How common are mental health problems in adolescence?

The onset of most adult mental disorders is between 15 and 25 years (Kessler et al. Reference Kessler, Berglund, Demler, Jin, Merikangas and Walters2005), with rates of child psychopathology in the United States ranging from 17% to 26% (McCabe et al. Reference McCabe, Yeh, Hough, Landsverk, Hurlburt, Culver and Reynolds1999; US Public Health Service, 2000). Many psychological disorders of adolescence are transient and self-limiting (Patton et al. Reference Patton, Ross, Santelli, Sawyer, Viner and Kleinert2014b) but some continue into adulthood, suggesting that adult mental disorders are really ‘chronic disorders of young people’. A large recent study conducted in Australia on 1943 teenagers assessed at baseline, at five 6-monthly intervals, and at three follow-up time points (20–21, 24–25, 28–29 years old) demonstrated that the strongest predictor of difficulties as an adult was having a long duration and multiple episodes of mental health disorder in adolescence (Patton et al. Reference Patton, Coffey, Romaniuk, Mackinnon, Carlin, Degenhardt, Olsson and Moran2014a). Although duration of illness was the strongest predictor, underlying anxiety and low mood was identified during at least one point in their teens for 30% of boys and 50% of girls. Evidence suggests that problems can exacerbate in children from low-income families, where there is risk of exposure to poverty, community violence, higher rates of social adversity, in addition to insufficient housing, health and poor contact with mental health services (Leventhal & Brooks-Gunn, Reference Leventhal and Brooks-Gunn2000; Self-Brown et al. Reference Self-Brown, LeBlanc, Kelley, Hanson, Laslie and Wingate2006). A review of 52 studies reported a median prevalence rate of psychopathology of 15% among adolescents, and an overall prevalence estimate ranging from 1% to nearly 51% (Roberts et al. Reference Roberts, Attkisson and Rosenblatt1998). A more recent UK survey found that 10% of 5–16-year-olds have a mental health disorder (Green et al. Reference Green, McGinnity, Meltzer, Ford and Goodman2005) with other studies reporting comparable results: 11% in 11–15-year age group (Office of National Statistics, 2004), and 20% in 16–24-year-olds (Budd et al. Reference Budd, Sharp, Weir, Wilson and Owen2005). Attempts at suicide are made by 2–4% of adolescents, with a completed suicide rate of 7.6 per 10 000 in the 15–19-year age group. In addition, 2–8% of young people experience major depression (Rushton et al. Reference Rushton, Forcier and Schecktman2002); about 2% have obsessive compulsive disorder; 0.5–1% of 12–19-year-olds (predominantly females) have anorexia nervosa and a further 1% bulimia nervosa (Wade et al. 2011). During the adolescence period, more serious disorders emerge with age (Petersen and Leffert, 1995). Four per cent of all cases of schizophrenia experience an onset in adolescence (Vyas et al. Reference Vyas, Kumra and Puri2010a, Reference Vyas, Patel, Nijran, Al-Nahhas and Puri2010b), which is reported to be associated with developmental deviance, and poor functional outcome (Vyas et al. 2007; Vyas et al. 2011). Evidence suggests that at least one in four to five young people will suffer from at least one mental disorder in any given year (Patel et al. Reference Patel, Flisher, Hetrick and McGorry2007) and when taken together, approaching 40% of youth aged 11–25 years will experience a period of major mental ill-health (Copeland et al. Reference Copeland, Shanahan, Costello and Angold2011).

Not only are mental health problems common in adolescence, these are also complex, with comorbidity particularly common. In community samples, 20% of those with an impairing psychiatric disorder have more than one disorder; comorbidity among those attending CAMHS is even higher (Ford et al. Reference Ford, Goodman and Meltzer2003, Reference Ford, Hamilton, Dosani, Burke and Goodman2007). The use and abuse of alcohol and drugs is high – in the United Kingdom 29% of 13-year-olds report drinking alcohol once a week; 16% of 16-year-olds regularly use solvents or illegal drugs; while 17% of older teenagers use cannabis (Fonagy et al. Reference Fonagy, Target and Gergely2000). For young people receiving CAMHS care, the rate of substance abuse or dependence increases dramatically, affecting nearly half of 21–25-year-olds (Greenbaum et al. Reference Greenbaum, Prange, Friedman and Silver1991). The Breaking the Cycle report (Social Exclusion Unit, 2004) found that 98% of young adults (16–25-year-olds) accessing services in the United Kingdom had more than one problem or need. Other common comorbid problems included homelessness, problems associated with leaving care, lack of training/education opportunities, barriers to employment, crime, poor housing, drug and alcohol misuse and learning disability.

Do adolescent mental health problems persist into adulthood?

