Small (but *long*) rant on occupational therapy & science
Hands up the occupational therapists who like statistics!  Thought not.

What about those who read journal articles comparing one treatment with another?  Use a treatment model consistently – or use a set of hypotheses and test to see which ones apply to the person we’re working with? Hmmm….  Are occupational therapists ‘epistemiphobic[1]’? or simply naive about science.

I recently read an assignment on clinical reasoning written by a very competent and educated occupational therapy graduate.  In it the views of Kuhn, cited in Mattingly and Fleming were described where ‘clinical reasoning within a biomedical frame is like puzzle solving, in that a clearly identifiable correct answer exists (e.g., a pathology, a cluster of physiologic deficits), and the player’s task is to find that answer.’ The argument continued that ‘Occupational Therapists in pain management often find themselves reasoning their work in a scientific and biomedical framework due to the nature of their work environments. This unfortunately often discredits the more difficult unspoken things that are done by the Occupational Therapist to the more recognisable and clearly defined instances that medical terms provide.’

My problem with the statements in this are that it assumes that science ‘discredits’ the ‘difficult and unspoken’ things done by occupational therapists – that somehow, occupational therapy is too complex for ‘science’ to unravel and understand.  Firstly this ignores the complexity of scientific methods such as constructivism, and secondly it suggests that what occupational therapists do is so complicated and unrepresented that no-one truly understands it. I also disagree that the prevailing model within pain management is a biomedical one – it’s clearly accepted that a biopsychosocial model is the one that most of the researchers and clinicians now accept. (Head here for one of the most popular books on pain I’m not advertising OK?!)

I’m going to deal with the first point, that occupational therapists have a limited understanding of science, and as a result misinterpret the multiple ways in which society, culture and indeed health care, has been built on the scientific method.  Then I want to turn to the suggestion that what occupational therapists do within treatment is so incredibly complex and individualised that it can’t be subjected to the kind of systematic scrutiny that other health professions have used, with success.

In the history of health care over the past 100 or so years, there has been a swing from humanism[2] toward empiricist[3] epistemology[4] and more recently, back again towards a constructivist[5] view of what makes up a ‘healthy’ individual.

This has been paralleled within the development of occupational therapy – from the Adlerian view of humans as being essentially self-correcting and requiring simply a context and ‘activity’ within which to flourish (Adler, 1931), through to the reductionist biomedical view of humans as an assemblage of body parts that can be pieced together using artificial limbs and principles of biomechanics (for example, as described by Ice, 2005).  As a reversal of this, the constructivist point of view has focused on the lived experience of individuals, rejecting the idea that data that is grouped together can truly represent what is true for any one person (Toombs, 1995).

Today, it’s not always clear, but plurality in how people define and reach wellbeing, especially while self managing chronic health conditions such as diabetes, rheumatoid conditions and chronic pain, is much more broadly defined than ‘the absence of disease’.  The WHO definition of health was, in my training in the early 1980’s, held up as an ideal rather than a reality.  Today, possibly as a response to the rise of chronic health conditions, the outcomes of health interventions are measured as often in terms of ‘quality of life’ and ‘disability’ than in complete cure.   Incorporating a ‘strengths’ approach within mental health (e.g. Cowger, 1997), looking at resilience within positive psychology (Seligman, 2008), and looking at fostering supported self management (Bycroft & Tracey, 2006), are not just passing fancies – they all have a sizable literature, and one that not only draws upon empirical science including randomized controlled trials, but also builds upon ‘the lived experience’ of individuals in all their complexity.

Science can often get a bad rap in the media.  Members of the public criticize scientists for being equivocal about results, for crying wolf over what may be harmful in one study and helpful in another. It’s not surprising that faith in what science can offer in health is somewhat low.

But occupational therapists are not the ‘general public’. Occupational therapists work within a health system that, for the most part, relies upon considered opinion based on fact.  Cold hard facts, often in the form of cost effectiveness, mean that well-designed scientific studies are needed to demonstrate that treatments are worth paying for.  This makes sense to me – I pay taxes, and I attend for treatment.  I’d much rather know that the treatments my money is spent on are known to work, and for whom; and I’d really like to know that when I need treatment, the options I’m given have data supporting their use rather than that old paternalist attitude of ‘the doctor knows best’.

