Cognitive-behavioral therapy for depression in children and adolescents

It is recognised that depression can severely impair a young person in many important aspects of their life, school, peer and social relationships, and also will frequently persist into adulthood. The term depression is used to describe a cluster of symptoms involving significant changes in mood, in thinking and in activity. These symptoms persist and result in changes in personal and social functioning over a period of at least two weeks. The main features are listed below.

The presentation of a depressive disorder depends on the developmental stage of the young person. The ability to communicate about experiences can vary widely between young people. Cognitive development influences the symptom profile. For example, feelings of guilt, existential thinking, nihilism and morbid introspection are usually only described by older, more mature adolescents. Younger adolescents may show more dependent behaviour with parents than usual.

The effects of depression are wide ranging, and involve changes in the young person’s behaviour, feelings and thoughts. Commonly a vicious circle is created, in which symptoms of depression enhance themselves. For example, inactivity leads to disturbed sleep and to increased time for worrying, both of which increase the symptom of low mood, which in turn leads to further inactivity. Lack of sleep and poor concentration can lead to problems with schoolwork and an increasing sense of failure. Irritability and sensitivity can lead to arguments and difficulties in relationships which escalate and prove to the young person that they are hopeless, worthless and that no-one cares about them.

The two international diagnostic systems, Diagnostic and Statistical Manual of Mental Disorders (DSM-IV; American Psychiatric Association, 1994) and International Classification of Diseases (ICD-10; World Health Organization, 1993) both categorise depression as mild, moderate and severe. ICD-10 puts the emphasis on symptoms whereas DSM-IV considers symptoms and functional impairment.

It appears that depressive disorder in younger children presents differently from in adolescence, with more somatic complaints, irritability and agitation, aggression, disoberdience and boredom and less hopelessness, worthlessness and despair. For youmger children the diagnosis is commonly associated with major family dysfunction (Brent & Malouf, 2009). There is equal prevalence between the genders and less likelihood of continuity into adulthood.

Epidemiological studies indicate that the prevalence of depression increases with age. A prevalence of 1–2 % is reported in school age children (Costello, Mustillo, Erkanli, Keeler &Angold, 2003), in adolescence 3–8 % (Fergusson Horwood, Ridder& Beautrain, 2005) and lifetime prevalence to young adulthood 20 % (Kessler Blazer & McGonagle, 1994). After puberty, depression is about twice as common in girls than boys.

The length of episodes of depression in community samples is reported to range from 3 to 6 months and in referred groups 5–8 months (Birmaher, Arbelaer & Brent, 2002). There is a high risk of recurrence with follow up studies indicating 30–70 % for 1–2 years in clinical samples and 11–33 % for 2–4 years in community samples (Birmaher et al., 2002).

Depression in adolescents is acknowledged to be under-recognised (NICE, 2005). Adults may find it difficult to recognise distress in children. Children do not communicate their mental state in similar ways to adults and so behavioural presentations such as irritability or temper outbursts may be seen as an adolescent behaviour problem rather than mood disorder.

Multiple problems (Comorbidity)

In specialist mental health services depression is rarely seen in isolation. Concurrent symptoms of behaviour problems or anxiety will be present in almost all cases and between 50 and 80 % of depressed young people will also meet criteria for another disorder. About 25 % will have conduct or oppositional defiant disorder and a similar figure for anxiety disorder (Angold & Costello 1995; Goodyer & Cooper, 1993). Other problems may have been present for several years. It may appear that the challenges presented by coping with difficulties in childhood become manifest as low self esteem as the young person enters adolescence and is more acutely aware of their problems. The presence of multiple problems challenges successful intervention.

A typical CBT programme

CBT begins with a comprehensive assessment of current problems and background, drawing on a range of information sources including the young person, their family or carers, other professionals involved, and probably their school. This is likely to include some questionnaire measures. The early stages of CBT involve a psycho-educational presentation, including information about the CBT model and about depression. Engaging the young person into therapy involves working to define the therapeutic contract, identify the problems from the young person’s point of view, and developing a shared formulation of difficulties. Goals are set for the outcomes of therapy. The vocabulary that the young person uses to describe their emotional experiences is considered. Most programmes begin with behavioural approaches;- activity scheduling, a systematic appraisal of the young person’s day-to-day activities and the impact that this has on mood and thinking followed by targeted activation. About 4–5 sessions into therapy, as mood begins to lift, cognitive techniques are used, including identifying negative automatic thoughts, recognising distorted thinking, and cognitive restructuring. The young person is involved in self-monitoring their thoughts, feelings and behaviour and challenging negative thoughts. Social problem solving may be used to work on social difficulties. As treatment draws to a conclusion, symptom monitoring continues. There is a shared discussion of work that has taken place and future coping strategies identified in a ‘therapy blue print’ that aims at relapse prevention.