Mental health problems in adolescence also predict who is likely to develop severe problems in adulthood (Silva, Reference Silva1990; Lamb et al. Reference Lamb, Hall, Kelvin and Van Beinum2008; Patton et al. Reference Patton, Coffey, Romaniuk, Mackinnon, Carlin, Degenhardt, Olsson and Moran2014a). The National Comorbidity Survey Replication in the United States found that 75% of people with a mental disorder had an age of onset younger than 24 years (Kessler et al. Reference Kessler, Berglund, Demler, Jin, Merikangas and Walters2005). Long-term outcome data from the 1946 National Birth Cohort survey shows that about 60% of 13–15 years old with an internalising mental disorder had persistent mental health problems in adulthood (Colman et al. Reference Colman, Wadsworth, Croudace and Jones2007). Teenage depression, which affects 3% of the adolescent population, is strongly predictive of adult depression disorder, substance misuse, unemployment, poor academic achievement and suicide. Although the prevalence of depression may vary across population depending on symptom and degree of severity measurements, the Youth Risk Behaviour Surveillance System study on high school students reported that 36.7% of females and 20.4% of males showed high levels of sadness, where Hispanic students reported higher rates compared with non-Hispanic peers (Youth Risk Behaviour Surveillance System, 2007). However, few seek help and fewer still (one in four) receive treatment. Only about half of depressed young people contact specialist services or a professional for depression (Kessler & Walters, Reference Kessler and Walters1998). There is continuity of depression from adolescence with adulthood, with 70% of those suffering depression in late adolescence being likely to develop recurrent depression into adult life. In the United States, effective intervention methods such as ‘TeenScreen’ have been used for many years (www.teenscreen.org) and intervention programmes established to manage emerging depression and to promote resilience (Stice et al. Reference Stice, Rohde, Seeley and Gau2008).

Other problems such as personality disorders and eating disorders that begin in adolescence also persist into adulthood. In eating disorders, a population-based cohort study showed that those who diet on a moderate or severe level at age 15 with low self-esteem, are up to 18 times more likely to develop an eating disorder 3 years later (Patton et al. Reference Patton, Selzer, Coffey, Carlin and Wolfe1999). Personality disorder, a contentious diagnosis in teen years, shows continuity of psychopathology from early adolescence to adulthood (De Clercq & De Fruyt Reference De Clercq and De Fruyt2007; Chanen et al. Reference Chanen, McCutcheon, Germano, Nistico, Jackson and McGorry2009; Winsper et al. Reference Winsper, Zanarini and Wolke2012; Winsper et al. Reference Winsper, Wolke and Lereya2014). The population-based Great Smoky Mountains Study prospectively assessed 1420 individuals approximately nine times from 9 to 21 years of age, reporting that by the age of 21 years, 62% of individuals met criteria for a major psychiatric disorder (Copeland et al. Reference Copeland, Shanahan, Costello and Angold2011). The overall prevalence for the development of a psychiatric disorder went up to 82.5% after including ‘not otherwise specified’ disorders. The study indicated that although there were a small proportion of young people with a DSM diagnosis, there is an increased risk of developing a clinical diagnosis by young adulthood.

The CAMHS–AMHS divide

There is increasing concern that the traditional CAMHS–AMHS divide does not meet the needs of adolescents with mental health problems. Problems at the interface between CAMHS and adult services are historical and rooted in the ideological, conceptual and practical differences in how these services are organised and delivered (Singh et al. Reference Singh, Cooper, Fisher, Tarrant, Lloyd, Banjo, Corfe and Jones2005; Singh, Reference Singh2009). AMHS emerged from a biomedical understanding of serious mental disorders such as schizophrenia and were influenced by neurology and phenomenology. Service provision is therefore based on treating individuals who suffer from diagnosable conditions, as elicited by individual psychopathology. Child psychiatry developed later, arising in a sociological context from concerns about delinquent, vagrant and traumatised youth. Over time it adopted a developmental framework in its understanding of child mental health problems and focused its interventions on the child as part of a family system. The change in care from CAMHS to AMHS is therefore accompanied by a radical change in the culture of understanding of mental ill-health and in the nature of services.

A recent US study used baseline data from the National Survey of Child and Adolescent Well-Being was conducted with 616 12–15 years old. The study reported a 33.3% reduction in the use of specialised mental health services during the transition period from baseline to 5–6-year-follow-up. The US Census Bureau in a nationally representative population-based sample on 1997 individuals reported reduced service utilisation at the peak transition age (16–25 years), an age group representing a period of rising incidence rates in adolescents with severe mental health disorders (Pottick et al. Reference Pottick, Bilder, Vander Stoep, Warner and Alvarez2008). They found an annual rate for service utilisation (inpatient, outpatients and residential) of 34/1000 for 16–17-year-olds and 18/1000 for 18–19-year-olds, suggesting a severe decline in service use at the time of significant risk.