I think science is misunderstood by many occupational therapists.  It’s seen as something removed from the daily reality of clinical life because it involves strange things like ‘control groups’ and ‘statistics’ and experiments, things that clinicians in everyday practice really don’t often do.  Science might not have even been part of some occupational therapists schooling, either before professional training or even as part of undergraduate training.  Girls and women are found to be fearful of science and mathematics as they go through school (e.g. Stipek, & Granlinski, 1991), particularly higher levels of schooling.  If these attitudes and beliefs continue in senior occupational therapy professionals and educators, it’s likely that undergraduate and newly graduating therapists will remain unaware of the place of science as part of the development of knowledge about our world and cultures.

Maybe it’s time occupational therapists took a second look at ‘science’.

Science simply means the ‘systematic study’ of phenomena to identify how it works, whether it will happen again, how it happened, where it happened, how similar or dissimilar that phenomenon was to another one – and so on.  Most of us are familiar with the ‘hypothetic-deductive’ process of experimenting to find out whether treatment X influences outcome Y.  This represents only one approach to the scientific study of phenomena.  There is more than one way to ‘do science’.  Here’s a very recent discussion about a definition of science.

Most of what we accept as ‘known’ today has developed as the result of years of serendipitous and systematic discoveries.  Serendipitous because we can all probably think of at least one discovery that was simply ‘a happy accident’ – think of Fleming and penicillin.  Systematic because once that finding was made, a concerted effort was made to understand what had happened, and to find out how to do it again – and explore the properties that caused such excitement in the beginning!

The view of science that I take is based on combining both aspects of scientific discovery.  The first is to identify something unusual that occurs – and then to see if it happens again.  Once is simply an aberration, twice may be coincidence, while three or more times starts to look seriously interesting. An example in occupational therapy practice might be to notice that several people with chronic pain describe being very reluctant to do their grocery shopping.  While in one or two people, this might be simply an area to work through to help them achieve independence, if it occurs in many people, especially in people with quite different pain presentations and durations, it might point to something worth investigating.  In the language of science, this is called ‘a phenomenon’, and if it’s demonstrated to be present in many people of different pain presentations and in different cities, then it could be called a ‘robust’ phenomenon.

It’s this first ‘aha!’ experience that is one of the most creative and serendipitous aspects of science – but that creativity need not stop there.  First of all, it’s really important to clearly and fully describe the phenomenon – using any manner of methods to identify who demonstrates it, how it’s portrayed, what settings it appears in, whether it’s simply in large supermarkets or maybe also in small convenience stores, whether there are any relationships between this problem with grocery shopping and any other pain-related phenomenon such as pain-related anxiety and avoidance (fear-avoidance).

The second part is to systematically study to understand how and why that phenomenon occurs. This involves painstaking hypothesis generation and testing, and the hypothesis generation part involves asking multiple ‘why’ questions, testing out with many experiments whether the predictions of that explanation hold true.  It also means a lot of disappointment as a favourite explanation turns out to be inaccurate, or incorrect.   The ‘experiments’ can be good old hypothetico-deductive ones, or more broadly-generated constructivist ones.  And to do this careful examination of evidence for and against any particular explanation means controlling for outside influences and chance.  This is whe re the statistics and other horrid mathematical things can come in handy because they allow us to factor ‘in’ or ‘out’ the amount of variability and trust we want to put into our experiments.  But this approach isn’t the only way – we can also use qualitative methods such as intensive description to examine a phenomenon, but even a constructivist approach to science uses inference to draw conclusions from one situation to help us consider and hopefully understand another situation.

So, science is about being systematic, recognising that each of us views the world slightly differently – but underneath all that, there are some commonalities at some point that we can recognise and then apply to another, similar situation (or by comparing what is dissimilar, use that information to inform a completely different situation).

I was with a group of therapists recently, and mentioned the word ‘model’ to them – to hear them talk, models are irrelevant to the ‘real’ day-to-day work of an occupational therapist.  Who’s got time to do all that reasoning and thinking in the business of helping people do things differently?  Well, whether you are aware of a model or not, everything we do especially within a therapeutic context, relies on the model we use. Models are simply a framework that describes more or less ‘how things work’, or how concepts hang together.  The good thing about a model is how easily it can organise how we gather information and what information is thought to be relevant or irrelevant.  A model is more or less helpful, depending on how well it works to explain the phenomenon we see.  The more accurate a model is, the better it will explain how and why a phenomenon occurs.

