Wednesday, 20 August 2014

Youth mental health in the UK - remember the children with intellectual disability and autism

There is interest today in the apparently dire state of child and adolescent mental health services in the UK including a headline BBC News story.

The purpose of this brief blog is to encourage people entering the debate, and those who may join a "task force", to remember that within this topic there is a serious issue relating to mental health inequalities. Various disabilities and physical health conditions are associated with an increased chance that children will develop mental health problems. My focus in this blog is on children with intellectual disability (learning disability in UK health services language), and those with autism.

Our research using data from the ONS Child and Adolescent Mental Health surveys of British children shows that children with intellectual disability are about three times more likely to have high levels of behavioural and emotional problems than children without intellectual disability or autism. Children with autism had even higher levels of mental health problems.

We also know that these mental health inequalities emerge early in life for children with intellectual disability and/or autism. Again using nationally representative population survey data from the UK, we found that these inequalities have emerged at least by the time children are five years of age (and probably much earlier).

Earlier this year, working with colleagues from various Universities we also published data on the mental health of children with autism who also had intellectual disability and attended specialist autism schools. Again, we found higher rates of mental health problems compared to what would be expected for British school age children generally. Also very worrying was that we asked parents (and teachers) whether these children had any contact with child and adolescent mental health services in the previous six months. Only 20-25% of those children who currently had likely mental health problems were reported as having any mental health services contact in the preceding six months.

These research studies are based on large population studies or otherwise large samples. The findings confirm a general pattern of significant mental health inequalites for children with intellectual disability and/or autism, the early emergence of these inequalities, and a lack of access to mental health services for one of the most vulnerable groups of UK children.

Any discussion about youth mental health in the UK must include questions about how best to support this vulnerable group of children, and also how to start with early intervention to help to prevent mental health problems.

Young people with intellectual disability and autism are not a side issue in this debate. One estimate, again from UK population data, is that 14% of ALL children with a mental health problem in the UK will have an intellectual disability.

So, don't ignore these children and young people!

Wednesday, 18 June 2014

Improving Access to Positive Behaviour Support (IAPBS)

I’ve been mulling over for a few months now the issue of what the solutions might be to the problem of moving people with a learning disability from Assessment and Treatment Units (ATUs) in the UK to community based services closer to home. I’ve also had several interesting conversations with people connected to the work relating to responses to Winterbourne View and more broadly with people working in the field of Positive Behaviour Support (PBS). Although I do take responsibility for the ideas presented in this blog, it is important to say that most ideas take shape in communities of people interested in and discussing topics together. Ideas, do not come from individuals having flashes of inspiration (however good or terrible these turn out to be).

I do think it is time, however, to share some views on the conversations I’ve had and what they mean. I’m also giving them a name – IAPBS (Improving Access to PBS). Will it catch on?!

The Problem(s)

The problems that people with learning disability and challenging behaviour currently experience in the UK are well-documented. Shocking stories about poor care and the lack of local high quality supports are not only personally devastating for those involved but the responses to these cases just seem wrong or “inadequate”. We have also seen numbers in ATUs rising – quite the opposite to the plan (and promise), which was to move people out. Research over the years already provided warnings about the services received by people with learning disability and challenging behaviour, and we recently synthesised this research systematically.

A core issue that strikes me, especially looking at the recent ATU data (which show large % reasons for remaining in ATUs as a “clinical decision”), is that challenging behaviour is a domain under strong medical control in the UK. I do think that existing research evidence highlights a core role for medically trained individuals as a part of a multi-disciplinary team, but challenging behaviour is primarily a social and psychological “problem” and not a disease of the mind or body.

Campaigners have been very good at pointing out the current problems in relation to challenging behaviour services in the UK, and I know many have been suggesting solutions too. I have heard some of these solutions. A worry I have is that these solutions are as yet not co-ordinated. In addition, many of them are not linked to a delivery mechanism that people outside of the learning disability field will understand. Therefore, I’m not going to present a fully formed and costed plan but I am going to suggest a model that may be worth exploring. This is also not a solution to where people can be housed in their local communities (which, I recognise, also needs to be thought through) – instead, it is about their access to high quality supports. In fact, I would argue that we must solve the issue of access to high quality supports BEFORE simply moving people back to their local areas.