Transition from CAMHS to AMHS is particularly problematic for many adolescents, with a large proportion dropping through a care gap at the interface (Singh et al. Reference Singh, Paul, Ford, Kramer, Weaver, McLaren, Hovish, Islam, Belling and White2010). Findings from the UK TRACK study show adolescents with a serious mental illness such as psychosis and bipolar disorder who are under CAMHS get referred to adult care especially if they are on medication or have been hospitalised. However, disorders such as ADHD, autism spectrum, mild learning disability or emerging personality disorder are either not referred to adult care or if referred, are not accepted. Those who do make the journey across services report that they were unprepared for the transition. Several report severe dissatisfaction and feel let down by the care they had received. For the majority, transition was poorly planned, poorly executed and poorly experienced. Many young people also experienced multiple other transitions during this period including leaving home and for females, getting pregnant (Singh et al. Reference Singh, Paul, Ford, Kramer, Weaver, McLaren, Hovish, Islam, Belling and White2010). Young people experienced a bewildering array of changes with many feeling overburdened and others feeling abandoned and neglected by services. There have therefore been major calls for action through intersectoral engagement that taps into the educational system, adolescent mental health services and health-related policy and legislation (Singh et al. 2010).

Have EIS improved care for adolescents?

A key mental health reform of the past two decades has been the development of early intervention in psychosis services which provide comprehensive, evidence-based high quality care for young people experiencing their first psychotic episode. Robust evidence from randomised trials (LEO and Danish OPUS trial) confirms that under specialised EIS, young people experience better clinical, social and vocational outcomes, have reduced inpatient stay and are better engaged (Craig et al. 2004; Petersen et al. 2005; Garety et al. Reference Garety, Craig, Dunn, Fornells-Ambrojo, Colbert, Rahaman, Read and Power2006). EI services appear to be highly valued by service users and their carers (Lester et al. Reference Lester, Birchwood, Bryan, England, Rogers and Sirvastava2009). When EI provision is supplemented by early detection teams in the community, people come to the services earlier, with less disabling symptoms, have better recovery and less suicidality (Melle et al. Reference Melle, Larsen, Haahr, Friis, Johannessen, Opjordsmoen, Simonsen, Rund, Vaglum and McGlashan2004; Larsen et al. Reference Larsen, Melle, Auestad, Friis, Haahr, Johannessen, Opjordsmoen, Rund, Simonsen, Vaglum and McGlashan2006; Melle et al. Reference Melle, Larsen, Haahr, Friis, Johannesen, Opjordsmoen, Simonsen, Rund, Vaglum and McGlashan2008). In the pre-psychotic (prodromal) phase, treatment has also been shown to reduce the risk of transition into full-blown psychosis (McGorry et al. Reference McGorry, Yung, Phillips, Yuen, Francey, Cosgrave, German, Bravin, McDonald, Blair, Adlard and Jackson2002; Morrison et al. Reference Morrison, French, Walford, Lewis, Kilcommons, Green, Parker and Bentall2004; McGlashan Reference McGlashan2005; Stafford et al. Reference Stafford, Jackson, Mayo-Wilson, Morrison and Kendall2013). EI services are also cost-effective as compared with generic teams and are recommended in the recent iteration of the UK NICE guidelines for ‘at risk’ and first episode cases. The success of EI services in therapeutically engaging young people (Lester et al. Reference Lester, Birchwood, Bryan, England, Rogers and Sirvastava2009) has led to calls for the EI paradigm to be extended to other mental and behavioural disorders (New Horizons, Department of Health, 2009), including the coalition policy on mental health (Social Care, Local Government & Care Partnership Directorate, 2014) and the Chief Medical Officer’s Annual Report on early intervention in youth disorders (https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/351629/Annual_report_2013_1.pdf).

Initial results from early intervention in other disorders are promising. For eating disorders, early intervention strategies have been developed for young women at risk, including internet-based behavioural therapy (CBT; Taylor et al. Reference Taylor, Bryson, Luce, Cunning, Doyle, Abascal, Rockwell, Dev, Winzelberg and Wilfley2006). A randomised 8-week, internet-based CBT intervention on college-age women with high weight and shape (n=480, high-risk group), showed significant reductions in their weight (assessed using the BMI) and shape concerns (assessed using the Weight Concerns Scale and Eating Disorder Inventory scale) for ∼2 years, which reduced the risk of eating disorder in a significant proportion of high-risk women. Similar work is being done for early intervention in emerging borderline personality disorder (BPD). Chanen et al. (Reference Chanen, Jovev, Djaja, McDougall, Yuen, Rawlings and Jackson2008, Reference Chanen, McCutcheon, Germano, Nistico, Jackson and McGorry2009) have shown that current evidence supports ‘indicated prevention’, targeting groups with precursor signs and symptoms such as substance use disorders or BPD traits, along with early intervention for first presentations for BPD.

