Pathological Demand Avoidance: does it exist, and if so what is it?

In recent years in the UK there’s been an increasing amount of talk about something called pathological demand avoidance, or PDA. It’s described as a subtype of autism that children (and some adults) are increasingly being diagnosed with. The National Autistic Society has been promoting acceptance of it, there’s a PDA Society, training courses and parent support groups are being set up. Paediatric and CAMHS services are seeing increasing numbers of referrals requesting assessment for PDA.

This is perhaps surprising because (and you wouldn’t know this from scanning the NAS or PDA Society websites) it’s not a recognised condition. Neither the DSM-5 nor the ICD-10 (the two main classification systems for mental disorder) have anything to say about pathological demand avoidance. The NICE guidelines (which set out best practice for UK clinicians) don’t mention PDA in their guidelines for diagnosing autism either for children or adults. It’s virtually unkown outside the UK (and a few parts of Scandinavia and Australia), and the research evidence base for it is wafer-thin.

So, I decided to look into PDA. What is it, does it really exist, and if it does exist, is it really a form of autism?

Pathological demand avoidance is described as a subcategory of autism characterised by “an avoidance of the demands of everyday life”. The child may use a number of avoidance strategies, including socially manipulative strategies. This is perhaps surprising, because autism affects a person’s social communication and interaction. Autistic people tend not to be good at social manipulation. So how do we know what’s being described is autism?

A quick bit of history. The term pathological demand avoidance was originally coined in the 1980s by Elizabeth Newson, a clinical psychologist at Nottingham University. She suggested it as a proposed diagnostic category for certain children who had previously been diagnosed with “atypical autism”. For a few years it didn’t really gain much traction, although Newson had a paper published in 2003 in the Archives of Diseases in Childhood. The paper drew a sceptical response from M Elena Garralda, professor of child and adolescent psychiatry at Imperial College London. Garralda suggested that Newson’s proposed diagnostic criteria lacked specificity (the ability to distinguish a condition from other conditions). She also suggested that some of Newson’s case studies seemed more indicative of a child who has a coexisting condition alongside autism.

Among those who have been sceptical of PDA, coexisting conditions are often raised as an alternative explanation for the features described. It’s very common for a person with autism to have a coexisting condition; for example, ADHD, or an anxiety disorder. It’s not the only alternative explanation though. Some point out that not everyone with autism presents in the same way, or is necessarily affected by all the core symptoms equally. There’s a saying, “If you’ve met one person with autism, you’ve met one person with autism.” It’s suggested that what’s being called PDA is simply part of that already-known variation.

There’s a third counter-explanation, which I’ll be talking about more later: that in some cases the child may not simply have a coexisting condition alongside autism, but actually isn’t autistic at all, and another condition has been misdiagnosed as autism.

PDA only really seemed to gain serious traction after 2007, when another paper, The Distinctive Clinical and Educational Needs of Children with Pathological Demand Avoidance Syndrome: Guidelines for Good Practice, was published by Phil Christie, another clinical psychologist at the Elizabeth Newson Centre in Nottinghamshire. It was subsequently picked up by the National Autistic Society, who started holding annual conferences on the topic, and in the last couple of years interest has been quickly growing.

So, what is pathological demand avoidance? In looking at the diagnostic criteria, I’ll be referencing Christie’s paper above.

So, let’s start with the first diagnostic criterion.

1. Passive early history in the first year

As with the more accepted criteria for autistic spectrum disorder, there’s some suggestion of delays in early development, but it’s all rather vague. The child in their early years may have been viewed as “puzzling in some way but not abnormal”.

2. Continues to resist and avoid ordinary demands of life, with strategies of avoidance
being essentially socially manipulative

This criterion is a particularly key one. According to Christie,

Children seem under an extraordinary degree of pressure from ordinary everyday demands and expectations and attempt to avoid these to an ‘obsessive’ extent. A key feature is that the child has sufficient social understanding to be socially manipulative in their endeavours and will often adapt strategies to the person making the demand. Strategies may include distraction, giving excuses, delaying, arguing, suggesting alternatives and withdrawing into fantasy.

“Being essentially socially manipulative”? “Adapting strategies to the person making the demand”? “Withdrawing into fantasy”? None of these sound like the actions of an autistic child. Autism is defined by difficulties in communicating, in interacting with others, and in having rigid, black-and-white thinking which impact on the person’s ability to use creative imagination.

But then I suspect the supporters of PDA would say that’s the point: that it’s an atypical form of autism.

3. Surface sociability, but apparent lack of social identity

According to Christie, the child appears socially interested, and will make use of social niceties, but only at a “skin-deep” level. “Greater empathy than in Asperger’s syndrome is apparent but sometimes it seems at an intellectual, rather than at an emotional level.” So, the child is able to empathise, albeit in a rather intellectualised way. Again, this doesn’t sound like autism, which affects somebody’s ability to understand what another person is thinking.

