Improving outcomes for people with autism spectrum disorders by reducing mental health problems: the IAMHealth research programme including one RCT
Background Autism is a neurodevelopmental condition whose core symptoms include impairments in social communication, restricted and repetitive behaviours and sensory atypicalities, which can have varying severity. Most autistic people experience additional, impairing mental health and behavioural pr...
Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
NIHR Journals Library
2025-05-01
|
| Series: | Programme Grants for Applied Research |
| Subjects: | |
| Online Access: | https://doi.org/10.3310/YRKP9867 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Background Autism is a neurodevelopmental condition whose core symptoms include impairments in social communication, restricted and repetitive behaviours and sensory atypicalities, which can have varying severity. Most autistic people experience additional, impairing mental health and behavioural problems, but these are often under-recognised by healthcare professionals, autistic people and their caretakers. Objective(s) We aim to improve identification of mental health problems by developing a tool for clinical use, which can also be used to monitor treatment response. Design Work package 1: we developed and validated a new instrument to provide improved detection of mental health and behavioural problems in autistic people from childhood through to adult life. Work package 2: we explored how autistic young adults understand and manage their mental health. Work package 3: we undertook a cohort study to identify risk and protective factors for mental health and behavioural problems in autistic adolescents. Work package 4: we undertook a pilot feasibility randomised controlled trial of Predictive Parenting compared to group-based psychoeducation and active control intervention. It was not the aim of the pilot feasibility randomised controlled trial to undertake hypothesis testing. Setting Participants in work package 1 were ascertained through clinical sites within London and Liverpool and through specialist autism schools in London. In work package 2, participants were selected from a cohort originally ascertained from 11 regions across south-east England. Participants were drawn from London Boroughs of Bromley and Lewisham (work packages 3 and 4) and London Borough of Lambeth (work package 4). Participants Work package 1: 255 parents of autistic children/adolescent; work package 2: 19 autistic young adults; work package 3: QUEST cohort of 277 children; work package: 62 children. Intervention Predictive Parenting – a novel parent-mediated intervention. Main outcome measure Work package 4: a blinded observational measure of child behaviours that challenge. Results We developed the Assessment of Concerning Behaviour to be completed by parents/caretakers, autistic children/young people/adults and teachers, and showed it has two reliable and valid subscales reflecting emotional and behavioural problems. We identified that poor or incomplete understanding of autism affected young adults’ and parents’ understanding, discernment and management of mental health difficulties. We showed strong continuity of emotional and behavioural problems as well as attention deficit hyperactivity disorder from early childhood to late adolescence, with prediction being largely within domain (emotional, behavioural or attention deficit hyperactivity disorder). Early childhood attention deficit hyperactivity disorder symptoms had a significant negative impact on adolescent everyday functioning. At an individual level, parents’ accounts suggested multiple factors may affect mental health trajectories and outcomes in the late teenage years. Our pilot feasibility trial of our new intervention, Predictive Parenting, directed at parents of young autistic children was highly acceptable and feasible to deliver. Limitations To date, only the parent version of the Assessment of Concerning Behaviour has had its psychometric properties ascertained. We combined clinical and non-clinical samples and the scale could have different psychometric properties for these two groups. The qualitative work in work package 2 was limited to young adults without an intellectual disability and there was under-representation of females and non-white ethnicity, as well as those with severe mental health problems. The QUEST cohort in work package 3 was derived from those receiving an early autism diagnosis, who are more likely to have severe autistic presentations and intellectual disability, so the results may not generalise to the full autistic population. The pilot feasibility study had a small sample size and hence modest power to detect group differences; the lack of an objective rating of intervention fidelity; the lack of a treatment as usual group to track the natural trajectory of child and parent behaviours over time; and the fact that although the researchers who coded the observational measure were blinded to intervention allocation, they were not blinded to time point. Conclusions The research undertaken in the current programme shows that mental health and behavioural problems are more common in autistic people and are strongly persistent over time, even when they commence in the early childhood period. Interventions for mental health and behavioural problems are a priority for autistic people and their families. However, we showed that autistic people and their families often find it difficult to discern the difference between autistic features and mental health and behavioural problems. Future work A definitive randomised controlled trial including an economic evaluation is needed to determine the effectiveness and cost-effectiveness of Predictive Parenting. Future longitudinal research could focus on modifiable risk and resilience factors related to mental health problems in autistic people and could determine whether routine use of mental health screening questionnaires increases the identification and treatment of mental health problems in autistic children and young people. Trial registration This trial is registered as Current Controlled Trials ISRCTN91411078. Funding This award was funded by the National Institute for Health and Care Research (NIHR) Programme Grants for Applied Research programme (NIHR award ref: RP-PG-1211-20016) and is published in full in Programme Grants for Applied Research; Vol. 13, No. 5. See the NIHR Funding and Awards website for further award information.