My core suggestion is that either we extend the Improving Access to Psychological Therapies (IAPT) system in England to encompass PBS, or that we replicate the general ideas within IAPT to deliver a large scale PBS infrastruture. Scaling up is actually the key problem when it comes to making PBS available to the several thousand people in ATUs but also to the thousands more people with learning disability whose behaviour challenges (estimated to be 10-15% of the population of people with learning disability).  At present, there are such a wide variety of solutions to challenging behaviour and a lack of focus on the main evidence-based approaches. This means that people with a learning disability and their families face the proverbial post-code lottery. National (England, and/or UK) action is needed and IAPBS may be an answer.

Why IAPT as a model? I see two driving forces behind IAPT, and both have parallels with the current problems around challenging behaviour. First, mental health problems were recognised as not (just) about medical understanding and treatment. Second, to invest in new services an economic case had to be made. IAPT provides access to evidence-based psychological therapies now for a wide range of mental health problems and has been extended to services for children as well as adults. The invest-to-save case was built initially on projections about savings in the UK benefits budget if adults of working age with mental health problems could be supported to stay in work, to have less time off work, or indeed to get into work in the first place. Providing evidence-based psychological therapies was the solution.

IAPT also has other features that I think are directly relevant to our current situation in relation to learning disability and challenging behaviour. First, there is coherence in terms of what therapies are offered – evidence based psychological therapies. Not just any therapy, and much less of the (what I consider to be dangerous) approach of eclecticism. Second, the competencies needed to deliver these therapies and the more general therapeutic skills needed to work with people are very clearly defined. For example, there is a distinction between high intensity trained therapists (who do the more difficult stuff), and low intensity trained therapists. Third, because competencies are defined clearly training programmes train people to “graduate” with the agreed competencies. Training contracts are also placed with expert institutions (mainly Universities I think). Fourth, there is also a national system of data gathering used throughout IAPT services which enables local and national progress to be monitored/audited in addition to the benefits of using data to inform therapeutic decisions with individuals.

An Initial Vision for IAPBS

My question in this blog is pretty simple. Can we use the IAPT model and infrastructure to deliver improved access to PBS? Many people will no doubt point out problems with this suggestion. However, we need action fast. We also need to ensure that people with learning disability and their families receive the highest quality evidence-based support for challenging behaviour. This support is needed now. To me, this means moving forward with a large scale delivery model that is established, that commissioners and service providers already understand, and so that could be got moving quickly.

PBS is not quite the same as many psychological therapies delivered to individuals or groups. For example, it is more of a coherent framework, rooted in behavioural theory and methods, and based on evidence about why challenging behaviours occur. In addition, PBS in many cases is likely to take longer than 8-12 one hour sessions to deliver for an individual with challenging behaviour. PBS is also likely to be delivered by a group of professionals working together with families and with people with learning disability themselves. However, I am not suggesting that these aspects are lost. I am talking about a delivery system and some ideas for how we can move forward much more quickly than we seem to be.

Some of the key pieces that need to come together quickly to then deliver IAPBS, are:

1.       Agreement about the evidence base. NICE is currently developing guidance on challenging behaviour in people with learning disability. I cannot see, considering the research evidence out there, that the Guideline Development Group for this work will be able to conclude anything else than a recommendation to use PBS as the main “treatment” and model for services.

2.       Agreement about the definition of PBS. Work has been ongoing relating to this in the UK, and a series of papers published in a special issue of the International Journal of PBS in 2013. Importantly, as I keep saying (apologies for repeating this), these issues of PBS definition start with an underlying model of why challenging behaviour occurs. It all starts with “theory”.

