Skip to main content
Table of Contents
Print

How do relapse or regression rates compare after ABA therapy versus alternative therapies for autism? 

Author: Hannah Smith, MSc | Reviewed by: Dr. Rebecca Fernandez, MBBS

Families often worry about what happens when intensive support ends and whether children might lose skills over time. According to the NHS Autism overview and NICE CG170 there is limited long term evidence on relapse or regression with any autism therapy and no single approach is recommended to prevent deterioration. 

Understanding the concept 

In research, relapse or regression would mean loss of skills, return of behaviours or a clear decline in day to day functioning after therapy stops. Most applied behaviour analysis or early intensive behavioural intervention studies focus on gains in IQ, adaptive behaviour or autism symptoms rather than relapse. Reviews of naturalistic developmental behavioural interventions such as the ESDM and the overview by Schreibman and colleagues also report maintenance more often than loss of gains. 

Evidence and impact 

Follow up work such as the Norwegian cohort reported by Eikeseth and colleagues found that children who had early intensive behavioural intervention initially made larger gains than eclectic treatment groups but by adolescence outcomes had largely converged, with many young people still experiencing significant autism related difficulties. A centre based follow up by Alhosaini and colleagues suggested that overall gains were maintained three years after an early intensive behavioural programme, although social skills scores fell on one scale. The NICE 2021 surveillance review notes stakeholder concerns about possible harms from ABA but reports that published studies do not provide clear evidence of harm, while evidence remains equivocal and focused on short term child outcomes. 

Practical support and approaches 

Alternative early interventions such as the ESDM and broader naturalistic developmental behavioural interventions reviewed by Sandbank and colleagues tend to show maintained or gently improving gains over a few years rather than regression, although samples are small and outcome measures narrow. The NHS Autism support for adults and NICE CG142 emphasise ongoing adjustments, communication support and practical help rather than relying on any single named therapy. The National Autistic Society highlights how supportive communication environments and family understanding can protect day to day functioning. 

Challenges and considerations 

Across both ABA and non ABA interventions relapse is rarely defined or measured directly. Most studies report average scores at follow up, which can hide individual patterns of burnout or loss of skills. The Project AIM meta analysis in The BMJ and the linked JAMA Pediatrics analysis suggest that early intervention effects are modest and often do not generalise strongly over time and they do not find a simple link between more hours of therapy and better long term outcomes. The National Autistic Society also draws attention to community concerns about intensive behavioural programmes and the lack of good quality data on possible adverse experiences. 

How services can help 

In the United Kingdom, guidance from NICE and the NHS focuses on multidisciplinary, person centred care, including communication support, help with routines and sensory needs, and support for parents and carers. Behavioural and psychological coaching programmes such as those being developed by Theara Change can offer structured ways to build coping skills, emotional regulation and everyday strategies but they sit alongside, rather than replace, clinical and educational services. 

Takeaway 

Current evidence does not show clear and robust relapse rates for ABA compared with other therapies and most studies describe maintenance or partial maintenance of gains rather than dramatic regression. At the same time there are major evidence gaps around long term, real world outcomes and potential adverse effects. For now guidance from the NHSNICE and the National Autistic Society stresses ongoing, individualised support and reasonable adjustments as the most reliable way to protect quality of life over time. 

If you or someone you support would benefit from early identification or structured autism guidance, visit Autism Detect, a UK-based platform offering professional assessment tools and evidence-informed support for autistic individuals and families. 

Hannah Smith, MSc
Author

Hannah Smith is a clinical psychologist with a Master’s in Clinical Psychology and over three years of experience in behaviour therapy, special education, and inclusive practices. She specialises in Applied Behavior Analysis (ABA), Cognitive Behavioural Therapy (CBT), and inclusive education strategies. Hannah has worked extensively with children and adults with Autism Spectrum Disorder (ASD), ADHD, Down syndrome, and intellectual disabilities, delivering evidence-based interventions to support development, mental health, and well-being.

All qualifications and professional experience stated above are authentic and verified by our editorial team. However, pseudonym and image likeness are used to protect the author's privacy. 

Dr. Rebecca Fernandez, MBBS
Reviewer

Dr. Rebecca Fernandez is a UK-trained physician with an MBBS and experience in general surgery, cardiology, internal medicine, gynecology, intensive care, and emergency medicine. She has managed critically ill patients, stabilised acute trauma cases, and provided comprehensive inpatient and outpatient care. In psychiatry, Dr. Fernandez has worked with psychotic, mood, anxiety, and substance use disorders, applying evidence-based approaches such as CBT, ACT, and mindfulness-based therapies. Her skills span patient assessment, treatment planning, and the integration of digital health solutions to support mental well-being.

All qualifications and professional experience stated above are authentic and verified by our editorial team. However, pseudonym and image likeness are used to protect the reviewer's privacy. 

Categories