4. Lability of mood, impulsive, led by need to control

Christie states, “The child may switch from one state to another very quickly (eg from contented to aggressive) in a way that parents describe as ‘like switching a light on and off’…The child seems driven by the need to be in charge and can change in an instant when this isn’t the case..”

At this point it starts to sound not only unlike autism, but it starts to sound like something else. It sounds like a child with an attachment disorder – a child who has difficulty maintaining and developing relationships, forming relationships with others and regulating their own emotions. These problems are most commonly caused by early exposure to abuse, neglect, social upheaval, trauma, poverty, or dysfunctional family relationships.

5. Comfortable in role play and pretend

Once again, this doesn’t sound like an autistic child, who generally aren’t comfortable with role play and pretend. The child may “mimic and take on roles of others”, for example pretending to be the teacher to other children, and, “At the extreme, some children seem to become a collection of roles and lose touch with reality.” This may not sound much like an autistic child, but it does sound like a deeply insecure child.

6. Language delay.

When I read Christie’s paper, here’s where alarm bells started to ring in my mind. “This seems as a result of passivity. There is often a striking and sudden degree of catch-up.”

Apart from autism, what else could cause language delay, seemingly due to passivity, with a “striking and sudden” catch-up?

How about a neglected child who’s just started nursery? They weren’t learning spoken language because they weren’t being exposed to it, and when they are exposed to language, suddenly they catch up.

After all, doesn’t the “striking and sudden catch-up” imply that the language delay was caused by something other than an inherent neurological deficit?

Pragmatics are not as disordered as in autism or Asperger’s syndrome with more fluent use of eye-contact (other than when avoiding demands) and conversational timing. Some pragmatic difficulties remain such as literality, understanding sarcasm and teasing.

Well, yes, a deeply insecure child will indeed struggle to respond appropriately to sarcasm or teasing.

7. Obsessive behaviour.

Christie gives an example of the sort of obsessive behaviour that might be seen in a child with PDA.

Tom, aged five…is very attached to a boy called Adam. He is only interested in emulating Adam’s work and often talks to him and ignores the teacher. He will only eat food if he thinks Adam is eating at the same time.

Again, this sounds more like the behaviour of a child with very insecure attachments, desperately trying to latch on to somebody, rather than an autistic child.

8. Neurological involvement

Finally some mention of neurological issues, that might differentiate autistic spectrum disorder from attachment disorder, but it’s rather vague and woolly.

Crawling is late or absent in more than half these children and other milestones can be delayed. Clumsiness and physical awkwardness is often seen, but Newson feels there is insufficient hard evidence as yet.

So, what we have is a proposed sub-category of autism, where the child can make eye contract, can engage in imaginative role play, can empathise with others and even manipulate them, can fantasise…so why is it autism?

There isn’t much by way of hard research evidence on PDA. One piece of research that does exist relates to the Diagnostic Interview for Social Communication Disorders, or DISCO for short. The DISCO is a diagnostic tool for autism – one of several that exist (others include the ADOS, ADI-R, 3di and AAA). 11 (previously 15) items in the DISCO have been identified as possible indicators of PDA. Although DISCO is by no means the only diagnostic tool available, it is (and this may be turning into a bit of a running theme) the one that happens to be promoted by the National Autistic Society.

The study is by Elizabeth O’Nions, Phil Christie and four others, testing the reliability of the 11 DISCO items they identified as PDA markers, among a sample of children who were being assessed for ASD using the DISCO. They concluded that these items were “pulling their weight” in identifying PDA among the children in the sample. However, the devil is in the detail.

Further limitations include the fact that cases not specifically suspected of social communication disorders were not included.

So, if the sample was purely of children already suspected of having ASD, what would happen if those exact same items were used to measure children who aren’t also believed to be on the autistic spectrum, but perhaps have other emotional and behavioural problems?

Here’s the 11 DISCO items for PDA.

• Lack of co-operation
• Apparently manipulative behaviour
• Awareness of own identity
• Behaviour in public places
• Difficulties with other people
• Repetitive acting out roles
• Fantasising, lying, cheating, stealing
• Inappropriate sociability (rapid, inexplicable changes from loving to aggression)
• Using age peers as mechanical aids, bossy and domineering
• Socially shocking behaviour
• Harassment of others

The hypothesis hasn’t been tested, but I suspect if you ran those 11 items past a sample of entirely neurotypical children with emotional and behavioural problems, you’d get positive scores left, right and centre.

O’Nions, Christie and their colleagues have also developed the EDA-Q, a 26 point questionnaire to be given to parents and teachers, to identify possible cases of PDA. Here’s the 26 points.