It is a general characteristic of the therapeutic model in CBT to have a session structure that has a clear agenda, work using a collaborative therapist style, and use home practice between sessions. Joint work with parents and carers and liaison with other agencies are important features. Specific challenges in working with people with depression compared to other disorders include recognising the impact of low mood on therapeutic engagement and on cognitive work. The young person may need to be activated first in order to have potentially positive experiences available. Depressive thinking will itself have a direct impact on cognitive work as well as being a focus for the tasks involved.

The use of both behavioural and cognitive techniques is important; they are complementary. Behavioural techniques serve to mobilise the young person, to increase their experience of normal peer-related activities and to prevent the potentially adverse effects of opting out and aviodance for instance in school. The young person can gain some sense of achievement from behavioural tasks as well as potentially learning new skills. Cognitive work addresses more directly the negative elements of depressive thinking and assists the young person in a more realistic appraisal of their experiences.

Evidence base

To date there have been about 50 randomised control trials evaluating psychological interventions for depression in young people. The majority of these trials are studies from the USA and involve cognitive and/or behavioural interventions. A minority of the studies involve samples of clinic populations. Most studies have participants with high levels of depressive symptoms rather than a formal diagnosis of depressive disorder.

Comparison groups

In most of the studies comparisons are made between treatment and no treatment, for instance waiting list controls. There are, however, some comparisons between treatments. cognitive behaviour therapy (CBT) has been compared to family therapy in one study (Brent et al., 1997); CBT was found to be slightly more effective. CBT was also compared to school counselling (Stark, 1990) and to interpersonal therapy (Rosello & Bernal, 1999); interpersonal therapy was found to be more effective.

Recent studies there have involved comparisons involving CBT and medication. In the largest of these, the Treatment for Adolescents with Depression and Study (TADS, 2004), CBT and medication together were found to be more effective than medication alone; and treatment with medication was superior to both placebo and CBT alone. Longer follow ups of participants have suggested that CBT either alone or with medication reduced suicidality (TADS, 2007). A further major trials (TORDIA; Brent Emslie, Clarke, Wagner, Asarnow, & Keller, 2008) showed CBT and medication to be superior to medication alone. However in the ADAPT trial (Goodyer, Dubicka, Wilkinson, Kelvin, Roberts & Byford, 2007) no difference was found between combination treatment and medication alone. The explanation of these different results remains a matter of active debate (see Delaney, 2009).

Group CBT for depression

This has been extensively evaluated in the USA by Greg Clarke and colleagues (Clarke, Rohde, Lewinsohn, Hops & Seeley, 1999). Their Adolescent Coping with Depression Course has groups of 4 to 8 young people aged 14 to 16 years involved in 2-hour school based sessions. The course is recommended for mild to moderate levels of depression. It uses a workbook that is integrated with course and homework tasks. Parallel parent sessions are available (nine 2-hour sessions). Manuals are available on the web (www.kpch.org/public/acwd/acwd.html).

There have been many studies of CBT, involving various different approaches within the model. Despite this, there has been no study where the efficacy of different elements of CBT has been explored. However, in a meta-analysis (Weisz, McCarty & Valeri, 2006) outcomes were compared for cognitive as opposed to behavioural approaches, and no evidence was found of any difference. Similarly, no difference was found between individual and group CBT.

A further dimension in relation to outcome is the effect of parental involvement in therapy. This has, however, rarely been explored, unlike with other disorders, for instance CBT with anxiety disorders (Bogels & Siqueland, 2006). The only outcome study is a group based treatment with a parallel parent group (Clarke et al, 1999), where no enhancement of impact was demonstrated.

There have been studies delivering CBT via bibliotherapy (Ackerson , Scogin, McKendree-Smith & Lyman, 1998) and computerised interventions (Abeles, Verduyn, Robinson, Smith, Yule & Proudfoot., 2009).