Plain language summary Background Two-thirds of autistic people have mental health disorders. These disorders are distressing to autistic people and their families/caretakers, may limit access to education and reduce employment opportunities and community participation. Autistic people and their families prioritise research that will improve their mental health. Methods We wanted to better understand how autistic people and their families recognise and seek help for mental health problems. To identify early risk factors for ongoing health problems, we followed a group of young autistic children to adolescence. To reduce distressing mental health and behavioural problems, we undertook a small pilot feasibility randomised controlled trial of a novel parent-mediated group intervention for young autistic children to see whether parents were interested and could take part in the intervention. Results We discovered that autistic young adults found it difficult to distinguish between autism and mental health. They were less likely to seek help for their mental health problems. Our new measure of mental health and concerning behaviours, co-produced with autistic people and parents, shows good reliability and validity. From childhood to adolescence, mental health and behaviour problems are likely to persist rather than go away on their own. Parents’ accounts indicate that many factors may affect an autistic child or teenager’s mental health – both in positive and negative ways. Our novel intervention showed high rates of completion and satisfaction. Conclusions Mental health and behaviour problems in autistic people are common and persist over time. Better identification and effective, targeted interventions are required to improve outcomes. These should start in early childhood. It is important to ensure autistic people and their families know about mental health problems. Mental health questionnaires that are easy for autistic people to complete could improve recognition by autistic people, parents/caretakers and professionals alike.
Scientific summary Background Autism involves pervasive impairments of reciprocal social interaction and communication as well as repetitive behaviours and interests and sensory atypicalities, including hyper- and hypo-sensitivity to different stimuli. Autism is common, occurring in ~1–1.5% of the population, lifelong and hugely expensive. Currently, there are very limited treatments for its core symptoms. There are also widely divergent views on whether core autistic traits should be a target for intervention or rather that societies and environments should be more flexible to accommodate neurodivergence (or both). However, previous research and consultation with autistic people and their families/caretakers highlighted the importance of commonly associated mental health problems (MHP). Referred to as mental health and behavioural problems (MHBP), these can lead to exclusion from everyday family life, educational and community activities, poor quality of life and increased stress for family members as well as high costs in service use and lost opportunities. Previous research has shown elevated rates of MHBP at all ages among autistic people, affecting up to two-thirds of individuals, but at the inception of this programme, it was not known whether these problems persisted over time in the same individuals, nor what are the risk factors for persistence. Aims and objectives This programme focused on decreasing MHBP as a strategy for improving outcomes for autistic people and their families. These outcomes include improved mental health, quality of life and community participation for autistic people; reduced family stress; and decreased economic costs by ultimately lowering the need for high-cost (often residential) care and integration into the community. To achieve this, we focused on improved recognition, early intervention and identification of the factors that predict MHBP and influence transitions to adolescence/early adult life. Autism is a lifelong condition, and this programme reflected this through work packages focusing on key time points from early childhood to young adult life. To achieve our aims, we addressed the following objectives: We developed and validated a measure of MHBP in autism, to assist professionals in detecting these problems, identifying their causes and monitoring treatment/intervention. (Work package 1 – instrument development.) We interviewed young adults and parents of autistic young people and young adults, in order to understand and describe their perception of the emergence of MHP their experiences of seeking help; and the impact of these on their lives. (Work package 2 – biographies.) We undertook a longitudinal study of a cohort of autistic adolescents in order to identify the personal, family and wider environmental risk/protective factors related to persistence/desistence of MHBP from early childhood to late adolescence. A nested qualitative study investigated parents’ accounts of their child’s mental health trajectories. (Work package 3 – predictors.) We developed and completed a pilot feasibility randomised controlled trial (RCT) of a novel intervention for parents of recently diagnosed children aimed at reducing MHBP, enhancing child and family functioning and decreasing parental stress. We compared it to a control intervention. (Work package 4 – treatment.) Work package 1: instrument development Methods We undertook focus groups with autistic adolescents and