Part 2. Patterns of service use

Mental health care in general practice

GPs are often the first port of call for children and young people with a new health concern. They play a key role in recognition and management of child and adolescent mental health problems and can provide referrals to specialised mental health services.

Analysis by NDL partners in North West London found that GPs account for an increased proportion of mental health-related service use by children and young people (Figure 4). Since 2015, the average monthly number of young people with a GP event or prescription related to mental health conditions increased by 194%. Over the same time, the number of young people seen at mental health trusts in the area increased by 138%. Even after adjusting for population growth, the rate of service use in both types of settings increased almost twofold.

Figure 4: Average monthly number of children and young people aged 0–25 years in North West London with a mental health-related GP event, and/or a mental health prescription, or seen at a mental health trust, 2015–2021

Source: NDL North West London analysis of Discover data.

Children and young people seen at a mental health trust for any mental health condition were mostly younger adolescents (Figure 5). In contrast, the vast majority of young people with a mental health-related GP attendance or prescription were older adolescents and young adults. This may partly be because GP diagnoses for conditions such as ADHD, which are more common in younger children, were not included in the analysis. Younger children may also be more likely to be referred to mental health services by schools.

Figure 5: Number of children and young people in North West London seen for mental health reasons in general practice or mental health trusts, by age group, 2019

Source: NDL North West London analysis of Discover data.

These findings suggest that in North West London, general practice has been playing an increasing role in the management and care of children and young people with mental health conditions. There is evidence that this trend exists across the UK, and that it is likely a symptom of both increasing need and overstretched specialist services.,, Most children and young people with mental health needs still receive no specialist support, and those who do often face long waits before treatment begins. The Children’s Commissioner found that in 2020/21 close to a quarter of referrals to CAMHS in England were closed before treatment started, and there is anecdotal evidence that the threshold for support may have risen further.,, For many children and young people who do not meet the criteria for specialist care, general practice may therefore be the main and only available source of mental health support.

Prescriptions for mental health medications

The frequency of antidepressant prescribing among children and adolescents in the UK has nearly doubled between 2006 and 2015. The group that accounted for most of this rise in prescribing was adolescent girls (aged 15–17 years) who were also more than twice as likely as boys of the same age to be prescribed antidepressants. A survey by NHS England in 2017 found that 1 in 50 (2.5%) of 5–19-year-olds were taking medication for a mental health-related problem. Among 17–19-year-olds, this figure rose to 1 in 20 (5.0%), with the most common type of medication being antidepressants. Children and young people with a hyperactivity disorder (ADHD) were the group with the highest prescribing rates (45.9%). Little published data are available on more recent trends, or on variation in mental health prescribing for children and young people by other demographic or socioeconomic characteristics.

NDL partners in Grampian analysed comprehensive, longitudinal local data on mental health prescriptions made in the community, using Scotland’s PIS dataset. They found that the percentage of children and young people (aged 0–24 years) with mental health-related prescriptions increased from 4.7% in 2012 to 6.4% in 2019. Prescribing rates were found to increase rapidly with age, driven by prescribing of antidepressants in the 19–24 age group (Figure 6). The inclusion of this age group also likely explains the higher prevalence of prescribing found in this analysis, compared with the earlier estimates for England. Young adults aged 19–24 years alone accounted for 55% of prescriptions.

Figure 6: Number of mental health prescriptions for children and young people in Grampian by age group, 2012–2020

Source: NDL Grampian analysis of Scotland's Prescribing Information System (PIS) data.

Figure 7: Types of mental health prescription for children and young people in Grampian by age group, 2012–2020

Source: NDL Grampian analysis of Scotland's Prescribing Information System (PIS) data.

NDL partners' analysis also found large age-related differences in prescribing between male and female children and young people. Children younger than 14 years were primarily prescribed medication for ADHD and close to 80% of prescriptions in this age group were for boys. Young adults aged 19–24 years were mostly prescribed antidepressants and 62% of prescriptions for young adults were for women. In 2019, 17.9% of young women aged 19–24 years had a prescription for antidepressants, compared to 9.1% of young men in the same age group. The Grampian team was also able to examine mental health-related prescription rates by area-level socioeconomic deprivation. They found that children and young people in the most deprived areas had twice as many prescriptions as in the least deprived areas (58 prescriptions per year per 100 children in the 20% most deprived areas, compared with 27 in the 20% least deprived areas).