  • Obsessively resists and avoids ordinary demands and requests.
  • Complains about illness or physical incapacity when avoiding a request or demand.
  • Is driven by the need to be in charge.
  • Finds everyday pressures (e.g. having to go on a school trip/ visit dentist) intolerably stressful.
  • Tells other children how they should behave, but does not feel these rules apply to him/herself.
  • Mimics adult mannerisms and styles (e.g. uses phrases adopted from teacher/parent to tell other children off).
  • Has difficulty complying with demands unless they are carefully presented.
  • Takes on roles or characters (from TV/real life) and ‘acts them out’.
  • Shows little shame or embarrassment (e.g. might throw a tantrum in public and not be embarrassed).
  • Invents fantasy worlds or games and acts them out.
  • Good at getting round others and making them do as s/he wants.
  • Seems unaware of the differences between him/herself and authority figures (e.g. parents, teachers, police).
  • If pressurised to do something, s/he may have a ‘meltdown’ (e.g. scream, tantrum, hit or kick).
  • Likes to be told s/he has done a good job.
  • Mood changes very rapidly (e.g. switches from affectionate to angry in an instant).
  • Knows what to do or say to upset specific people.
  • Blames or targets a particular person.
  • Denies behaviour s/he has committed, even when caught red handed.
  • Seems as if s/he is distracted ‘from within’.
  • Makes an effort to maintain his/her reputation with peers.
  • Uses outrageous or shocking behaviour to get out of doing something.
  • Has bouts of extreme emotional responses to small events (e.g. crying/giggling, becoming furious).
  • Social interaction has to be on his or her own terms.
  • Prefers to interact with others in an adopted role, or communicate through props/toys.
  • Attempts to negotiate better terms with adults.
  • S/he was passive and difficult to engage as an infant.

Reading those 26 points, my thoughts are pretty much the same as for the 11 DISCO markers: if you handed that questionnaire to the parent or teacher of any child with emotional and behavioural problems, the likelihood is that there would be items that could be ticked off. It wouldn’t matter whether they were autistic or not.

All this feels very unsafe. If PDA were to become accepted, there could be a risk that an autistic child could have features of an entirely different problem (possibly abuse or trauma) mislabelled as a feature of their autism. Worse, a child who may have been abused or traumatised, and isn’t autistic at all, could be misdiagnosed as autistic.

Browsing around Intenet forums, there’s a lot of anger from parents who are going to doctors asking for a PDA assessment, only to be told it’s not a recognised disorder. A common refrain is, “It makes me so angry that these doctors are refusing to diagnose PDA when it’s recognised by the National Autistic Society.”

The trouble is, “recognised by the NAS” may sound impressive and official to the lay person, but it doesn’t mean anything at all to a doctor. The NAS aren’t the DSM-5, or the ICD-10, or NICE. They’re a campaign group. They simply don’t have any authority to declare what is or is not a bona fide condition.

At the moment PDA is a proposed condition that isn’t formally recognised, doesn’t have much of an evidence base, and looks a lot like other problems than the one it’s purported to be. It’s hard not to draw the conclusion that the National Autistic Society is behaving in a hugely irresponsible way.


9 thoughts on “Pathological Demand Avoidance: does it exist, and if so what is it?

  1. A primary indication for autism is a lack of the ‘theory of mind’ which doesn’t work in a description for this so called disorder. Manipulation relies on a developed ‘ToM.

  2. As a retired psychotherapist of approximately 30 years, I am pleased to read that there are questions about the diagnosis of PDA and the association with autism. While I have read extensively about autism/Asperger, it was not my area of expertise. I continue to believe that “PDA” is more associated with early trauma and maybe a physiological predisposition of some type at the most, but I suspect that there is more neglect/trauma involved in childhood than autism.

    • Quite frankly I find that mode of thinking insulting. Both myself & my daughter identify as having PDA & neither of us have experienced anything in the way of childhood trauma. I believe it is a genetic predisposition, a way of being that is inherent to who we are. That is not to say that therapy can’t help overcome the more challenging aspects of PDA. I accept it’s difficult for me to be objective in our case, but your comments also suggest that there is ‘something wrong with us’ rather than focussing on the many positives that come from being neurologically atypical.

      • Why would you feel insulted??? If I was ill I would try to explore ALL paths to get better. But maybe it is easier to blame other things than to take a look at self

      • Ill? Having Pathological Demand Avoidance is not an illness. It is innate and an example of a neurotypical difference. It isn’t wrong, it is simply different.

  3. Thank you for taking the time to look at this – I can see why the literature appears confusing and there are people out there who consider ‘PDA is just a fad’. Nevertheless, I find that professionals who have worked with PDA young people are very clear on the matter, and individuals who have the condition are certainly the most expert in their difficulties and experiences.

