As an Attention Deficit Hyperactivity Disorder (ADHD) specialist, I have often wondered why children and young people with emotional dysregulation frequently require trials of more than one ADHD medication to optimise outcomes and support day-to-day functioning.
Neurodevelopmental disorders such as ADHD and autism are commonly encountered and managed by Child and Adolescent Mental Health Services (CAMHS). Many children and young people with these conditions also present with problems managing emotions, low mood, anxiety, self-harming behaviours and other challenges with their behaviour.
Mood instability, one facet of emotional dysregulation, involves frequent and significant changes in emotional state, such as increased irritability, anger, sadness, or anxiety. These changes can be difficult to control and often lead to problems with social, educational or family relationships (Broome et al., 2015). They have also been associated with longer hospital stays and more frequent admissions in adults aged 16 to 65 years (Mann F, 2015). But what about children and young people?
Mood instability continues to be a challenging co-morbid presentation. However, its impact on the need and duration of contact with mental health service is an unknown. Seker et al. (2026) sought to explore this by examining the association between mood instability at diagnosis and the frequency and duration of contact with CAMHS.
Methods
This study used anonymised electronic health records from children and young people with a diagnosis of ADHD, autism or both who had been referred to South London and Maudsley NHS Foundation Trust (SLaM) CAMHS between 2008 and 2022. Children diagnosed before the age of 5 and those admitted to an inpatient unit were excluded.
A natural language processing tool was used to search free text clinical notes and correspondence for key words relating to mood instability within three months either side of diagnosis. Outcomes included frequency of CAMHS attendance per year and length of time before discharge from the service.
Time to discharge was analysed using Weibull accelerated failure time models (to estimate how long young people remained in the service), and annual contacts using negative binomial regression (to analyse contact counts while accounting for differences in how frequently young people attended). Both models were adjusted for gender, age, ethnicity, deprivation, co-morbid mental health conditions and medication treatments.
Results
Study sample
The study included 21,906 children and young people: 7,674 (35%) with autism, 7,889 (36%) with ADHD and 6,343 (29%) with both conditions. Mood instability was identified in 13.6% of the autism group, 13.8% of the ADHD group and 20.1% of those with both conditions.
There was an association between the autistic population and lower deprivation, and ADHD and greater deprivation. Intellectual disability was the most common co-occurring diagnosis in the autistic group, whereas emotional disorders were most common in the ADHD group.
Contacts with services
Mood instability was associated with greater use of mental health services in children and young people with a diagnosis of autism, ADHD or both. The median total number of CAMHS contacts was highest in the group with both ADHD and autism compared to those with a single diagnosis. After adjusting for other factors, the researchers found that mood instability was associated with an increase in annual CAMHS contacts by 24% in the autistic group, 47% in the ADHD group and 27% in those with both conditions.
Children and young people with a diagnosis of ADHD and mood instability were noted to require more CAMHS input compared to those without. This association was strongest among those not receiving ADHD medication, who had 67% more annual CAMHS contacts. However, the study cannot tell us whether this greater use of services reflected more crises, behavioural challenges or emotional difficulties requiring intervention. The researchers suggested that crisis presentations, such as self-harm, may lead to intensive but short-lived CAMHS involvement.
Time to discharge
On the other hand, it was interesting to note that mood instability affected discharge time differently across the diagnostic groups. For the autistic group, mood instability did not clearly affect the time they spent in CAMHS. The study did not establish why this was the case, but suggested that mood instability may be a more enduring pattern in autism, intertwined with core features of the condition. On the contrary, in the ADHD group, mood instability was associated with shorter time to discharge. This was particularly evident among those not receiving ADHD medication, whose time to discharge was 30% shorter. There was no clear association in those receiving medication.
Conclusions
The researchers found that mood instability is associated with increased use of CAMHS for children and young people with neurodevelopmental conditions including autism, ADHD and both conditions being present. This was more evident in the population of young people with a diagnosis of ADHD not being treated with medication. The link may be weaker for those receiving treatment for underlying difficulties, such as ADHD medication. However, although data showed that mood instability did not prolong the time spent in the service, it was linked to greater need for support and contact with CAMHS.
They emphasised that their findings highlight the importance of identifying and assessing emotional dysregulation during initial neurodevelopmental assessment and treatment pathways within child and adolescent mental health service.