Their findings about the rise in prescription rates are consistent with the observed rising prevalence of mental ill health, as are the findings about the variations in age and sex, for example the high rates of antidepressant prescribing for young adults. But these data beg contextual information not consistently recorded in administrative health care data, for example whether those receiving prescriptions for depression are also accessing appropriate psychological therapies and being regularly monitored according to NICE guidelines.

Referrals and contacts with specialist care

National prevalence data and other research suggest that mental health support needs vary by age, sex, ethnicity and socioeconomic circumstances. In the absence of more granular information on underlying need among different population groups, local bodies responsible for planning and delivering specialist CAMHS need to understand such variations, to better target services according to need.

Local data analysis of this kind can highlight variation that might indicate services are not meeting needs. In the West Midlands, for example, analysis by the Strategy Unit found that black children and young people had more frequent contact with mental health services but shorter contact time, had the highest re-referral rates and were most likely of all ethnic groups to have prolonged service needs. While NDL teams found that ethnicity data were too poor to use, and interpreting them was challenging due to missing data on population size by ethnicity, they did find striking differences in the use of services by age, sex and deprivation.

Referrals and contact patterns by age and sex

Despite differences in data sources and analytical approaches, NDL partners found similar patterns of mental health service use by age and sex. The NDL team in Grampian examined the number of referrals to specialist mental health services for children and adolescents (aged 2–17 years). Two NDL partners in England analysed the number of care contacts with specialist mental health services (including those aged 0–25 years in Liverpool and Wirral, and 11–25 years in Leeds).

Among children younger than 12 years, boys accounted for the majority of mental health referrals in Grampian and of care contacts in Liverpool and Wirral (Figure 8). Consistent with the national prevalence estimates of mental health disorders in England, from the age of 12 years girls were more likely than boys to be referred to or receive specialist mental health care, with particularly stark sex differences around the time of adolescence. Referrals and contact rates for adolescent girls increased with every year of age, up to the ages of 15 and 16, respectively. In Leeds, adolescent girls accounted for 72.6% of all contacts of those aged between 15 and 17 years at the time of referral.

Research has found plenty of evidence of gender disparities in mental health among young people. A large study of primary care records in England found that the incidence of adolescent girls presenting with self-harm has been rising sharply since 2010, and children and adolescents who harm themselves were also found to be at an increased risk of suicide.

Figure 8a: Number of referrals to CAMHS for children and young people aged 2–17 years in Grampian by age at referral, 2015–2021

Source: NDL Grampian analysis of CAMHS data.

Figure 8b: Number of children and young people aged 0–25 years with mental health contacts in Liverpool and Wirral by patient age, 2019–2021

Source: NDL Liverpool and Wirral analysis of MHSDS, SUS and ECDS data.

Transition to adult mental health services (AMHS)

Adolescence and young adulthood are often turbulent and vulnerable times, marked by key developmental milestones such as leaving school and moving out of the family home. Both the incidence of mental illness and the risk of disengagement from mental health services are high in this age group. In addition, between age 18 and 25 young people are required to transition to adult mental health services (AMHS), which may have different acceptance thresholds. This service ‘cliff edge’, and the often poor support offered during transition, has been a long-standing concern.

While there is little quantitative evidence on the proportion of young people who successfully transition to adult services, qualitative work has suggested that up to a third of young people are ‘lost from care’ during this period, with another third facing disruptions to care. Previous research also found that young people struggle with the cultural shift between CAMHS and AMHS, from the nurturing environment in CAMHS, which emphasises the family unit, to a more impersonal atmosphere in adult services, which are more likely to treat young people as autonomous adults and offer limited family involvement in favour of personal privacy.

While this transition typically occurs between ages 18 and 21, there is large variation between areas in the age ranges covered by children and young people’s services, and in how young people are supported beyond CAMHS. Due to the poor quality and completeness of mental health data, our understanding of young people’s experience of transition and how services can improve transitions is extremely limited.