3.       The competencies needed to deliver PBS. Again, work on this is currently in progress. Explicit in this work is the need to define core behavioural competencies and all the associated professional competencies and values that go with these to define PBS. In addition, competencies are clearly needed at different levels (from expert consultants and supervisors supporting and training other staff and also dealing with the most complex cases, to the competencies needed by front line support staff in various settings, and also competencies for family members).

4.       A large scale training programme. Based on a nationally agreed competencies framework, there need to be training contracts funded with recognised expert organisations (similar to those who have been used to delivering IAPT training contracts). Initially, such a training programme could be focused on up-skilling, refreshing, and bringing into line with a nationally agreed PBS model, those professionals who already “market” themselves as PBS experts. This could happen quickly with dedicated resources. Accreditation of training and competencies is important so that commissioners can move towards only commissioning services working within the nationally agreed IAPBS system. In later years, an army of newly trained people would be needed to support the delivery of IAPBS.

5.       Service design. In some ways similar to IAPT, a range of different services is needed to deliver a full set of supports to people with learning disability and their families. For example, we know we need early intervention/prevention models working directly with young children but also providing skills to their families. We also need ways of screening for/identifying problems as they start to emerge (at any time in a person’s life) and associated early response supports. Then there are various levels of supports needed when significant problems have emerged. These may involve short term work from expert PBS professionals who carry out a functional assessment, design a PBS plan, and monitor outcomes. At the extremes, there may be a small number of cases where individuals (and their families too?) need to move out to a special place for assessment, design of “treatment”, and then “discharge” back to their local community. This last point does suggest a role for ATUs, but much more locally based and as a part of a clear model for services. Most importantly, all services would be working to the same underlying model (PBS) and the same people with expertise would likely be involved in the delivery of several aspects of a local/regional service. A vision for joined up services (joined up by a working model, and by the same competent professionals) is something families have been asking for. If you want to think about this ATU point and how it might work locally IF linked to other services working with the same underlying PBS model, then take a look at this research paper from Iowa in the USA [you should be able to download the pdf for free]. It is rather too scientific to make my point perhaps, but you can get a feel for what is going on in the inpatient facility – and by the way, average length of stay in this facility was 10 days (range 4-15 days), and two thirds of “patients” had a 90% reduction in their challenging behaviour or more. That’s proof that assessment and treatment services can deliver high quality and effective supports to people with learning disability and challenging behaviour. Also, note that this is NOT a medical facility – in fact, there is no mention of medical doctors as a part of the team working there.

6.       Systems for data collection. Some data collection as a part of PBS will be bespoke to an individual and their situation (part of the point of an individualised functional assessment/analysis). However, like with IAPT, a national system of at least outcome measurement could be used to increase accountability in the system. Commissioning could also shift to include aspects of direct outcomes (e.g., improvements in quality of life and reductions in challenging behaviour) if such a data collection were to be implemented. I should also mention research at this point – research on the outcomes of IAPBS would be crucial to ensure further development and testing of innovation. Focused UK funding support for applied learning disability research would be so good to see.

I’m no economist, but I recognise that it will cost money initially to realise an IAPBS vision such as that outlined here. I can also see that considerable savings are possible. We spend an awful lot of money on ATU placements at present. What do we get in exchange for this? Imagine how much cheaper it would be to have average 10 day placements (Yes, still at very very high cost but maybe only for 10 days and not months or years) in ATUs. If we get prevention right, more children and young people with learning disability in particular should also be able to stay with their families in their local communities – better for cost, and more importantly for quality of life for all concerned.

Friday, 6 June 2014

Can we have early behavioural intervention for all children with disabilities in the UK who need it?

I’ve already written a blog recently about some thoughts emerging from the campaign ABA4All []. This new blog is partly in support of this campaign, and partly to build on arguments from previous blogs over the past couple of years.

ABA (Applied Behaviour Analysis) is a dirty “word”

Let’s face it, the term ABA carries an awful lot of baggage. People think it is an intervention/treatment in itself (rather than an applied science), think it is about supporting children with autism, and ABA is also associated with traumatic experiences for many professionals and policy makers. By traumatic experiences, I mean being subjected to sometimes fundamentalist proponents of ABA who insist that things must be done in particular ways because “that’s what the evidence says” and also argue that ABA “has the best evidence in the field of autism”. 