    I gather you don’t have so much personal experience, and in fact, as PDA is relatively rare, one of the difficulties is that even clinicians don’t see many cases and so it is taking time for understanding to spread.

    While your research appears to make sense on the surface, I would say it dismisses too easily the professionals at the NAS who are truly experts (in both clinical practice and research), and studied carefully before amending their terminology around PDA (and in doing so, they would be the first to say that terminology doesn’t equate to ‘diagnosis’).

    Much of the PDA literature identifies the differences between PDA and other forms of ASD, and because ASD is a given, doesn’t talk very much about the things in common. Sensory issues, passionate interests, underlying difficulties with theory of mind (which are clear) and so on are all very much part of the condition. The EDAq is recognised as relating to the avoidance dimension, so is not itself considered a diagnostic tool.

    While there are undoubtedly some overlaps in characteristics between those with PDA and those with attachment disorder (and other conditions), the clear ASD characteristics, along with the very different responses to management techniques mark it out as different.

    It is the need for this different type of support (which so very clearly helps) that is the reason that recognition is needed and being demanded by schools, speech and language therapists and so on, not just by parents (the nature of that recognition – the ‘label’ itself – isn’t especially important… ASD sub-type PDA, or ASD characterised by extreme dem avoidance is fine… it is the PROFILE that is important).

    You are absolutely right that there needs to be very careful diagnosis in what are often complex cases, ensuring that past history is taken into account and seeing individuals in different settings. Acceptance of the cluster of ASD traits that identify the PDA group actually helps to tease out difficulties, causes, and therefore suitable approaches.

    We are undoubtedly still in the early stages in our understanding of this condition – not unlike the consideration of Autism in the 50’s – but there are now a number of institutions conducting research which is already starting to provide useful additional information.

    Thank you again for your contribution- and I hope that’s a helpful explanation.

  4. My own experience is that our son started ‘to crash’ at age 16. He was eventually given an Asperger diagnosis. Following advice from professionals such as ‘be very honest’, ‘give lots of structure and have clear rules that all obey’ did not help us to help him at all if anything it made things worse. His CAHMS doctor gave him a PDA descriptor without prompting. It was the only diagnosis that fitted. All too late for us as a family, he had entered the care system by then, aged 17.

    He has a high IQ ( that enables him to ‘mask’ in many situations but severe problems around ‘theory of mind’. His anxiety levels are sky high if he is not in control of his environment and the people in it to an extreme extent. It is hugely debilitating. I understand your skepticism but I have no doubt that compassionate engagement will bring answers in the same way as Lorna Wing’s engagement brought widespread recognition of Asperger Syndrome. It is heartening that Judith Gould who collaborated with her, is one of the foremost experts in the area of PDA

  5. Not all autistic advocates and academics agree with PDA or support it. I myself (2017) and Damian Milton (2013) have written about our concerns on the topic. I argue it does not exist. Damian Milton works for NAS, while I suspect it is the non-autistic members of NAS such as Judith Gould who expose PDA. These replies support my observation that PDA supporters tend to have emotive arguments for supporting PDA, while the arguments used to support PDA do not stand up to critical thinking. The key problem for PDA it is based on 1970s understanding of autism, with different explanations being provided by recent parts of the autism literature from the autistic population and autistic academics. This recent and emerging information is often missing from the PDA debate, thereby causing the debate to lack context of the wider pressing issues of how the entire autistic population is poorly treated.

    NAS’s support is an issue; I suspect NAS only supports PDA because Judith Gould supports PDA. Lorna Wing originally did not support PDA, with is being deliberately left out of the triad of impairment. The scuttle butt in Nottingham suggests Elizabeth Newson in 1980 created PDA in response to Wing’s triad of impairment (1979) as there was a rivalry between the pair. If NAS went of Damian Milton’s opinion, PDA would not be supported. There is a good argument that NAS is abusing its position as the primary stakeholder in the UK autism community. Phil Dore is correct NAS is a campaign group, which focuses mainly on the autism advocacy (parents and professionals), not the autistic voice (autistic individuals ourselves).

    As Damian Milton points out just because some adjustments work for persons believed to PDA does not mean it exists, the key issue with autism is that most autism adjustments do not work for a lot of us. Our exclusion rates are higher than the SEND average, along with our employment rates being notoriously low.

    EDAQ is being used as a screening tool in recent PDA articles as a replacement for diagnoses due to the issues in diagnosing PDA. As Damian Milton also says PDA is discounted from the DSM5 by the American Psychological Association as it PDA counts as a false diagnosis of an attachment disorder. Critically reading recent PDA articles supports Damian Milton’s assertion of “who needs to control whom?” of a possible struggle for power in the PDA discourse.

    I hope this clears up some clears up some points on the discussion.

  6. Pingback: AN OPEN LETTER TO ICD11 PRODUCTION TEAM. – Rational Demand Avoidance

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s