Although NDL teams attempted to examine this using patient-level linked data, poor quality and completeness of data fields prevented them from reliably identifying whether recorded care contacts were with child and adolescent or adult mental health teams. As a proxy measure to study the relationship between patient age and ongoing engagement with the service, NDL partners in Leeds quantified the proportion of young people still in contact with services after 1 year, shown by age at first contact (Figure 9). They found that this proportion decreased at around age 17, when young people approach transition age, and remained lower for older ages. This could indicate that young people around transition age are more likely to lose contact with services or that the length of treatment may be shorter for adult mental health services.

Figure 9: Percentage of children and young people in Leeds still in contact with services 1 year after their first contact, by age at first contact

Source: NDL Leeds analysis of MHSDS data.

Socioeconomic deprivation

There is considerable evidence of the links between socioeconomic deprivation and child mental health problems., Responses to surveys in England show a clear correlation between lower parental income and poorer mental health among children and young people. However, data on the relationship between socioeconomic deprivation and mental health service use among children and young people are currently only available for Scotland and remain experimental. These data show a socioeconomic gradient in referral rates, with rates being consistently higher for people in the most deprived areas.

To explore this relationship in more detail, NDL partners were able to link patient records to the Index of Multiple Deprivation (IMD) from England (2019), Scotland (2020) and Wales (2019), based on the area where patients lived. They then compared mental health referrals, contacts and outcomes between young people living in the 20% most deprived areas and the 20% least deprived areas.

Analysis of referrals to specialist mental health care from the NDL team in Grampian found that there were 1.7 times more referrals for young people in the most deprived areas compared with young people in the least deprived areas. Young people from the most deprived areas were also younger when referred: the average age at first referral in the most deprived areas was 10.4 years, compared with 11.6 years in the least deprived areas.

The NDL team in Leeds found similar patterns in care contacts, with children and young people living in the most deprived areas most likely to be in touch with specialised mental health services. Interestingly, they also found a high proportion of young people in contact with services in the least deprived areas, which warrants further investigation, and may indicate differences in self-recognition of support need or in the ability to navigate access in a complex system (Figure 10). This hypothesis is supported by previous analysis highlighting that children and young people in less deprived areas are more likely to self-refer to mental health services. It is important to note that none of the data used for this analysis capture privately funded mental health services, which will affect the steepness of the deprivation gradient, as they may be more frequently used by children and young people in affluent areas.

Figure 10: Percentage of children and young people aged 11–25 years in Leeds who were in contact with CAMHS by IMD quintile, 2017–2020

Source: NDL Leeds analysis of MHSDS data.

As national estimates on the prevalence of mental health disorders are not available by socioeconomic deprivation, it is difficult to draw conclusions about how these patterns relate to underlying support need. However, among existing patients aged 11–25 years in the most deprived areas, the rate of both referrals and crisis referrals in Leeds was higher (Figure 11). The difference in crisis referrals is particularly pronounced, with 60% more referrals for patients in the most deprived areas than for patients in the least. This may indicate that children and young people in more deprived areas are not only more likely to have mental health support needs, but that these needs may also be more severe.

Figure 11: Number of referrals and number of referrals to a mental health crisis team (crisis referral) per 100 patients aged 11–25 years in Leeds by IMD quintile, 2016–2021

Source: NDL Leeds analysis of MHSDS by NDL Leeds.

These findings reinforce the importance of better understanding how children and young people navigate the complex referral pathways and how this varies by socioeconomic deprivation. For example, analysis by the Strategy Unit found that children and young people in more deprived areas are more likely to have their referral deemed ‘unsuitable’, have shorter contact time during an appointment and are more likely to be re-referred back into services within a year, despite a higher percentage completing their treatment plans.

Mental health crisis presentations

For children and young people experiencing mental health crises, timely access to appropriate support is essential. Children and young people may present in crisis at many different entry points to the NHS, and not all may require hospital care. In order to target action to improve crisis support, service planners and decision makers need a better understanding of the incidence of mental health crisis presentations, and which children and young people are at a high risk. Pioneering analysis from NHS Wales has already demonstrated how examining crisis presentations across the emergency care pathway can inform a more joined-up, whole-systems approach to crisis care. However, the data sources currently available at the national level only provide insight into small parts of the crisis care pathway, such as A&E attendances or hospital admissions, making it challenging to assess the scale of demand and whether young people’s needs are being met in an effective and timely manner.