adults and their parents/caretakers as well as mental health professionals. With autistic people and their parents, we wanted to identify the most effective and understandable ways of conveying item content and appropriate scoring. To assess the instrument, participants were identified through mental health and paediatric clinics, schools for autistic children and schools with special units for autistic children. We then asked autistic people, their parents and teachers to complete the questionnaire. A subset completed the questionnaire on two occasions to obtain test–retest reliability. Exploratory factor analysis (EFA) identified the structure which was replicated in an independent sample using confirmatory factor analysis (CFA). Reliability and validity were assessed for the final solution, for each factor separately. Convergent and discriminant validity were measured against existing measures. Key findings Item content, presentation and response format are very important to autistic people. The Assessment of Concerning Behaviour (ACB) was completed by 255 parents, 149 autistic children and young people and 30 teachers; test–retest data were available from 121 parents and 61 children/young people. Target participants (across all respondents) had an age range of 7–29 years; self-reports were completed by youth aged 8–14 years. Male preponderance varied from 75% to 83.6%. Mean IQ varied from 63.8 to 77.8 with a range from the profound intellectual disability (ID) range to superior IQ. EFA supported a two-factor model as providing the best fit [χ2/degrees of freedom (df) = 1.7, root-mean-square error of approximation (RMSEA) = 0.053, Comparative Fit Index (CFI) = 0.91] compared to a one- or three-factor model. This was validated by CFA on a second sample (χ2/df = 1.7, RMSEA = 0.057, CFI = 0.88) which was compared to a one-factor model. Within-factor reliability and stability were judged satisfactory with Cronbach’s weighted kappas ranging from 0.51 to 0.72 and per cent agreement from 83% to 95.5%. Concurrent, convergent and discriminant validity was supported by the pattern of correlations with other measures. Limitations There were relatively low completion rates by children, adolescents, and young adults, as well as teachers, in comparison with parents, meaning that a full psychometric profile could only be generated for the parent version. However, up till adult life, among autistic populations, it is usual to rely on parent report as the primary informant. Interpretation Co-design of a questionnaire with autistic people and their families led to a different structure and response format than is typical for questionnaires about MHBP. The ACB was subjected to stringent psychometric evaluation, including replication of the structure in a second sample, and was found to be robust. Work package 2: biographies Methods Using an existing research cohort, we purposively sampled autistic young adults with previous experience of a range of MHBP. Nineteen autistic young adults aged 23–24 years were recruited. Parallel interviews were undertaken with parents. In-depth interviews explored how they understood and managed MHP. Data were analysed thematically, and this framework was shared at an early stage with the patient and public involvement (PPI) panels. Key findings Young adults adopted self-management strategies rather than seeking advice or intervention from more conventional sources, including clinical services. Factors contributing to this included beliefs about the causes of MHP and increased vulnerability with the context of a diagnosis of autism, knowledge of self-management and, based on prior experiences, a view that professional support or intervention was unavailable or inadequate. Where help was sought, this was only at the point of psychological distress becoming very apparent to parents (typically due to concomitant physical symptoms such as significant weight loss) who typically either initiated or supported help-seeking. Limitations The study focused on young adults without learning difficulties (IQ < 70). There was an under-representation of females and people from ethnic minority backgrounds in the cohort from which we recruited. This means that we have only a partial understanding of these issues for autistic adults. Interpretation Young autistic adults and their families may hold erroneous beliefs about autism and mental health, and, as a result, struggle to discern when they might need mental health support. Negative or unhelpful experiences of mental health support during childhood and the teenage years may engender a suspicion or reluctance to seek help from mental health services. There were few systematic opportunities for autistic young people or their parents to learn about autism, including its implications for mental health. In addition, health and social care professionals need to be aware of the high rate of MHBP in autistic people and that autistic people may not recognise their own MHBP or feel confident in seeking professional help for them. Work package 3: predictors Methods We followed up the QUEST cohort, a sample of 277 autistic children first assessed at age 4–9 (Wave 1) and followed up at ages 11–16 (Wave 2) and 13–18 years (Wave 3). A particular focus was on the role of family factors, including maternal stress and mental health and family child-rearing practices, alongside wider environmental experiences, such as type of schooling and bullying, on MHBP. MHBP were assessed with parent- and