Special Report: Children's Mental Health - Research evidence

Study 1

Review: Effectiveness of mindfulness in improving mental health symptoms of children and adolescents: a meta-analysis

Kannan Kallapiran, Siew Koo, Richard Kirubakaran and Karen Hancock, Child and Adolescent Mental Health, (2015)

There has been an upsurge in the use of mindfulness-based interventions (MBIs) in treating adolescent mental health issues. This systematic review considers the impact of different MBIs when managing a variety of mental health disorders in children and adolescents.

There is a consensus that mindfulness refers to the development of a sustained attention to the present but there is no conclusive definition in use. Here, the authors conceptualise mindfulness as developing an attitude of non-judgment and acceptance of the present. While MBIs developed for use with different groups practice mindfulness in a variety of ways, the majority are brief, group-based interventions utilising meditation techniques.

Popular MBIs include mindfulness-based stress reduction (MBSR), mindfulness-based cognitive therapy (MBCT), dialectical behaviour therapy (DBT), and acceptance commitment therapy (ACT). Previous reviews have suggested that techniques might be successfully adapted for use with children and adolescents, though the studies reviewed were deemed to be of poor quality.

Kallapiran et al included 11 randomised control trials (RCTs) in this review conducted in both clinical and non-clinical settings. A total of 1,454 participants were included. The quality of the interventions reviewed was assessed as mostly good (assessed on factors including structure of the intervention, experience of therapist and sample size). This reflects the fact that the methodological rigour of the studies evaluating the impact of MBIs is improving. Key findings for some specific interventions were:

• MBSR/MBCT were more effective than non-active treatment in improving stress (the studies compared results for a group with similar issues not receiving any treatment)
• ACT was comparable to active treatment control in managing anxiety (the studies compared results for a group with similar issues receiving different treatment)
• Other MBIs were significantly more effective than non-active treatments in improving stress symptoms.

Most of these studies found positive outcomes for MBIs for stress, anxiety, depressive symptoms, and quality of life. ACT and MBSR appear to meet the standard definition of empirically supported therapies - demonstrating significant results in at least two RCTs and two good quality independent settings. Participants in these groups demonstrated sustained or improved outcomes at the three and six-month follow-up points. ACT was widely used in clinical and MBSR in non-clinical settings in these studies, so the issue of differing responses to techniques in different settings is worthy of further exploration.

While the authors found MBIs were comparable to active interventions and more successful than non-active controls in most of the studies, there is insufficient evidence to conclude that the effectiveness is mediated by the mindfulness interventions or how the different components of specific mindfulness interventions mediate effect. Further investigation of the processes of change is necessary to draw such conclusions.

Implications for practice
The authors conclude that mindfulness-based interventions appear useful in improving stress, anxiety, and depressive symptoms and quality of life in children and adolescents in both clinical and non-clinical samples. MBSR and ACT satisfy the criteria for empirically supported therapy in this population.

However, large-scale empirical research is required to further improve our understanding on the impact of mindfulness interventions on young people and assessing mindfulness as a mechanism of change.

Study 2

Alternatives to inpatient care for children and adolescents with mental health disorders

Ka Ho Robin Kwok, Sze Ngar Vanessa Yuan and Dennis Ougrin, Child and Adolescent Mental Health, (2015)

In recent years, there has been a shift of focus from hospitalisation to community-based care for children and adolescents with severe mental health problems. While hospitalisation may be of benefit, 2011 research found a lack of adequate aftercare is associated with significant risk of self-harm and suicide.

This study provides a systematic review of randomised controlled trials (RCTs) which can help answer the question: Is inpatient psychiatric hospitalisation of young people with severe psychiatric disorders more clinically effective than intensive and comprehensive community-based services?

A number of databases were searched and relevant RCTs involving children aged 18 and under were selected for analysis. Six RCTs involving 569 children were identified. The RCTs explored specialist outpatient treatment, multi-systemic therapy (MST), day patient treatment, intensive home treatment and supported discharge services versus inpatient care. A number of key findings were evident.

Clinical outcomes

• Significantly, studies tended to show similar or better clinical outcomes in intensive community services compared to inpatient care in a number of contexts:
• Similar rates of improvement between specialist outpatient treatment, inpatient treatment and general child and adolescent mental health services (CAMHS) in the UK at one and two-year follow-up for teenagers with anorexia.
• Treatment at home was better at improving all behavioural measures in patients with behavioural disorders.
• MST in young people with psychiatric emergencies was at least as effective, and sometimes more effective, than hospitalisation.
• MST participants, externalising symptoms and family functioning improved and they experienced significantly reduced rates of attempted suicide at one-year follow-up.

