adolescent); and/or (i) participant gender (parent and/or adolescent)? Data permitting, we will fulfil this objective through subgroup analyses. This review will include randomised controlled trials (RCTs) where participants have been randomly allocated to an intervention or control condition, and cluster-RCTs where predefined clusters or groups are randomised to different conditions. In these designs, the intervention condition refers to participants who take part in the parenting programme for parents of adolescents. The control group refers to those in a comparison condition involving no intervention, an alternative intervention, service provision or treatment as usual, or waitlist control. In addition, eligible studies may include follow up assessment. We define participation in a study to be comprised of two parts: (1) participation in the intervention and (2) the provision of outcome data. This means that participants can be either parents or adolescents. We envision that some studies might include adolescent data and not parent data, but to be included in the review the study must have one or more parent as a programme participant regardless of whether they provide outcome data. Studies will be included if the participants are parents or caregivers who have adolescents aged between 10 and 18 years at the start of the intervention; accordingly, adolescent participants must be aged between 10 and 18 years. Parents or caregivers will be included if they are biological, adoptive, kinship, or foster caregivers. Adolescents may be the primary recipient of the intervention providing there is a substantive parenting component and that a key objective of the intervention is parenting support and capacity building. Where studies include a proportion of adolescents or parents with adolescents outside of the age range, we will contact the study authors to obtain data for participants in the specified age range. When that data is unavailable, we will include the study if at least 80% of the sample falls within the specific age range. If these data are not available, we will exclude the study. We will include studies in any resulting meta-analyses regardless of whether the sample is comprised entirely of the eligible age group or comprised of at least 80% in the eligible age range. However, we will conduct sensitivity analyses to determine whether the results change by including studies with different proportions of eligible participants. We will exclude parents with an uncontrolled serious mental illness (i.e., major depression, anxiety disorder, substance use disorder, bipolar disorder, psychotic disorder, personality disorder). These parents are arguably a unique category of participants and have been included in other reviews specifically targeted at these populations (Bee et al., 2014; Kersten-Alvarez et al., 2010, 2011; Reupert et al., 2012; Siegenthaler et al., 2012; Thanhäuser et al., 2017). Adolescent participants will be excluded if they have a serious, uncontrolled mental illness or a diagnosis of intellectual disability, global developmental delay or traumatic brain injury. These individuals are likely to differ from neurotypically developing adolescents and the interventions would require a conceptually different and tailored approach that addresses the specific developmental issues and contexts of these individuals (Brown et al., 2013; Tellegen & Sanders, 2013). Studies will be included that evaluate a parenting intervention that primarily targets parenting practices and/or the parent–adolescent relationship. The intervention can take a prevention, early intervention or treatment focus but must have an active psychological focus such as active skills teaching, CBT, acceptance or mindfulness, or emotion-coping. We will include interventions based on any theoretical paradigm or approach, including parenting interventions based on behavioural principles and social learning theory, attachment theory (including Interpersonal therapy), family systems theories, relational frame theory/acceptance and commitment therapy and nonviolent resistance theory. Interventions involving an adolescent component (i.e., where the adolescent is a direct participant in the intervention) will be included only if the parenting component is primary, not supplementary, and is of equal or greater dosage than the adolescent component. Interventions will be excluded if the sole focus or content of the intervention is related to the psychological functioning, social support, relationships or wellbeing of the parent (e.g., psychological interventions for mental health problems; case management or practical support for employment, housing and finances; social support groups, couples counselling). We will, however, include studies that have a parenting component alongside this type of content as long as the parenting aspect is of equal or greater dosage. We will exclude interventions that do not have an active psychological focus, such as those that involve peer support or psychoeducation only. Studies will be included if the outcomes are measured using questionnaires (self or other report), independent observation, clinician ratings, administrative data or diagnostic interview. The timing of outcome assessment will be categorised as: short-term (immediately postintervention to 3 months), medium-term (3–6 months after the intervention) and long-term (7–12 months vs. >12 months). Adolescent mental health, including disruptive, oppositional or conduct behaviour problems; anxiety symptoms; and depressive symptoms; as measured by targeted, problem-specific scales such as the Conners Rating Scales (Conners, 1997), Eyberg Child Behaviour Inventory (Eyberg & Ross, 1978), Child Depression Inventory (Finch et al., 1987), Screen for Child Anxiety Related Disorders (Birmaher et al., 1997), Spence Children's Anxiety Scale (Spence, 1998), and broad-based questionnaires such as the Child Behaviour Checklist (Achenbach & Rescorla, 2001), Strengths and Difficulties Questionnaire (Goodman et al., 2003) and Youth Outcome Questionnaire (Burlingame et al., 2004). Adolescent positive development (the skills and competencies that adolescents require to thrive and successfully transition into adulthood (Bowers et al., 2015; Lerner et al., 2009) including measures that assess social skills, educational achievement and engagement, persistence and dealing with setbacks, forward planning and preparedness, problem-solving and optimism or hope for the future. This includes domain specific scales such as the Student Subjective Wellbeing Questionnaire (Renshaw et al., 2015), Social Skills Improvement System (Gresham & Elliott, 2008) or Children's Hope Scale (Snyder et al., 1997), as well as broad-based measures assessing only positive development such as the Five Cs of Positive Youth Development (Geldhof et al., 2014) and those that include a subscale to measure positive development, such as the Adolescent Functioning Scale (Dittman et al., 2016) and Strength and Difficulties Questionnaire (Goodman et al., 2003). Parent–adolescent relationship functioning or quality, including both positive (e.g., warmth, support, acceptance, connectedness) and negative (e.g., rejection, hostility, conflict) dimensions of the relationship and global measures of relationship quality. Parenting practices, including effective (e.g., encouragement and praise, clear limit setting, appropriate consequences, consistency, appropriate monitoring) and ineffective (e.g., coercive or harsh discipline, physical discipline, neglect) dimensions of parenting and parenting styles. We will include studies with eligible interventions that are delivered in any setting (e.g., home, community, school, hospital, clinic, adolescent mental health services) or via any delivery modality (e.g., individual, group, workshop, seminar, online, telehealth). Included interventions can be of any duration or intensity, and may take a prevention, early intervention or treatment focus. Interventions can also be practitioner-led or self-directed (e.g., online programme, self-help w