teacher-reported questionnaire measures at Wave 2 and parent- and self-reported questionnaires and parental psychiatric interview for the intensive subset at Wave 3. These were conceptualised in three domains: emotional problems, behavioural problems and attention deficit hyperactivity disorder (ADHD) symptoms. Parents reported on their own MHP at each Wave. Regression analysis and structural equation modelling were used to examine longitudinal relationships. A nested qualitative study of parents (n = 33) of autistic teenagers (15–19 years), purposively recruited from the cohort, sought to collect parents’ accounts of their child’s mental health from diagnosis to the present, their beliefs and observations about the factors which affected it, and the impacts of MHBP on them as parents. Key findings We demonstrated moderate to strong persistence of mental health symptoms and diagnoses in autistic children over more than 10 years. Once initial comorbidity of symptom domains was accounted for, stability was largely within domain. Adolescent adaptive functioning was predicted not only by early childhood autistic symptoms and IQ and ADHD symptoms. Higher parental MHP at Wave 1 was found to be associated with lower child IQ (β = 0.2), but not autistic symptoms. The nested qualitative study revealed that multiple factors may protect against, or increase the risk of, MHBP during a child’s life. These include bio-psychological (e.g. IQ, communication skills, puberty, social and cognitive development) and social ecological factors (e.g. parenting skills, family and school environment) factors. Parents described feeling skilled and competent in supporting their autistic child until the early teenage years, when they encountered new, more challenging difficulties. Limitations The QUEST cohort is ascertained from a population of children diagnosed in two London boroughs before age 4 years. Thus, while this is a carefully characterised population, it reflects children diagnosed early in life and does not include those with more subtle presentations who may only be recognised as having autism later. The nested qualitative study focused only on parents’ accounts. Interpretation Mental health and behavioural problems showed moderate to strong stability from early childhood to adolescence, supporting the importance of early detection and appropriate intervention. While parents self-report high levels of MHP, these do not appear to be strong predictors of subsequent child MHBP. Nevertheless, the association between parent MHP and child IQ suggests that clinicians should attend to the well-being of parents, especially those whose children have ID. Efforts to minimise the risk of MHBP among autistic children and teenagers need to be multifaceted with interventions and support available for children, parents and schools. Work package 4: treatment Methods This was a pilot feasibility RCT comparing a 12-week group behavioural parenting intervention [Predictive Parenting (PP)] to an attention control [psychoeducation (PE)]. Parents of 62 4- to 8-year-old autistic children were randomised to PP (n = 31) or PE (n = 31). The primary outcome was a blinded observational measure of child behaviours that challenge. Secondary outcomes included observed child compliance and parenting behaviours; parent- and teacher- reported child emotional and behavioural problems; self-reported parenting practices, parental stress, self-efficacy and well-being. Key findings Recruitment, retention, completion of measures, treatment fidelity and parental satisfaction were high for both interventions. There were no significant differences on other measures. Limitations Predictive Parenting was compared to an active intervention of PE delivered by experienced clinicians. Although recommended by many professionals, PE is not routinely available as treatment as usual and thus this comparison does not reflect the potential augmentation of current practice that could be conferred by PP. This is a pilot feasibility trial that requires a definitive evaluation including estimation of cost-effectiveness. Interpretation Predictive Parenting is an acceptable and feasible intervention to deliver. The MHBP it is tackling are important targets for intervention. Conclusions We have shown that MHBP in autistic people at different time points show high levels of persistence in the same individuals, highlighting the importance of early recognition and targeted, autism-specific interventions. Furthermore, our finding that young autistic adults may have difficulty recognising MHP as distinct from autistic symptoms increases the need for autism-specific instruments to detect MHBP, such as the ACB. Co-design of instruments with autistic people and their parents may be important in using language and formats that assist autistic people and those informing on their symptoms to provide accurate accounts leading to timely help. Interventions should be offered from early childhood and more work is required to identify the most effective and cost-effective treatments. Study registration This study is registered as Current Controlled Trials ISRCTN91411078. Funding This award was funded by the National Institute for Health and Care Research (NIHR) Programme Grants for Applied Research programme (NIHR award ref: RP-PG-1211-20016) and is published in full in Programme Grants for Applied Research; Vol. 13, No. 5. See the NIHR Funding and Awards website for further award information. |
|---|---|
| ISSN: | 2050-4330 |