Building on their previous work, NDL partners in Wales created a novel linkage of routine secondary care data with data from the Welsh Ambulance Services Trust (WAST). A strength of this approach is that the data are national and capture all ambulance callouts in Wales, allowing a more comprehensive picture of acute care use by children and young people experiencing mental health crises.

Figure 12: Children and young people aged 11–24 years presenting to WAST with mental health crises by outcome, 2018–2020

Unknown/other includes cases where clinicians requested transport, hoax calls or erroneous data, no patient found at the scene or ambulances were cancelled pre-arrival.

Source: NDL Wales analysis of Welsh Ambulance Service Trust, emergency department and admissions data.

The analysis included 4,638 ambulance callouts related to children and young people aged 11–24 years presenting with mental health crises between 2018 and 2020, including self-harm, suicide attempts, overdose, psychosis, and other serious mental illnesses requiring urgent care. The analysis found that in 2 out of 3 cases (63%), patients were subsequently transferred to an emergency department (Figure 12). Of these, the majority were not admitted to hospital (but they might have been referred to outpatient services or to their GP). Overall, only 21% of ambulance callouts related to mental health crises resulted in admission to hospital. Some of this difference may reflect challenges in interpreting ambulance callout and emergency attendance data.

Improving the completeness and quality of these data has the potential to provide much richer information on the care pathways for children and young people in crisis attended to by ambulance. The findings highlight that data available at the national level, such as A&E attendances and emergency admissions, only capture a fraction of mental health crises that present to ambulance services. Just under 1 in 10 patients (9%) attended by ambulance teams refused care. To improve service planning and provision, it will also be important to understand the underlying reasons why some young people refused treatment.

The Wales NDL team also investigated patterns in crisis events in more detail, including ambulance attendances, visits to A&E and emergency hospital admissions. They found that in 2019, girls (11–15 years) and young women (16–19 years) were twice as likely to present in crisis than boys and young men of the same age (Figure 13). The rates of crisis events generally increased with age, but they were highest among young women aged 16–19 years. Crisis event rates were also strongly patterned by socioeconomic deprivation, with children and young people living in the 20% most deprived areas in Wales having almost double the rate of crisis events compared with those living in the 20% least deprived areas (Figure 14).

Figure 13: Rates of mental health crisis events among children and young people in Wales, by age and sex, 2019

Mental health crisis events include WAST attendances, A&E attendances, and emergency hospital admissions.

Source: NDL Wales analysis of Welsh Ambulance Service Trust, emergency department and admissions data.

Figure 14: Rates of mental health crisis events among children and young people in Wales, by Welsh Index of Multiple Deprivation (WIMD) quintile, 2019

Mental health crisis events include WAST attendances, A&E attendances, and emergency hospital admissions.

Source: NDL Wales analysis of Welsh Ambulance Service Trust, emergency department and admissions data.


** GP events include mental health diagnoses, observations and referrals to other mental health services. For a full list of codes for events and prescriptions see the technical appendix.

†† Anxiety, depression, bipolar disorder, eating disorders, personality disorders, schizophrenia, self-harm, or harmful thoughts.

‡‡ According to NICE guidelines, antidepressants (fluoxetine) should only be offered to children and young people (5–18-year-olds) if moderate to severe depression is unresponsive to a specific psychological therapy after 4–6 sessions, and then only in combination with concurrent psychological therapy.

§§ This analysis was performed by linking patient-level prescription information to quintiles of the Scottish Index of Multiple Deprivation (SIMD, https://www.gov.scot/collections/scottish-index-of-multiple-deprivation-2020).

¶¶ In Grampian, this included referrals to CAMHS, and in Leeds, and Liverpool and Wirral it included referrals to tier 2 and above services included in the Mental Health Services Dataset (MHSDS).

*** Between 2015 and 2021, there were 29.1 referrals per 100 children living in the 20% most deprived areas and 17.1 referrals per 100 children living in the 20% least deprived areas.

††† Referrals to tier 2 and above services included in the Mental Health Services Dataset (MHSDS).

‡‡‡ The proportion of referrals that were accepted were similar across deprivation quintiles (not shown).

Previous Next