Most clinicians moving into autism assessment work end up asking the same question fairly early on. The literature talks about the ADI-R and the ADOS-2 as the two "gold standard" instruments, training courses for both exist, and senior colleagues seem to use them slightly differently in different services. So which one should you actually learn first?
The honest answer is that it depends on what you assess, where you assess it, and how much time and money you can put into training in the next twelve months. But there is a sensible default, and there are clear reasons to choose one route over the other.
This guide breaks down what the ADI-R and the ADOS-2 actually do, where they overlap, where they differ, and how UK clinicians can think about sequencing their training. It is written for clinicians who are already qualified and looking to move into autism diagnostic work, rather than for general readers.
Short answer: For most UK clinicians, the ADI-R is the more practical first instrument to train in. It is interview-based, can be administered remotely, requires no materials kit, and the core training is a single full day. The ADOS-2 is essential alongside it for direct observation, but the training pathway is longer and the setup is more demanding. In well-resourced services, clinicians are trained in both.
Modern autism assessment in the UK is built on two complementary sources of information. The first is the developmental history. The second is current observable behaviour.
The ADI-R covers the first. The ADOS-2 covers the second. When used together they produce what the international research literature has, for over twenty years, described as the most reliable approach to autism diagnosis.
That phrase, "gold standard", gets used loosely. In practice it means three things. The instruments have published psychometric properties showing acceptable reliability and validity. They use standardised administration and coding rules that allow different clinicians to reach similar conclusions. And they are recommended in the international autism research community, including in NHS, academic, and tribunal contexts.
Neither tool diagnoses autism on its own. Both feed into clinical judgement made by a qualified clinician or multidisciplinary team, in line with NICE CG142 for adults and NICE CG128 for under 19s. The tools sharpen the judgement. They do not replace it.
The Autism Diagnostic Interview-Revised is a structured, investigator-based interview conducted with a parent or primary caregiver. It was developed by Catherine Lord, Michael Rutter, and Ann Le Couteur, and is published by Western Psychological Services.
A full ADI-R takes between 90 minutes and three hours. The clinician works through a defined set of questions about the individual's developmental history and current behaviour, focusing on three domains drawn from international diagnostic criteria.
Responses are coded against behavioural anchors, and an algorithm produces scores that the clinician interprets alongside the broader clinical picture. The output is not a yes or no answer. It is a structured account of the developmental history with clear evidence for or against an autism formulation.
The key practical features that matter when you are choosing where to start:
The Global ADHD Network ADI-R training course covers all three teaching domains for the instrument: foundations and psychometrics, administration and coding, and scoring and clinical interpretation. It is delivered live online over a full day, carries 8 CPD hours, and the certificate is submissible to the BPS, HCPC, NMC, GMC, and GPhC.
The Autism Diagnostic Observation Schedule, Second Edition, is a semi-structured, standardised assessment of communication, social interaction, play, and restricted and repetitive behaviour. Unlike the ADI-R, it is administered directly to the person being assessed.
The ADOS-2 uses a series of "presses". These are activities and conversational prompts designed to give the individual the opportunity to display particular behaviours. The clinician records and codes what they see using standardised criteria, then maps the codes to algorithm scores and comparison scores.
It is structured into five modules, selected based on the individual's age and expressive language:
A full ADOS-2 administration usually takes 40 to 60 minutes. Training is more involved than for the ADI-R. Most providers run a multi-day course followed by a period of supervised practice and inter-rater reliability checks, particularly for clinicians who want to use the tool in research or in high-stakes diagnostic contexts.
Global ADHD Network does not currently run a dedicated ADOS-2 training course. The standard UK routes for ADOS-2 training are through specialist providers and university-affiliated programmes. Many clinicians complete the ADI-R first, build experience with it, and then add ADOS-2 once they are working in a service that supports the observation setup.
The table makes the contrast look sharper than it is in practice. Most experienced clinicians use both, because each compensates for the other's blind spots. A caregiver interview without observation misses the current presentation. An observation without developmental history can miss masked autism or read social anxiety as social communication difficulty.
The ADI-R is at its strongest in cases where you need a thorough, structured developmental history and where the current presentation alone is not enough to reach a clinical conclusion.
This includes:
There are also cases where the ADI-R is less useful on its own. Adults without a reliable historical informant are an obvious example. In those situations, clinicians often combine the ADI-R with other adult-focused instruments, school records, employer feedback, or self-report measures such as the RAADS-R and AQ-10.
The ADOS-2 is at its strongest when you need standardised, real-time observation of how the individual currently communicates and interacts socially.
This includes:
The ADOS-2 is less useful when the practical conditions are not in place. If you cannot run the assessment in a quiet, controlled room with the right kit and a co-rater available where needed, the quality of the data drops quickly. This is one reason it is harder to roll out across remote or hybrid services.
In well-resourced NHS pathways and established private services, clinicians are typically trained in both. The reasons are clinical, not bureaucratic.
The ADI-R answers the question: across this person's life, what has their pattern of social communication and behaviour looked like? The ADOS-2 answers the question: in a standardised setting today, what do we observe? A confident diagnosis brings both answers together. When the two converge, diagnostic certainty is high. When they diverge, the clinician has a clear signal that more investigation is needed before reaching a conclusion.