When we’re being educated as therapists, we spend time learning about various models, and for the more theoretical amongst us, models bring to life the various ways we can learn to view a ‘problem’.  Occupational therapists are famed for being ‘problem solvers’ because we learn to view a ‘problem’ as something that can be solved or ‘got around’.  Initially it’s not easy to solve various problems, but as we gain experience, see more problems, learn what works and what doesn’t work, we become much more capable of seeing ways to solve problems.  But as we do this, the visibility of our reasoning can slip behind this wonderful thing called ‘intuition’.  Intuition simply means that ‘we don’t know how we do it, but we come up with the right answer’ – or as a psychology lecturer once said, ‘intuition is overlearned skill’. We’ve got to the point where we are unaware of all that background reasoning and those assumptions that we have made because we’ve done it so often.

To give an example: we don’t think much about how we walk, or at least most of us don’t.  By the time we’re adults, walking is ‘intuitive’ – we just do it.  But there was a time when we were learning to walk, and it took all of our attention and we made a lot of mistakes and only after a couple of years did we actually ‘get’ how to walk.  Similarly, for most of us, driving the car is now pretty simple – we ‘intuitively’ know when to change gear, when to apply brakes, when to turn the wheel – but we probably all remember what it was like to learn how to drive!

Why am I raving about this? Well, the models we use in occupational therapy (& indeed in any therapy) initially feel quite difficult, require a lot of thought and we need to concentrate to ensure we collect information on all the relevant areas in that model. The models are based on things we ‘know’ about the world – scientific studies that verify the relationships that we recognise within that model.  For example, if we’re using a biomechanical model, we base this on the physics of movement – lever arms, fulcrum, equal and opposite forces.  We rarely go back and re-test the physics because they’re ‘common sense’.  Ermm… maybe they are now – but there was a time when humans didn’t know this and had to learn it.  So when we’re faced with a problem that appears to be solved using a biomechanical solution, we ‘intuitively’ solve it – but our intuition is based on the accumulated wisdom of all those people who did set out and test the laws of physics.  We just quickly jump to that conclusion because we don’t need to check out those laws any more.

The problem is when we try to use a biomechanical model to solve a problem that isn’t fully understood using a biomechanical model.  A good example might be manual handling.  We have assumed that the problem with back pain is essentially that we lift using poor biomechanics, placing additional loads on structures that shouldn’t be placed under such strain.  The solution seems clear – let’s all learn how to lift using ‘proper’ biomechanics and all will be well.  Sorry – that doesn’t work.

The problem of sore backs is not fully explained by a biomechanical model, but because to many therapists it seems ‘common sense’, other solutions don’t get a look-in. It’s been the work of scientific endeavour that has expanded the range of factors that are known to influence back pain (and more especially disability from back pain) that has started to provide some explanations for how and why some people have trouble managing their pain.  And the solutions derived from this research are not ‘intuitive’ to many therapists yet.  They’re still being examined, and they still need expanding.  But if we’re scientifically illiterate, feel afraid of reading scientific articles, rely instead on ‘intuition’ and either don’t measure outcomes or rely on inaccurate measures of outcome, we might continue to believe the biomechanical model is accurate.

Is occupational therapy so complex that what it does can’t be examined using science? I don’t think so.  If it is, then we’re saying that social systems, cognition, emotions, politics, neuroanatomy and physiology and perception are all much more simple than making a cup of tea in the morning when we get up.   Hmmm.  I don’t think so.  It’s more about looking at aspects of that activity in the light of various models (scientifically validated/tested hypotheses) and coming up with an explanation that can explain the most of what we see, with the least number of assumptions.  And while that might be quite distant from the day to day work of a therapist drafting housing alterations or a therapist establishing whether a person is safe to leave the mental health services and manage alone, the work that both of those therapists does uses models – and relies on assumptions that those models inherit.  To ignore the reality that we simplify our worlds so we can get our heads around them and to call it something woolly like ‘intuition’ does the cognitive work we do a disservice.

Is there anything wrong with simplifying the world? No – psychologists have been doing it for at least a century! The cognitive behavioural model is good example of a simple model that explains part of an individual’s behaviour.  It doesn’t explain everything, but it does provide some useful assumptions that help us help another person make changes.  Put together with other models from within psychology, and others from other disciplines, it starts to explain something about how people ‘tick’.  Maybe that’s the best that we can do at this point in time to understand each other.

