Response

Sayal and colleagues report findings from a secondary analysis of the STADIA trial involving 1225 children and young people aged 5–17 years with emotional difficulties and referred to child and adolescent mental health services (CAMHS) in England. ^{Reference Sayal, Holt, Partlett, Wyatt, Bhardwaj and Gledhill1} STADIA examined the use of a standardised diagnostic tool in supporting referral and assessment, ^{Reference Sayal, Wyatt, Partlett, Ewart, Bhardwaj and Dubicka2} but the trial also provided an opportunity to collect real-world data examining outcomes following referral. The findings raise important questions about the role of CAMHS and how well services identify and meet the needs of the populations they serve. In the context of rising demand, this speaks to the challenge of how limited resources can be allocated as effectively and equitably as possible to improve the mental health of children and young people. We have selected three main areas for comment.

First, although we would anticipate that severity of emotional difficulties would be the major predictor of referral acceptance, the authors find that participant-reported symptoms and impairment were not associated with index referral acceptance by 12 months, nor were they associated with receipt of a diagnosis or treatment being offered by 18 months. Furthermore, in adjusted models, participants living in more deprived areas were less likely to have their index referral accepted. This finding may reflect ‘inverse care law’ dynamics, such as increased pressures on referrers in deprived areas impacting referral quality/detail, or the role that family resources (or lack thereof) might play in ‘successful’ referrals. There might be relevant, unmeasured factors that may have informed decisions to accept referral or offer treatment, such as comorbidity, markers of social or clinical complexity or availability of alternative services. However, Sayal and colleagues appear to be describing a system that does not accord with the principle of delivering care proportionate to need, painting a picture of compounding inequity, especially for those living in the most deprived areas.

Second, the paper provides insights into the relationship between diagnosis (as collected from CAMHS records) and service receipt. It is remarkable that only 23% of those with an accepted referral were given a diagnosis (of an emotional disorder), whereas 71% were offered treatment or intervention, with the only predictor of treatment being older age. This may partly reflect reluctance to assign and/or record diagnoses within CAMHS, perhaps due to differences in perspectives on conceptualisation and the value of formal diagnosis in children and young people. ^{Reference Gill, Jones, Price, Goodison and Tyson3} Older children and those with previous/existing diagnoses at the time of referral were more likely to have a diagnosis recorded – factors that may increase both diagnostic certainty with the passage of time, and the acceptability of assigning a formal diagnosis. Nonetheless, the lack of recorded diagnostic information makes it challenging to achieve a more granular understanding of who is being seen, and what outcomes we can expect from the interventions offered, which arguably hampers child mental health research to a greater degree than in other fields.

The observed mismatch between rates of diagnosis and receipt of treatment suggests that children and young people who do not meet diagnostic criteria are still being offered interventions within CAMHS, possibly reflecting the use of a needs-based approach (e.g. the THRIVE framework). ^{Reference Wolpert, Harris, Hodges, Fuggle, James and McKenna4} However, from the latest national child mental health survey, we also know that many who do meet diagnostic criteria are not in contact with specialist mental health services, and that children from socioeconomically disadvantaged and minoritised ethnic groups are less likely to report contact, ^{Reference Mathews, Ford, White, Ukoumunne and Newlove-Delgado5} linking back to our earlier comment on compounding of disadvantage.

Third, the authors find that referral acceptance was not associated with improved outcomes in terms of meeting ‘caseness’ criteria at 18 months post-referral. It is important to note that caseness criteria may not reflect improvement in other domains that are important to the child and family. The analyses are also impacted by a common limitation of studies of routine clinical outcomes, which is a lack of detailed information on what was provided, to whom and at what ‘dose’, meaning that we do not know whether participants received an adequate intensity of evidence-based treatment. It remains concerning that, of those referred to CAMHS, being from a more deprived area was associated with higher caseness and increased severity of depression at 12-month follow-up. This may speak to the role of socioeconomic disadvantage in the development and maintenance of emotional difficulties, ^{Reference Chua, Schlüter, Pearce, Sharp and Taylor-Robinson6} and leads us to consider whether traditional CAMHS may be less effective in some contexts, with greater efforts needed to increase equity.

This commentary is not intended as a critique of the efforts of colleagues in CAMHS: we can safely assume that they have set out to provide effective and equitable services. However, why is it that the current systems and processes of referral, assessment and management appear to result in these outcomes? The interim report from the Department of Health and Social Care’s independent review into mental health conditions, attention-deficit hyperactivity disorder and autism ⁷ identifies the problem as follows: ‘The service challenge, therefore, is not simply that more people may be in difficulty, but that increasing numbers are presenting to systems that are not always well configured to sort, prioritise and support them efficiently’; and notes, ‘The Review also does not yet know enough about who may still be missed by current systems… This reinforces the need for more fine-grained analysis of who is, and is not, entering diagnostic pathways’. We would agree that this is the first part of the task. Arguably, this paper raises as many questions as it answers, with much more work to do, both on our qualitative understanding of what is going on in CAMHS, and on improving data capture to study the nature of need and the match with the treatment on offer, particularly in socioeconomically deprived communities. The STADIA trial design somewhat limits our ability to draw conclusions about equity for disadvantaged populations more broadly: first, predominantly online participation may have introduced barriers for more disadvantaged families. Second, we note that urgent referrals were excluded from the trial, and children and young people from deprived backgrounds may be more likely to present later and with higher acuity.

The next part provides a thornier challenge: if need and services are not matched, and the system is not ‘sorting’ or ‘supporting’ effectively in the face of rising demand, how do we rebalance the system? In our view, a more equitable and active approach to identifying children and young people who could benefit from specialist help is required, with rigour in referral, triage and assessment. Attention and resources are also required to ensure that those who do access CAMHS can engage in targeted, evidence-based treatment at sufficient intensity and duration to be effective, coupled with prevention and early intervention efforts as part of a holistic system of support to promote the mental health of our children and young people.