This combined approach is consistent with the recommendations in the NICE guideline on autism in adults (CG142), which emphasises comprehensive assessment built on developmental history, current presentation, and consideration of differentials. It is also the standard expected in most peer-reviewed autism research and in services that publish assessment quality data.
In day-to-day practice, however, very few clinicians learn both tools at the same time. The realistic question is which one to start with, and how to sequence the rest of your training behind it.
There is no universally correct order, but there is a sensible default and a small number of cases where the default flips.
Start with the ADI-R if:
Consider starting with the ADOS-2 if:
For most UK clinicians moving into autism assessment from psychiatry, clinical psychology, nursing, paediatrics, or allied health, the ADI-R is the more practical entry point. The lower barrier to entry does not mean lower clinical value. It means a faster route to building real diagnostic experience.
The longer-term plan is usually the same regardless of where you start. Train in one, embed it in your practice, then add the other within twelve to twenty-four months.
For ADI-R training, the main UK route is a CPD-certified course delivered by a recognised provider. The Global ADHD Network ADI-R Training Course covers the three core teaching modules: foundations, administration and coding, and scoring and clinical interpretation. It runs as a full-day live online programme via Zoom, awards 8 CPD hours, and the certificate is submissible to all the major UK regulators and to insurance providers for credentialling.
If you are earlier in your autism assessment journey and want the wider context around qualifications, registration, and the realistic route into UK autism assessment work, the companion guide on how to become an autism assessor in the UK covers the full career pathway.
For clinicians who want to build a stronger clinical foundation before specialising in diagnostic interviews, the Autism Awareness Training course is the most appropriate starting point. It grounds practice in neurodiversity-affirming approaches and aligns with current NICE guidance.
If a significant portion of the patients you assess present with both autism and ADHD, the Co-Occurring Autism and ADHD course is worth adding. AuDHD presentations are common, frequently missed, and require specific attention to differential and combined assessment.
For ADOS-2 training, UK clinicians typically train through specialist providers, university-affiliated programmes, or in-house NHS pathways. ADOS-2 is not currently part of the Global ADHD Network catalogue.
Whichever order you choose, plan for supervised practice afterwards. Sitting in on assessments led by experienced colleagues, then leading your own cases under supervision, is what turns a training certificate into clinical competence.
Training in either tool is a real investment of time and money. A realistic snapshot for a UK clinician planning the next twelve months:
ADI-R
ADOS-2
Bodies such as the Royal College of Psychiatrists and the British Psychological Society recognise both as appropriate CPD for autism-related practice, and certificates are typically accepted for revalidation and insurance credentialling.
There is also a quieter cost worth flagging. New assessors are most likely to underperform in the first ten to twenty cases, regardless of which tool they have been trained in. Building in supervision time, slower clinics, and protected reflection in your first few months is not optional. It is part of the cost of doing this work well.
Neither is more important. They answer different clinical questions. The ADI-R covers the developmental history. The ADOS-2 covers current observable behaviour. The strongest diagnoses use both, with clinical judgement integrating the findings against DSM-5-TR or ICD-11 criteria.
In some cases, yes, particularly in adult assessments where direct observation is less informative than developmental history. UK guidance does not mandate that any single tool be used, but it does expect comprehensive assessment. Using the ADI-R alongside other evidence such as self-report measures, school or work history, and clinical interview is usually acceptable. A standalone ADI-R is rarely sufficient in high-stakes or medico-legal contexts.
Not on its own. The ADOS-2 captures current behaviour but does not establish developmental history, which is essential for an autism diagnosis under both DSM-5-TR and ICD-11. It needs to be paired with a developmental account, either through the ADI-R or through detailed clinical interview with the individual and an informant.
No. The two tools have coexisted as complementary instruments since the late 1990s. There is no indication in the published literature or in UK clinical guidance that one is being phased out in favour of the other.
The ADOS-2 is more demanding to learn to a clinically reliable standard. It requires familiarity with multiple modules, observation and coding skills, materials handling, and supervised practice. The ADI-R has a steeper learning curve in coding interview responses against behavioural anchors, but the practical setup is simpler and the training is more compact.
Not always, but a service that uses both will usually be seen as more thorough. Many UK private clinicians use the ADI-R alongside other instruments rather than the ADOS-2, particularly for adult assessments. The decision usually comes down to the population assessed, the available space and resources, and the standards required by the clinician's insurance and regulatory body.
Yes. The ADI-R is well suited to remote administration over secure video call, provided the caregiver consents and the environment supports a long, focused conversation. This is one of the key practical advantages it has over the ADOS-2, which is designed for in-person, controlled observation.
If you have read this far and you are still unsure which tool to train in first, the sensible default for most UK clinicians is to start with the ADI-R. It builds the core diagnostic interviewing and coding skills that transfer to almost every other autism instrument, it can be slotted into existing remote or hybrid services, and the time and cost commitment is manageable inside a single CPD cycle.
A realistic next-twelve-months plan looks like this:
The clinicians who build strong autism assessment practices are the ones who treat this as a layered, multi-year skill set rather than a one-off course. Pick a sensible starting point, embed it in your practice, and build outwards from there.
You can view the full ADI-R training course details and upcoming dates here, or browse the wider Global ADHD Network course catalogue to plan your full pathway. For the broader picture of qualifications, experience, and the career route into UK autism assessment work, see the companion guide on how to become an autism assessor in the UK.
%20(1).webp)