The prevalence of unipolar depressive disorders is 0.4–0.8%, with 20% of adolescents being diagnosed by the age of 18 years (Lewinsohn et al. Reference Lewinsohn, Hops, Roberts, Seeley and Andrews1993; Birmaher et al. Reference Birmaher, Ryan, Wiliamson, Brent, Kaufman, Dahl, Perel and Nelson1996). Introducing early intervention strategies in young children (13-year-olds) has been shown to dramatically reduce the risk of depressive disorders in adulthood (Harrington & Clark, Reference Harrington and Clark1998). Longitudinal studies have reported that intervention in the first episode of depression is effective in halting the development of negative cognitive styles, which are typically associated with recurrent episodes (Lewinsohn et al. Reference Lewinsohn, Allen, Seeley and Gotlib1999). An evidence-based review reported efficacy of CBT in young people with anxiety and depressive disorders (Compton et al. 2004; McCrone et al. 2004), which was far more effective compared with tricyclic medications (Hensley et al. Reference Hensley, Nadiga and Uhlenhuth2004).

Along with neurodevelopmental disorders, emerging personality disorders often fall through the care gap between child and adult services (Singh et al. Reference Singh, Paul, Ford, Kramer, Weaver, McLaren, Hovish, Islam, Belling and White2010; Paul et al. Reference Paul, Ford, Kramer, Islam, Harley and Singh2013). The NICE guidelines have recommended more specialised interventions on antisocial personality disorder, such as parent-training programmes, brief strategic family therapy and multi-systemic therapy (Department of Health 2009; National Collaborating Centre for Mental Health 2009; Vizard et al. Reference Vizard, Jones, Viding, Farmer and McCrory2009), as well as group-based cognitive and behavioural interventions. Dialectical behaviour therapy has also been recommended for female service users with self-harm behaviour (Department of Health, 2009).

Launching of youth-based services: Birmingham initiative

Given the epidemiology of adolescent onset mental disorders, the lack of care continuity between CAMHS and AMHS, the high risk of disengagement from services in this age group and the success of EI services in bridging the CAMHS–AMHS divide, Birmingham and Solihull Mental Health Foundation Trust piloted a dedicated youth-based mental health service (www.youthspace.me). The Youthspace clinical team was initially a pilot from July 2011 to January 2012 and targeted young people up to the age of 25 years. The purpose of the pilot was to explore what clinical approaches are effective in working with a service user group who engage poorly with traditional mental health services and are at risk of developing long-term serious mental health problems. The service aimed to be rapidly responsive, youth friendly and intervene early using a broad range of interventions. Access to the youth services for young people aged 16–25 years was established through two pathways. First, the adult community mental health services introduced a youth access pathway across Birmingham, where the youth access teams conducted an assessment and diagnostic formation to the referring GP within 1 week of referral. This was followed by a brief CBT intervention and symptomatic treatment via medication needs by the GP following advice by the consultation team. Second, individuals were screened for risk of psychosis, bipolar disorder, eating disorders and personality disorder using an established framework (Lin et al. Reference Lin, RLEP and Wood2013).

The clinical team offered a quick response to the young person’s referral; flexibility/choice of venue with emphasis on non-stigmatising youth friendly environments; active signposting of young people to activity aimed at reducing NEET (the number of young people not in employment, education or training) status via community partnerships; an expert assessment and personalised plan of support; support and intervention from a named youth mental health practitioner; access to on-line support and information via www.Youthspace.me; and a focus on remaining independent and building resilience - quick discharge following completion of intervention but with quick re-access if required.

The pilot phase was evaluated externally through Health Innovation and Education Cluster (HIEC). Over the evaluation period, the service received 247 referrals (62% female, mean age 22) about the same number and profile as comparator CMHTs. As compared with CMHTs, Youthspace was able to offer a faster first contact following referral (mean 2 days as compared with 12), quicker first assessment (16 days as compared with 45), and markedly reduced ‘did not attend’ proportion (5% against 28%). Only 10% of those referred to the service actually required secondary care – the rest responded well to one-off expert assessment and personalised plan, brief to medium psychological intervention and active signposted to other support networks (32% were signposted to Princes Trust for focussed work relating to education, employment and training with 65% having positive outcome). There was very high level of user satisfaction and 67% reported making continued use of maintenance techniques provided through Youthspace intervention 12 months on from discharge (information collected post HIEC evaluation). Based on these positive results, the Birmingham commissioners are now radically re-commissioning services to develop an integrated care pathway for the population aged 0–25 (http://www.bhamsouthcentralccg.nhs.uk/patient-and-public-engagement/0-25-mental-health-services) to radically reform child and young AMHS to begin in October 2015. This will represent the first major service reform for young people’s mental health in the United Kingdom for over 30 years.