Meta-analyses

The rapidly growing number of outcome studies for psychological interventions for depression in young people has permitted four meta-analyses since 1998 (Michael & Crowley, 2002; Reinecke Ryan & Dubois, 1998; Reinecke et al., 1998; Watanabe , Hunot, Omori, Churchill & Furukawa, 2007; Weisz et al, 2006 Weisz et al., 2006). Weisz and colleagues pooled data from 35 studies. Their approach included a wider range of studies than earlier meta-analyses, including for instance data from unpublished theses. As a result, they had a larger sample size, which may be more representative of research conducted and also used a different calculation method designed to prevent over-estimation of treatment effect size. Their conclusions were that CBT shows significant but modest treatment effects. They report an effect size of 0.34, which ‘lags behind’ that found for treatment of other disorders, notably anxiety. They reported that cognitive and behavioural treatments are equally effective. If there is no evidence that working cognitively enhances outcome, this finding has important implications for the cognitive theory of depression. However, cognitive ‘treatments’ also have a large behavioural component. It was also reported that treatment gains in the long-term were not demonstrated. There is a consistent finding that over time untreated populations improve. Again this is a problem for cognitive theories, as an anticipated benefit from CBT is in changing core beliefs, which should reduce relapse. A further conclusion was that CBT is most effective in mild to moderate depression.

Problems with research studies

There are major theoretical difficulties in appraising the evidence for effectiveness of psychological interventions for depression in young people. These are not necessarily addressed by aggregating the data in meta-analysis. The variation between randomised control trials is considerable in the nature of samples, length of treatment, and length of follow-up. The follow-up after treatment is often very short (3–6 months) which, given the nature of depression, inevitably will make long-term outcome dubious. This is a major issue in appraising the effectiveness literature. There are different forms of CBT in use; these include cognitive and behavioural, social skills training, and problem-solving approaches. There is also wide variation in the number of sessions, from 5 to 15. It seems that different forms of CBT are differentially effective but there is no indication from the literature about how this might come about. There are wide ranging comparison groups, although mostly wait list control. There are no replication studies, particularly important where a treatment has been shown to be effective. There is a lack of systematic assessment of co-morbidity or of other moderating and mediating variables for outcome. There has been neglect of treatment processes important in adolescence, for instance, of the engagement dimension, and of treatment fidelity.

Clinical guidance

The most recent systematic review in the UK of treatments for depression has been conducted by the National Institute for Clinical Excellence (NICE, 2005). This includes robust examination of research but also takes into account other forms of evidence where no randomised control trials are available. The Institute makes recommendations and sets clinical standards. The NICE evaluation of treatments for depression in young people concluded that CBT and other individual therapies are likely to reduce the length of a depressive episode (a reflection that controls improve over time). Two studies were considered where CBT showed no difference in outcome from waiting list or treatment as usual and three studies where CBT performed better than a control comparison, although the advantage was lost by follow-up. For group therapy, CBT showed consistently better outcomes than wait list or no treatment control. Mixed results were described when comparing CBT to other treatments.

NICE defines the UK national standards and the recommendations were, in summary, that it is important to train front line professionals working with young people in detection and awareness of depression, and that within CAMHS we need to ensure that clinicians involved in assessment adequately consider the presence of depression. For treatment of moderate to severe depression in specialist CAMHS, psychological therapy should be first line. Psychological therapy should be conducted for at least a 3-month trial. Therapies available should be CBT, IPT (interpersonal psychotherapy; Mufson , Dorta, Weickramaratne, Nomura, Olfson & Weissman, 1999) or short term family therapy (Brent , Holder, Kolko, Birmaher, Baugher & Roth, 1997). Antidepressants may be considered, but should be closely monitored and used in combination with psychological treatments.

Areas for development

Treating depressed young people successfully is difficult. The evidence base gives us little reassurance that we are effective. Clinical populations are often harder to reach than research populations and working with multiple problems is the norm. There is higher prevalence of high risk behaviour, particularly suicidality, and persisting adversity in clinical as opposed to research populations. Patterns of relapse and recovery in depression in adolescence can make engagement and sustaining therapy difficult.