So… should occupational therapists know ‘science’ – oh yes! both quantitative and qualitative methods too.  And probably some philosophy, and certainly lots of theory-building, and probably more psychology and sociology and biomechanics and anatomy and physiology.  Why not?!! At the very least occupational therapists should have some idea that what they’re doing is not ‘simply intuition’ nor ‘common sense’, that they do hold assumptions about how the world works when they view a problem, and their solutions are based on those mental models that we’ve learned.  So science is integral to how we do what we do.  Let’s not be afraid of science, let’s not think it’s irrelevant – let’s honour it and celebrate it and get on and DO IT!


[1] Epistimi – Greek word for ‘science’ or ‘knowing’, phobos – Greek word for ‘fear’ – a term I coined!


[2] Humanism – humanism is defined as a doctrine which seeks to account for the specificity, uniqueness and dignity of human life (retrieved from 7 September 2009 at 9.34am).  Occupational therapy is considered a strongly humanist profession.

[3] Empiricism – this school of philosophy claims that what we experience through our sense is the ultimate source of all our concepts and knowledge.  In other words, unless someone can show us something that is able to be touched, examined or measured, it cannot be known.  By contrast, a rationalist viewpoint suggests that we already know about things, or we can deduce things from what we do know, and it’s a matter of simply uncovering these known things through logic and reasoning. (retrieved from 7 September at 9.43am)

[4] Epistemology – the branch of philosophy devoted to studying the nature, sources and limits of knowledge (retrieved from 7 September at 9.43am)

[5] Constructivism – Constructivism states that our knowledge about what the world is entirely dependent on the meaning we as individuals make of what we perceive within our cultural context. In health constructivists assert that to call a condition a disease is to make a judgment that someone in that condition is undergoing a specific kind of harm that we explain in terms of bodily processes. But the bodily processes are not objectively malfunctioning; rather, they are merely judged by us to be unusual or abnormal because they depart from some shared, usually culturally specific, conception of human nature. (retrieved from 7 September 2009 at 9.48am)


Bycroft, J., Tracey, J (2006). Self-management support: A win-win solution for the 21st century, NZ Family Physician, 33(4), 243-248

Cowger, C. (1997). Assessing client’s strengths: Assessment for client empowerment. In D. Saleebey (Ed.), The strengths perspective in social work practice (2nd ed., pp. 59-73). White Plains, NY: Longman.

Ice, GH., (2005). Biological anthropology and aging. Journal of Cross-Cultural Gerontology, 20(2) 87-90

Mattingly, C. & Fleming, M.H. (1994). Clinical reasoning: Forms of inquiry in a therapeutic practice. Philadelphia: F.A. Davis Co. U.S.A.

Seligman, M.(2008). Positive Health, Applied Psychology: An International Review, 57, 3–18

Stipek, D, Granlinski, H (1991) Gender Differences in Children’s Achievement-Related Beliefs and Emotional Responses to Success and Failure in Mathematics, Journal of Educational Psychology , 83 (3) pp. 361-71.

Toombs, S K., (1995) The lived experience of disability. Human Studies, 18(1) 9-23

Ice, G. (2005). Biological anthropology and aging Journal of Cross-Cultural Gerontology, 20 (2), 87-90 DOI: 10.1007/s10823-005-9084-6

Doing, being, and creating a myth

A couple of things have drawn my thoughts to this topic: the first is a post on the Salford University Occupational Therapy Blog called ‘Create your own destiny’ in which they ask how educational institutions should prepare new graduate occupational therapists for the Brave New World of health care in which we work. They suggest we look for opportunities to promote ‘thinking out of the box’ and working from a nonmedical model – but this poses the question of how, in doing this, occupational therapists will manage to still meet the needs of those within a medically-based health care system.

One of the respondents to this post made the point that ‘I believe that occupational therapy is about occupation, health and well being’. I replied with the thought that ‘occupational therapists identify ‘doing’ (occupational performance) as their core domain, and all their clinical efforts are focused on helping ensure that people can ‘do”. (more…)

Finally – truth and opinion

This is the last post in this mini-series on why I use science when deciding what interventions to use as a therapist.  As I did yesterday and the day before, I refer to William Palya’s book on research methods – it’s easy to read, available on the internet for free, and although it gives only one view of scientific method, it’s a good start.