I think we should all be using more precise terms here. So, let’s stop talking about ABA at all. Instead, let’s use labels that clearly say something about the focus of any intervention/treatment approach. As an example, the approach that combines ABA understanding and methods with a strong focus on values and attention to the person’s context to “treat” challenging behaviours in people with developmental disabilities is called Positive Behaviour(al) Support (PBS). Everyone loves PBS. It sounds nice, and it carries little of the negative baggage of “ABA”. However, PBS is simply good ABA. An ABA treatment/intervention approach or package has been developed to make a positive difference to the lives of people with developmental disabilities. When we talk about evidence based treatment/intervention and best practice working with individuals whose behaviour challenges, we don’t say that they are receiving ABA. It would make little sense to do so, because “ABA” does not communicate the specificity that is PBS. We just say they are receiving a PBS intervention or service.

The same case applies to early intervention for children with significant disabilities (intellectual disability, any number of syndromes associated with intellectual disability, and autism). I’m suggesting that we use a term that makes clearer what the treatment/intervention is. My suggestion is Early Behavioural Intervention (EBI). This term is not new – we and others have used it in international peer review journals when reporting on research studies. EBI can be used to refer to high quality support using behavioural principles and methods (i.e., drawn from the science and practice of ABA) delivered explicitly with early intervention in mind.

EBI, I would argue, refers to a number of intervention/treatment approaches that have behavioural methods at their core but emphasise different aspects of teaching methodology and also different settings in which intervention/treatment takes place. A non-exhaustive list would include: the Lovaas method, Discrete Trial Training, “ABA”, Pivotal Response Training, Verbal Behaviour, Natural Environment Teaching/Training, the LEAP model (from the USA – in schools), CABAS schools, and the Early Start Denver Model.

Why not “Intensive”?

Many research outcome studies in the field of autism have used the term Early Intensive Behavioural Intervention. The intensive part really emerges from the earliest evaluation of an EBI approach in autism by Lovaas. Lovaas’ treatment study published in 1987 compared a 40 hours per week intensive EBI approach, to a lower intensity (10 hours per week) model, and a control group who did not receive either treatment/intervention. Now, the 40 hours per week group of children did much better than either of the other two groups and so the argument for Intensity was born.
There are, however, two problems with using the Lovaas study to support an argument that EBI MUST be intensive to be effective. First, intensity was not manipulated experimentally – children were not allocated randomly to one of the treatment conditions. So, the groups could have differed on some other variable that explained the difference in outcomes (or at least some of the difference). Second, the 10 hours per week group differed in other key ways to the 40 hours per week group. These included: some treatment procedures could not be delivered to the 10 hours group because of lack of resources/staffing time, more other interventions were also used alongside the 10 hours, and the children were slightly older.

A second argument is that very positive outcomes for children with autism and children with an intellectual disability have been seen in studies where lower numbers of hours have been delivered (e.g., 15 hours and fewer). A third argument “against” intensity is the results from our large scale analysis of hundreds of children with autism who received EBI as a part of research evaluation studies around the world [see]. When we looked at variables associated with outcomes for the children, higher intensity (number of hours per week) did predict better outcomes. However, intensity explained only a small amount of outcome. Something else (or some combination of other factors) explains an awful lot more!

The argument FOR intensity in early intervention is an international consensus about the fact that any early intervention for children with significant disabilities (not specific to EBI at all) ought to be intensive. The assumption is that a higher “dose” will lead to more cumulative learning and so better outcomes overall.

Coming back to the UK again, a further reason for arguing for EBI and not EIBI is that 40 hours per week is completely unrealistic. The more we push for this, and it clearly cannot be afforded now or probably ever, the more that the wider population of children with significant disabilities will lose out to the small number of children who are able to access this sort of intensive intervention.