In clinical work the challenges include engaging depressed young people into therapy. There can be problems in ensuring attendance at sessions and for young people in being optimistic about help; this takes time. In addition, there can be difficulty in maintaining the pace of therapy, particularly if there are attendance issues. Sometimes drop out can relate to rapid improvement but then there can be a subsequent deterioration following a negative experience. There are challenges in working cognitively with some young people who find it difficult to access their thoughts. The nature of young people’s social contexts needs to be fully considered. When considering comparison with adult populations, NICE 2005 states:

Whereas adults with depression are often treated for the disorder specifically, children with depression are often not thought of as ‘having’ depression but as affected by a set of emotional, behavioural, learning, relationship and family problems that need to be considered together and may still need to be addressed together even if depression in the child is the primary concern. (p.75)

Enhancing the impact of CBT

In clinical practice, CBT for depression has tended to focus more on one-to-one work than CBT for other childhood disorders, probably because of the personal nature of the experience of depression. CBT therapists need also to consider addressing systemic change and context in meeting young people’s therapeutic needs. There are several useful approaches to successful engagement. These include providing age-appropriate information and working from the outset with a clear focus on the young person’s experience. Motivational interviewing (see Schmidt, 2005), developed from work with substance abusing adults, can give a framework for supporting the start of the change process. It is crucial that interventions address the problems defined by a formulation. Work needs to be supported by appropriate supervision. The evidence suggests that both cognitive and behavioural elements are effective. There tends to be an emphasis from CBT practitioners on purely cognitive approaches. CBT can be delivered not only one-to-one and face-to-face but with supplementary systems such as texting, e-mail and telephone. CAMHS professionals bring skills in creative work with young people, particularly in non-verbal and action techniques and these can be implemented within the CBT model. There needs to be flexibility in treatment duration, including creating an expectation of sustained intervention from the outset. Booster sessions can be used, with longer follow-up, possibly in groups (see Clarke et al., 1999).

In clinical practice techniques have developed for working with multiple problems. They include having a goal focus and identifying the difficulties that are most distressing to the young person. Problems that are easiest to change are addressed first in order to enhance motivation. There is a negotiation of goals on behalf of the young person with parents and others. A structured approach can help the young person to cope with seemingly overwhelming problems in a range of different areas.

Young people with depression will be at high risk of suicidality and self harming behaviour. Even though effective engagement in CBT reduces this risk, it is important that a non-judgemental and supportive approach is taken by the clinician and that appropriate goals are set by the young person in managing these problems. Self harming behaviours commonly encountered include self cutting, burning, hitting, excessive use of drugs and alcohol, risk and promiscuity. Chaotic eating patterns including bingeing and vomiting are also common. Suicidal thinking and urges needs to be identified and incorporated into a risk management structure.

If the initial presentation to mental health services has been an act of self harm, following assessment it will be important to determine key problems as the young person sees them. Risk assessments will continue to play a part in sessions and it will be important to agree reviews of treatment and progress and contingencies for deterioration in the clinical situation.

Self harm or suicidal behaviour can be included in a formulation of the young person’s difficulties. For many young people self harm is seen as a coping mechanism and they can be interested and engaged in therapy on the basis of developing alternative strategies for dealing with the feelings that lead to self harm. Suicidal behaviour clearly needs to be managed within a safe environment with close working with parents or carers. The formulation will include other depressive symptoms and an analysis of the underlying beliefs and assumptions and negative automatic thoughts and styles of thinking that maintain them.

Strategies used commonly in CBT in the management of suicidality and deliberate self harm include distraction from thoughts of self harm and developing alternative ways of managing overwhelming emotion. Suicidal content may emerge in challenging negative automatic thoughts, cognitive distortions and core beliefs. Work with parents may include techniques for «grounding», that is, parents being aware of and responding to a need for greater supervision. The clinician may be able to make a ‘no self-harm’ contract with the young person.

There have been some recent attempts to present clinical thinking on working with CBT and families (Dummett, 2006; Wolpert, Doe, & Elsworth, 2005). These approaches go beyond the ‘parallel parent session; psychoeducation’ or ‘advice to parent about how to support change’ descriptions of parental involvement. Models could be developed that address working with the parent and young person together as essential, and directly integrated with the work with the young person, focusing particularly, for example, on the young person’s core beliefs. Much systemically driven work is compatible with CBT, particularly some elements of solution focused approaches.

It is important to maintain a balance of work with families and individual work. Issues of confidentiality within the individual work will continue to be important. Differing goals between young people and their families may need to be considered. It is likely that other members of the family will have experienced symptoms of depression.