After having discussed the first onus – which is to be ethical, and the second, which is to be pragmatic, the third is to use a method to help achieve one or both aims. There are two basic things you need to do:
(1) Demand truth and
(2) Have good understanding

So, what is truth?
Well this can get into murky waters – especially if you listen to the philosophers! But for practical purposes, we can assume that ‘truth is an accurate description of something that is real’. It’s process of building up evidence from many sources, at different times, in different places that describe the same thing, using the least number of assumptions or appeals to special factors that can’t be tested, and describing the majority of the thing under examination. We can use the word ‘phenomenon’ instead of ‘thing’, or ‘event’ or ‘factor’.

Empiricism is one way that is used to determine ‘truth’. Something that is empirical is observed – through technology, to be sure, but can in some way, correspond with something that exists in the real world. As Palya puts it ‘If we wish to claim that something we cannot experience is real then the burden is on us to prove it to a skeptical audience; that is only fair.’

The evidence needs to be reliable – that is, if you look at it more than once, it should be the same. It should also be the same if anyone looks at it.

There should be more than one source of evidence for the ‘thing’. Palya’s example may help – ‘The more evidence from the wider a variety of sources, the more believable. If the police find a finger print the same as yours at a murder scene, maybe it means you are guilty, maybe it doesn’t. However, if the police also find your wallet there, and the murder weapon in your house, and the tire tracks of your car at the murder scene, and the victim’s jewelry at your house, and your teeth marks on the victim’s throat, and a VCR tape of the murder with you in the starring role – well, then you’re in trouble.

You can’t be the only person to say it’s so – and the others that agree with you also need to hold to the same ideas about what is ‘true’ and ‘real’. ‘If several observers who abide by the “rules” of science all agree concerning an event then it is probably true. It is reliable, it is objective. If only one person observes something and others do not observe the same thing then it is subjective.’

The phenomenon needs to be carefully defined so we all agree on what it is and that it describes the essence of the phenomenon. ‘The concept of a horse is false if it includes the saddle or fails to include four legs; it is false if it includes speaking English or fails to include galloping.’
The implications of this aspect of ‘truth’ is that the words we use to describe need to include the critical or essential elements, while excluding those that are not essential. A good definition is unambiguous with respect to what is included and what is not.

The definition you provide must actually have an impact on something that can be measured – because if you can’t confirm that it affects something, it might as well not exist.
Things cannot be said to exist outside the impact they have on sensation (resulting measures) or the impact on other things (functional definition). If your idea of the correct concept of a thing exceeds its operational/functional definition, the burden of proof or burden of communication is on you to prove, explain, and communicate the difference.’

I leave the best summary again to Palya: ‘we start with the notion of empirical, reliable evidence with multiple converging support which is operationally/functionally defined and has consensual validation and ask what is beyond. If someone wants to offer something else as a “truth” it must be proven. Truth does not mean anything anybody wants it to mean. Anyone wanting to extend the meaning of truth to something beyond what science has already substantiated must explain to us what they are talking about.’

Some people say that because various ideas that were once strongly supported by scientists have been rebutted in recent years that there is no such thing as ‘truth’ and science is nothing more than a set of opinions that change all the time. (eg disc prolapses on MRI were once thought to indicate the source of back pain and therefore needed surgery, now they are thought to be incidental and possibly a ‘normal’ variant in many cases)

Some things do change over time – not because the ‘truth’ part changes, but because more information comes to hand that explains more, or explains more with fewer special assumptions, or has more robust support than the previous ‘truth’. This is, in part, why we describe ‘truths’ as theories – theories can be and should be continually tested and as a result, refined. If a theory cannot be tested – then it’s really a model and needs to be evaluated in terms of how useful it is. If it doesn’t help with making decisions that can be tested, then it’s not useful at all.

Whew!! That’s a lot of theory and philosophy of science!
I think though, that it’s really important that we, as therapists, work out why we use the interventions we do, and that we can point to a method that means we feel we can rely on the interventions – and that we really do understand what we mean by evidence and science. Otherwise we are only reciting by rote, or working by habit and convention rather than seeking to understand.

What’s understanding?
It means you can describe, predict, know how to influence, synthesise and explain what you are actually doing.

This is from Palya’s chapter – summarises it quite neatly I think!

Have a great weekend – it’s Friday here, and I’m about to look for a Friday Funny. Be back soon!