Hospitalisation and cost
Using intensive community services was associated with fewer days in hospital during the follow-up period, and lower costs. In particular, MST was more cost-effective at reducing externalising and internalising behaviours.

Family satisfaction
Parents and young people in specialist outpatient treatment and MST tended to report higher satisfaction than those in inpatient treatment. However, there was no significant difference in parental satisfaction between home treatment and inpatient treatment.

Factors behind findings
The authors suggest possible explanations for these results might include the fact that alternative treatments:

• Avoid lengthy admissions, which can contribute to interpersonal distrust and low self-esteem;
• Facilitate autonomy and self-confidence outside of hospital, elements crucial to the development of all adolescents;
• Take place in less restrictive settings, increasing child and family engagement and therefore treatment effectiveness, which could lead to improved outcomes for the child; and
• Working in family environments helps professionals to understand problems experienced in daily life and so better target their work.

Several characteristics of intensive community services were found to be important:

• A comprehensive first assessment
• Tailored treatment plans
• Services delivered at home or in the community
• When inpatient treatment was required, patients were discharged early and received later treatment at home or in the community
• The crucial role of involving the family
• The need for professionals to respond quickly and work flexibly with children and families

The findings of the RCTs echo other studies exploring the effectiveness of alternative models for inpatient treatment for "delinquent" young people. For example, studies looking at MST versus criminal justice or community service for juvenile delinquents and offenders have demonstrated that MST is more effective in reducing delinquent activity, aggressive behaviour, recidivism and drug use, while also improving family cohesion over a four-year follow-up period. These findings add weight to this review because up to two-thirds of male and nearly three-quarters of female young offenders in detention suffer from a mental health problem.

Implications for practice
While the evidence here mostly excludes children and adolescents with the most complex mental health disorders and is largely derived from studies outside of the UK, the authors conclude that intensive community services appear to be a promising alternative to inpatient care in children and adolescents with severe mental health problems.

Study 3

Measurement issues: Review of four patient reported outcome measures: SDQ, RCADS, C/ORS and GBO - their strengths and limitations for clinical use and service evaluation

Miranda Wolpert, Helen Cheng and Jessica Deighton, Child and Adolescent Mental Health, (2015)

There is a drive for routine use of patient-reported outcome measures across child and adolescent mental health services (CAMHS). While the validity and reliability of common wellbeing and mental health measures have been reviewed, there have been fewer attempts to consider the suitability of measures for routine use.

This review considers four child self-report measures: the Strengths and Difficulties Questionnaire (SDQ), the Revised Child Anxiety and Depression Scale (RCADS), (Child) Outcomes Rating Scale (C/ORS) and Goals Based Outcomes (GBOs). The authors consider the strengths and limitations of each and how they can be used to support both clinical practice and service evaluation.

Of the measures reviewed, the SDQ was found to be of most use in service evaluation, while GBOs, C/ORS and RCADS are most useful for informing direct work. However, each measure reviewed had different strengths and limitations and it is important that these are kept in mind when using them.

The Strengths and Difficulties Questionnaire (SDQ)
The SDQ is a well validated measure with an extensive literature detailing its psychometric properties. In terms of clinical utility: while the SDQ is widely used in CAMHS, its usefulness as a feedback measure is unclear. Clinicians say that the broad nature of the measure can mean that feedback does not provide enough detail about case-specific issues.

In terms of use for service evaluation, SDQ has the best normative data (data obtained from using the measure with a representative sample of the wider population. This provides a baseline for scoring against which other scores can be compared) so is most suitable for interpreting results. While the parent/carer SDQ has been used to develop an "Added Value Score" designed to assess improvement in mental health scores following an intervention, no equivalent measure has been developed for the online self-report SDQ for young people.

The revised Child Anxiety and Depression Scale (RCADS)
Evidence supports this questionnaire's value in predicting anxiety and depression. Furthermore, clinicians report that it can help with discussions around particular problems; understanding children's difficulties; informing problem analysis, and monitoring between sessions. The RCADS, ability to help inform diagnoses, track clinical change and distinguish between these two disorders is useful for both clinical and evaluation purposes.

(Child) Outcomes Ratings Scales (C/ORS)
The C/ORS is a measure of psychological distress developed in 2003 by Duncan; practitioners are encouraged to use it to frame a narrative around how children experience services. The scales are popular, endorsing their clinical utility. However, there are concerns about the broad nature of individual items - for example one covers both work and friendships, which service users might feel very differently about.

The few available studies of C/ORS suggest the measures have good internal consistency, reliability and validity; however, some studies note "ceiling effects", when the highest score is not sufficient to accurately assess someone's improving levels of outcome over time. Other studies note that users are sometimes reluctant to use the extremes of the 0-10 response scale. Thus, it may be possible to use the C/ORS for evaluation but these effects should be considered.