Strengths and limitations
One of the strengths of the study was its large sample size, consisting of over 21,000 children and young people identified over a period of approximately 15 years. This means the results reflect how mood instability is recorded in routine care. The sampling process appeared thorough and robust, with electronic health records used to identify eligible participants. A strong analytical methodology was also adopted, including adjustment for potential confounding factors.
However, it is questionable why children admitted to inpatient units were excluded, as this may have prevented the inclusion of more complex cases of mood instability commonly encountered within mental health services.
Sampling was also restricted to a single hospital trust in South London. This may limit the generalisability of the findings, particularly when considering other trusts that may have different care systems and approaches to managing mood instability within CAMHS.
Another limitation relates to the measurements used within the study. The researchers stated that participants had received diagnoses according to ICD-10 criteria. However, only 58% of autism diagnoses and 48% of ADHD diagnoses came from structured ICD-10 codes. The rest were picked up from clinical notes by a natural language processing tool, so some diagnoses may have been misclassified.
The records did not include reasons for discharge, so the study could not tell whether a shorter time in the service meant young people had disengaged or that their needs had been met, a gap the authors acknowledge. This matters for ADHD, a long-term condition that often persists into adulthood, where discharge from CAMHS may not mean the difficulties have gone away.
Overall, although sampling, analysis and findings from the study are applicable to CAMHS services with a similar patient management pathway, they may not be generalisable to others that do not conform to this system.
Implications for practice
The findings from this study emphasise that proper assessment, identification and early management of mood instability are important in mental health services, especially for children and young people with co-morbid autism and/or ADHD.
Emotional dysregulation should be considered a clinically important difficulty amongst children and young people accepted by CAMHS, and as such management pathways should be incorporated into every mental health service.
Assessment of mood instability or emotional dysregulation should be included as one of the baseline neurodevelopmental assessments within CAMHS to help improve patient care and release capacity for mental health services. This may mean lesser need for an individual to be seen or supported by CAMHS if addressed earlier on in the CAMHS journey.
Policy makers should also consider addressing the issue of mood instability as early as within primary care before presentation to CAMHS. Emotion regulation training has been proposed for inclusion into school curriculums as a strategy to improve outcomes (Albajara Saenz, A, 2022). Also, this “indicated prevention” strategy has been suggested to reduce the risk of developing depression in later years (Cuijpers, P, 2023).
As young people with ADHD and mood instability who were not taking ADHD medication had more CAMHS contacts and left the service sooner, it is important for CAMHS to ensure neurodevelopmental assessments are thoroughly done and treatment offered for ADHD as soon as practical. Services may want to review whether ADHD pathways offer treatment soon after diagnosis. This study cannot show that medication made the difference, and the authors note that children not on medication may have other complications, such as contraindications or limited family support.
For future research, there is a potential benefit from having digital health records systems that will allow for tools similar to natural language processing to be used to retrieve patient data with more ease. It would also be useful to conduct the study in other NHS trusts and settings to test whether the findings hold.
Statement of interests
Adepeju Deekae (she/her) has no conflicts of interest to declare.
Editor
Edited by Éimear Foley. Claude and ChatGPT assisted with language refinement and formatting during the editorial phase.
Links
Primary paper
Asilay Seker, Seungyoung Kim, Susie Chandler, Craig Colling, Rashmi Patel, Edmund Sonuga-Barke and Johnny Downs (2026). Mood instability and mental health service use in autism and attention-deficit/hyperactivity disorder: a natural language processing analysis of CRIS electronic healthcare records from 21 906 children and adolescents. BMJ Mental Health, 29(1), e302488. doi:10.1136/bmjment-2026-302488
Other references
Albajara Saenz, A. (2022) Emotion regulation skills for depression and anxiety in young people. The Mental Elf, 8 February. Available at: National Elf Service blog post
Broome, M.R., Saunders, K.E.A., Harrison, P.J. and Marwaha, S. (2015) ‘Mood instability: significance, definition and measurement’, The British Journal of Psychiatry, 207(4), pp. 283-285.
Cuijpers, P. (2023) Preventing the onset of depressive disorders #DepressionSolvingTheToll part 1. The Mental Elf, 17 November.
Maddox, L. (2014) Longer adolescent duration of worry and low mood predicts problems in adulthood: suggests early intervention important. National Elf Service, 15 April, 2014.
Mann, F. (2015) Mood Matters: mood instability is common and associated with poor outcomes. National Elf Service, 29 July. Available at: National Elf Service blog post