Quality delivery by experts

It is not the case that just anyone can deliver high quality and effective EBI. All of the existing research evaluation studies (including all of the “treatments” I claimed are EBI –PRT, ESDM etc etc.), tested treatments/interventions delivered by highly trained and supervised individuals. A couple of hours of training generally in “ABA” or an eclectic mishmash of approaches is not the same as what has been delivered in research studies and evaluated as evidence-based.

In the UK, hundreds of students have now graduated from post-graduate University courses that teach the theory and practice of ABA. Universities delivering, or who have delivered, these courses include: Bangor, Swansea, Cardiff, Kent (Tizard Centre), South Wales, Queens University Belfast, and Ulster. Of course, individual practitioners need additional training in particular intervention approaches (like EBI) and to understand the context in which they will be working. They also need ongoing supervision (just like any health, social care, or education professional in the UK). However, the point is that trusted and respected UK higher education institutions already have training courses that can support the experts who will design and deliver EBI services. More training courses, run by these experts, are also probably needed in the UK for staff of various kinds who will do most of the hands-on delivery of EBI (e.g., early years staff, teaching assistants).

EBI can be delivered effectively by specially trained and supervised professionals. UK state funded Universities can be trusted to help with this task of workforce training and development.

EBI for every child with significant disabilities in the UK

Here’s where we return to the ABA4All campaign. A central pillar of ABA4All’s work is parent choice. Even if you do not believe that EBI works better than other approaches typically available to children with autism and with other disabilities, it is a respected evidence-based approach delivered by well-trained professionals in the UK. There is every reason to offer parents EBI, or to make sure they are able to choose EBI for their child and family. Not everyone will want to choose EBI and that’s fine. At the moment though, ABA4All argue that parents and children are explicitly denied EBI. I agree.

ABA4All also campaign on the basis that what I’m calling EBI should be a choice that is available universally in the UK, so for all children who may need this boost early in life. Again, I agree. I think this does have some implications. The main one is that 40 hours per week is not going to be the answer when it comes to universal choice/provision. In any case, the key argument I make above is for quality and not necessarily quantity. So, what might be affordable – what existing funding mechanisms might be extended to open up the choice for parents of being able to have EBI for their child? How can we make sure that enough people are trained, properly supervised, and properly quality-assured to make this happen? The fact that universal availability is a challenge – to scale-up from the current provision across the UK – is not a reason to do nothing.

I also suggest that we must campaign outside of the autism box. Yes, children with autism can benefit from EBI, we have plenty of data to show that, and many parents want EBI for their young children with autism. However, there is nothing really autism-special about EBI. With only very minor changes, the model works very well for children with intellectual/learning disabilities. All of those children and their parents need the choice of EBI to maximise their early life chances.

Parental choice and the relevance of behavioural methods for teaching skills to children with significant disabilities also extend beyond early intervention. Older children desperately need skilled behavioural practitioners supporting them at home, and in mainstream and special education settings. Let’s not forget these children and their families. It is important to start somewhere though, and I think I have to agree that winning a battle about EBI may be the best way to establish a beachhead. 

Good luck to ABA4All.

Tuesday, 20 May 2014

The social model of disability, diversity, ABA, and autism

I’ve just listened to a short radio piece on the use of Applied Behaviour Analysis approaches to help children with autism learn. The segment is from BBC Radio 4’s You and Yours programme. Classic BBC in that they like to see a balance of views presented. There is some anti-ABA and some pro-ABA approaches.

For those of you outside of the UK who may not be able to listen to this piece, you may well have heard similar media discussions in your own countries. Several parents (from the group ABA4All - tell their stories about how their children’s skills have changed significantly (they argue) as a direct result of the use of ABA approaches. The parents are seeking to challenge the NHS in the UK who, they argue, spends money on support for children with autism with less evidence than ABA approaches and so ABA ought to be made available via the UK NHS. Discussion with experts following this piece includes an attempt to clarify that ABA is NOT a treatment for autism (see also my blog on what ABA is - and that the approaches are very much focused on the needs of individual children. The critical perspective is a little unclear, because it is somewhat indirect. Luke Beardon ( argues that he prioritises a social model of disability rather than a medical one, and that no one treatment for autism will suit all children.