Goals based outcomes (GBO)
GBO attempts to measure progress towards goals identified by a young person or their family. Progress ratings for each goal allow assessment of how well treatment is addressing identified issues.

GBOs have been criticised for untested assumptions about how to scale and compare issues, but there is anecdotal evidence of positive use by Law and Wolpert in 2014. GBOs may remove the sense of just completing a "tick box" exercise, about which both practitioners and service users express concerns and the facility to jointly agree goals means the measure is tailored to the work carried out.

GBOs are entirely focused on clinical utility as opposed to evaluation. There is some evidence of use for overall measurement of outcomes, and for tracking progress across sessions to improve services, but more research is needed and the potential for manipulation and data error should be noted.

Implications for practice
By way of conclusion, the authors state these different measures can be viewed as complementary tools. They also point out that determining the best way to make use of them on their own and in combination is the current focus of trials across England.

Study 4

Child maltreatment and adolescent mental health problems in a large birth cohort

Ryan Mills, James Scott, Rosa Alati, Michael O,Callaghan, Jake M Najman, Lane Strathearn,
Child Abuse and Neglect, (2013)

The authors examine whether child maltreatment is associated with adverse psychological outcomes in adolescence, and whether differing patterns of outcomes are seen depending on the type of maltreatment.

Child maltreatment has been repeatedly linked to adverse mental and physical health consequences, and there is some evidence of strong associations between certain maltreatment types and outcomes; for example, studies by Grogan-Kaylor et al, and Manly et al, have suggested that physical abuse may have the strongest association with later delinquency and aggression. It is also well recognised that multiple maltreatment types often occur within the same family and studies have indicated that young people suffering this are more likely to experience psychological dysfunction than peers suffering one form of maltreatment.

This study constituted 7,223 mother and child pairs enrolled in a birth cohort study (a series of surveys with people from birth throughout their lives) in Brisbane, Australia. Suspected child maltreatment was gauged using agency data. When aged approximately 14, young people completed the Youth Self Report (YSR), a screening tool for behavioural and emotional problems in young people. Outcomes measured were internalising (withdrawal, anxiety/depression, somatic symptoms) and externalising behaviours (delinquency and aggression). This method is distinctive because socio-demographic factors were accounted for including the young person's age, gender and race; maternal age, marital status, educational achievement, and family income. It also avoided bias associated with the retrospective recall of child maltreatment.

The YSR was completed by 72 per cent of the cohort. Mills and colleagues found that maltreatment was significantly associated with both internalising and externalising behaviours at age 14. A major finding was the strong association of multi-type maltreatment with adverse outcomes, which was significant for the groups notified for neglect and emotional abuse (with or without physical abuse). Physical abuse, neglect, and emotional abuse were also significantly associated with adverse outcomes on their own.

The following combinations of types of maltreatment saw significantly higher internalising behaviour:

• emotional abuse (with or without neglect)
• multi-type maltreatment including physical (but not sexual) abuse with neglect and/or emotional abuse.

The following groups were associated with externalising behaviours:

• emotional abuse (with or without neglect)
• multi-type maltreatment including physical abuse (with neglect and/or emotional abuse)
• sexual abuse (with neglect and/or emotional abuse, and/or physical abuse).

Findings highlight the significance of emotional abuse and neglect as factors in a significant proportion of child maltreatment-related psychopathology in adolescents, supporting existing evidence. This adds weight to the idea that the variation in observed outcomes of childhood physical and sexual abuse across different studies is likely to be due not only to socioeconomic factors but other family issues including neglect and emotional abuse.

Implications for practice
The authors highlight that agency data almost certainly underestimates the number of young people who experienced each type of maltreatment, and also the rate of multi-type maltreatment. The relative lack of association between sexual abuse (on its own) and adverse psychological outcome in adolescence in this study is acknowledged, and reasons explored as to why this may be the case.

This study underlines that child neglect and emotional abuse have serious adverse effects on adolescent mental health and warrant the attention given to other forms of child maltreatment. Additionally, it confirms that young people who are notified for more than one type of maltreatment are at particular risk of adolescent mental health problems.

Study 5

Getting behind the closed door of care leavers: understanding the role of emotional support for young people leaving care

Natasha Adley and Victoria Jupp Kina, Child & Family Social Work, (2014)

In this paper, a small number of care leavers reflect on their experiences of the transition to independence. The importance of their emotional support networks during transition and the professional support they received to explore these are identified.