A formulation of depression based on Beck’s theory (Beck, Rush, Shaw & Emery, 1979) will include a developmental aspect suggesting that schema developed in childhood, in all probability in relation to family events or relationships. In working with families it may be possible to address these issues or elicit further information that informs the nature of early challenges. The active parental role in therapy, including the evaluation of progress, should not be under-estimated. Placing individual therapy within a conceptualisation that includes family work has implications for services and in some cases more than one therapist may be deemed appropriate.

Case examples

The following case vignettes are provided in order to exemplify the diverse presentations of depression in young people referred to services and to demonstrate a range of ways in which CBT can be applied.

Britta age 9

Britta was described as always having been a bright sensitive girl, generally a competent coper in the face of new experiences. At the initial assessment key features were anxiety and depression. She was only able to sleep with the light on in her room, she was over-eating and had gained weight. Her mood was irritable and tearful although she was not usually badly behaved. Britta’s parents had separated 2 years prior to referral. Her father had a history of alcohol abuse but contact with him had been maintained. Britta and her mother had moved house and she had had to change school. Subsequently a friend had moved out of the country, the family dog had died and Britta experienced bullying in her new school.

As is usual with children of this age the therapeutic intervention included some individual work with Britta in identifying goals, thoughts and feelings. This was very important particularly as in the course of an individual session with Britta it came to light that she had many negative thoughts resulting from an incident when alone overnight in her father’s care she was aware that he had been too inebriated to look after her. She had been afraid at the time but was also fearful that her contact with him would be stopped. The majority of the following sessions involved work with mother and child together to help Britta improve communication about her feelings, to increase shared positive activities and then boost self esteem by encouraging age appropriate independence in activities and routines. The matter of contact was managed after discussion with Britta. Her mother was able to intervene effectively to reassure Britta that contact would continue safely, with the help of paternal grandmother. Some individual work also was required with Britta’s mother and also telephone communication with her father.

Adam age 14

Adam was referred with a 6 month history of depression following the death of his paternal grandfather. His father was also depressed. Adam found it difficult communicating his negative thoughts but the therapist was able to use Adam’s high level of interest in football in describing how thoughts affect performance. Adam was able to talk about feeling that he had always to succeed and that he should not admit to difficulties. Following this, work with Adam and his father together on core beliefs helped Adam to understand that his father had felt similarly in relation to his own father. His father emphasised the importance that he placed on understanding how Adam felt and that he valued being able to help Adam if he could so needed to know of any problems.

Erica age 15

Erica had experienced sexual abuse when age 9 from an adolescent male babysitter. A sexual assault in adolescence led to the onset of depression. Since starting High School Erica had experienced learning and peer problems in school which were exacerbated by recent school absences. There were stresses within the family as a result of her father’s unemployment. Her mother worked full time. She had an older sister and a younger brother both of whom were outgoing and successful at school. Erica had symptoms of low mood but also frequent headaches and abdominal pain. Individual CBT involved activity scheduling and a typical programme of cognitive work. Erica had no symptoms of trauma but her unwanted sexual experiences had directly affected her self esteem and sense of self worth. Work with Erica and her family focused on reducing conflict and supporting a successful school intervention in obtaining additional help with difficult lessons.

Conclusion

In recent years there has been a significant increase in research activity in the area of depression in children and young people. The focus of research has moved away from establishing the nature of depression in young people to the task of delivering effective treatments. The earliest intervention studies generally involved comparing CBT to no intervention in mild to moderately depressed young people, usually not those accessing clinical services. More recently the emphasis has been on moderately to severely clinically depressed young people and includes use of medication as well as psychological intervention. This has led to challenges to increase efficacy for the most troubled populations.

Thus the current state of research evidence would suggest that CBT for depression has modest effects. There is a need for the further developments taking place in clinical practice to be reflected in research. Salkovskis (2002), quoted in the NICE Guideline of Depression in Adults (2004), stated; ‘in most instances CBT for any particular psychological problem is quite different now to CBT as practised 5 to 10 years ago. Practice is evolutionary and interactive, and pragmatic outcome trials play a relatively minor part in this development.’ Elements of practice that are developmentally appropriate for adolescents, including family work, are an example where child and adolescent therapists are developing practice. This needs integrating both theoretically and clinically in order to be able to be evaluated in research trials. This may include wider consideration of models for delivery of CBT including group work and computer and new technology applications. l