Interestingly, we also hear a statement from the UK National Institute for Care and Health Excellence saying they found no evidence for ABA when they reviewed treatments for children with autism. This is an incorrect statement in my view, and I have written about this before ( NICE actually do recommend several ABA approaches in their guidance. In addition, their methodology is so medical model focused that it misses the importance of how to consider evidence in support of complex (often more educational) interventions such as those based on ABA (

I’ll not write more about the evidence base for ABA approaches in autism since my views on that are available for anyone to read and to consider whether they agree with me. I do want to pick up on the issue of the social model and also the diversity in autism (hence, no one treatment fits all).

The social model and ABA

I don’t intend a detailed review of the social model of disability here, and I am no expert. However, my understanding is that the social model at its core is about the responses of other people/society being what disables people. A medical/disease model sees conditions as things to get rid of – just as physical illnesses are subjected to a cure process. 

We know that ABA approaches, by definition, are firmly grounded in the view that the context around someone and how they behave is where we look to understand why things are happening as they are. Similarly, the environment (the social context in particular – what other people do) is what is changed in ABA approaches to deliver learning and other positive outcomes for people. This is exactly the approach that is applied in working with children with autism. ABA professionals should look to what it is about the environment, especially the social and educational environment, that needs to change to adapt to the child with autism. Most obviously this is seen in the fact that in early intervention in particular the teaching approach is different – more structured. There is little more significant in a child’s environment than how they are supported or taught during school (and also of course at home).

Behaviour analysis is also not informed by a medical model. It doesn’t recognise psychiatric or other medical labels for “disorders” as helpful explanations. Given this, ABA professionals should not be talking about “curing autism” – although I recognise that some do. ABA approaches in autism are instead focused on helping children (and adults) lead full lives, achieving the things that are important to them (and to their families). It is not about taking away autism at all. In fact, this is often a criticism of ABA approaches – that either they do not treat autism (note the medical model language used here), or that they don’t measure the core features of autism as outcomes. On the latter point: Of course not! First, we’re not trying to get rid of autism. Second, getting rid of autism is also unlikely to be the priority of a person with autism or their family so these “core features” are unlikely to be targeted.

At the very core of things here, my argument is that ABA approaches are compatible with a social model of disability. ABA approaches are enabling, emphasise that the environment needs to change to accommodate and support those with disabilities, and are not about getting rid of a disabling condition.

Therefore, it makes no sense to criticise ABA approaches as being anti-social model of disability. It certainly makes no sense to then add to the argument against ABA using perspectives derived from a medical model (e.g., ABA doesn’t “treat” the autism or the underlying neurological differences).

Diversity in Autism and ABA

My comments above make it clear that ABA approaches fully respect diversity. In fact, I have argued before that ABA approaches are the ultimate person/child-centred approach. Everything is focused on the child and their needs. Understanding the individual child’s strengths and weaknesses, understanding their preferences, understanding how their autism affects them day-to-day, designing a bespoke programme of support/learning, and also measuring progress at the level of specifically recorded outcomes that are individual to the priorities for that child. This is what a good ABA “programme” for a child will be like. There may be some similar pieces in several children’s programmes (e.g., perhaps several of them will use PECS as a communication tool), but the overall programme/support around each child will be different.

Because ABA is not one specific “treatment” for autism, and flexes to incorporate the full range of diversity, I also do not understand a criticism of ABA approaches that starts with “one approach to autism is not going to suit all children”. Actually, ABA offers a coherent framework within which probably the vast majority of needs and diversity can be understood and responded to (in terms of individualised supports). I can’t think of any other approach that offers that flexibility, respect for diversity, and coherence. Perhaps someone can put me right on that if there is another similarly diverse model out there. 

There is no darkness behind ABA (it is not the one ring to rule them all and in the darkness bind them), just people trying to understand and work out how to best support children with autism and their families. It is time for rapprochement on the common ground that is a social understanding and a respect for diversity in autism.