Adley and Kina highlight substantial changes to the legislation and guidance surrounding social work with young people leaving care over recent years and report that the transition to independence is becoming increasingly challenging. Evidence shows that care leavers continue to start living independently at a much earlier age and achieve poorer outcomes than their peers. Despite evidence of the intense emotional impact of leaving care and the central role of stable and caring relationships in mediating this impact (Centre for Social Justice, 2014), research has consistently highlighted that current practice pays insufficient attention to sources of emotional support for care leavers.

This study drew on in-depth interviews with six care leavers aged 18 to 21 in an urban local authority in England, including use of a visual tool to promote reflection and guide the interview process.

Five of the six young people rated an emotional support network 10 out of 10 in terms of its importance, yet participants often had only a couple of people in their closest circle. None of the interviewees felt that any professional had worked with them to explore their emotional support network.

Two themes emerged as central factors in relation to participants, support networks:

Misperceptions of the transition process
Most participants said that at the time of leaving care they had felt ready to move on, but came to realise how unprepared they were. One said: "I had this glamorised thought of living on my own and having all my friends around and life being a big party but it wasn't that. I was studying too and... I had to come home and cook and do the washing and the laundry... I was basically on my own."

This indicates to professionals that finding the right moment to offer support is vital. A further issue highlighted was adapting to being alone, which participants described as often being forgotten.

The complexity of accepting professional support
A young person's refusal of support did not always mean they did not want it, but was associated with broader issues of timing, identity, lack of trust, sense of pride or their experiences of the care system in general. One young person said: "Some people feel shame to say they need help, it's easier to say no. I felt ashamed ‘cos I'm not good at budgeting."

This highlights the importance of professionals offering support in a timely manner and exploring the past history of the young people they work with to understand how this might be affecting their present engagement.

Implications for practice
Care leavers said emotional support from professionals should be individually tailored, emerging from relationships of trust that are essential in strong practice with looked-after young people. The onus must be on the professional to overcome resistance to support, to recognise that "no" may not always mean "no", and to ensure that offers of support are repeated.

This paper provides valuable insight into the emotional support networks of young people from the perspective of care leavers themselves.

It provides further evidence of need for social workers to consider, understand and contribute to care leavers, emotional support networks. The use of simple visual tools can help to highlight gaps in young people's support networks.

Links between maltreatment and adolescent mental health problems

• Notified maltreatment was significantly associated with both internalising behaviour and externalising behaviour at 14 years old
• Physical abuse, neglect, and emotional abuse were each significantly associated with both internalising and externalising behaviour
• The following groups had significantly higher internalising behaviour (withdrawal, anxiety, depression) after adjustment: emotional abuse (with or without neglect), and multi-type maltreatment including physical (but not sexual) abuse with neglect and/or emotional abuse
• The following groups were associated with externalising behaviour (delinquency and aggression): emotional abuse (with or without neglect), and multi-type maltreatment including physical abuse (with neglect and/or emotional abuse), or sexual abuse (with neglect and/or emotional abuse, and/or physical abuse)

Source: Findings from a youth self report survey completed by 5,172 young people and undertaken as part of the Child Maltreatment and Adolescent Mental Health Problems in a Large Birth Cohort, 2013

Further reading

Research summaries on The Mental Elf:

• How does parenting impact on the emotional wellbeing of children?
• IPT and CBT best for depression in children and young people
• Social media, cyberbullying and young people: the evidence so far

The impact of early childhood experiences on adolescent suicide and accidental death by NSPCC Northern Ireland

Measuring wellbeing: A Literature review by University of Bristol and Coram Voice

A systematic review of mental health outcome measures for children and young people by CAMHS Evidence Based Practice Unit, University College London and the Anna Freud Centre

The association between participation in cultural activities and self-perceived health, life satisfaction and mental health: the Young HUNT study, Norway

A US study into the impact of cardiorespiratory fitness in protecting against depression in adolescents

An evaluation of the ‘FRIENDS for Life, emotional resiliency programme on anxiety levels, anxiety type and school adjustment

Resources by Research in Practice

Adolescent mental health, Frontline briefing, tool and workshop

Risk-taking adolescents and child protection Frontline briefing

That Difficult Age: Developing a more effective response to adolescence Evidence scope and 1-2 day tailored support workshop

Models of adolescent care provision Evidence scope

Promoting resilience in children, young people and families Frontline briefing and tool

Research and Policy Update 166 January 2015 on adolescent mental health

Why it is important to understand adolescent development Blog by John Coleman, My Frontline blog

My Life, My Future: supporting young people's emotional resilience Blog by Natasha Rego, project lead at the Mental Health Foundation

This article is part of CYP Now's special report on children's Mental Health. Click here for more