Clinical reasoning – and cognitions

Possibly one of the most hotly discussed aspects of clinical reasoning and pain relates to thoughts and beliefs held by both people experiencing pain and the clinicians who work with them. It’s difficult to avoid reading papers about “pain education”, “catastrophising”, “maladaptive thinking”, but quite another to find a deeper analysis of when and why it might be useful to help people think differently about their pain, or to deal with their thoughts about their experience in a different way.

Cognition is defined by the APA Dictionary of Psychology as

1. all forms of knowing and awareness, such as perceiving, conceiving, remembering, reasoning, judging, imagining, and problem solving. Along with affect and conation, it is one of the three traditionally identified components of mind.

2. an individual percept, idea, memory, or the like. —cognitional adj. —cognitive adj.

Cognitions are arguable The Thing most accessible to ourselves and most distinctive about humans – indeed, we call ourselves “homo sapiens” or “wise man” possibly because we can recognise we have thoughts! Although, as you can see from the definition above, many aspects of cognition are not as readily available to consciousness as we might imagine.

From the early days of pain management, explanations about the biology of pain have been included. Indeed, since 1965 when Melzack and Wall introduced the Gate Control Theory, in which modulation and descending control were identified, clinicians working in pain management centres have actively included these aspects of pain biology as part of an attempt to help people with pain understand the distinction between hurting – and being harmed (see Bonica, 1993).

The purpose behind the original approaches to “explaining pain” were to provide a coherent explanation to people in pain as to the “benign” nature of their experience: in other words, by changing the understanding people held about their pain, people were more likely to willingly engage in rehabilitation – and this rehabilitation largely involved gradually increasing “up time” and reducing unhelpful positions or activity levels. Sound familiar? (see Moseley & Butler, 2015).

Of course, in the early days of pain management, specific relationships between thoughts and both automatic and volitional behaviour were unclear. What we know now is that if I wire someone up to a biofeedback machine, measuring say heart rate variability, respiration and skin conductance, and then I mention something related to the person’s appraisals of their pain – maybe “Oh this really hurts”, or “I don’t think I’ll sleep tonight with this pain” those parameters I’m measuring will fluctuate wildly. Typically, people will experience an increase of physiological arousal in response to thinking those kinds of thoughts. In turn, that elevated arousal can lead to an increased perception of pain – and increased attention to pain with difficulty taking attention off pain (see Lanzetta, Cartwright-Smith & Eleck, 1976; Crombez, Viane, Eccleston, Devuler & Goubert, 2013).

So, the relationship between what we think and both attention to pain and physiological response to those thoughts is reasonably well-established, such that if someone reports high levels of catastrophising, we can expect to find high levels of disability, and reports of higher levels of pain. So far, so good. BUT how do we integrate these findings into our clinical reasoning, especially if we’re not primarily psychologically-oriented in our treatments?

The answer has been to dish out “pain education” to everyone – giving an explanation of some of the biological underpinnings of our experience. But for some of our patients this isn’t useful, especially if they have already heard the “pain talk” – but it has only hit the head and not the heart.

As Wilbert Fordyce was known to say “Information is to behaviour change as spaghetti is to a brick”. In other words – it might hit the brick and cover it, but it doesn’t change the brick, and neither does it move the brick!

You see, cognitions are not just “thoughts”, nor thoughts we are consciously aware of. Cognitions include implicit understanding, attention, the “feeling of what it is like to” and so on. And as occupational therapists and educators have found over the years, experiential learning (learning by doing) is one of the most powerful forms of behaviour change available (Kolb, 2014). People learn by experiencing something different. This is why cognitive behavioural approaches such as Acceptance and Commitment Therapy (ACT) so strongly endorse experiential elements.

Rather than attempting to change someone’s head knowledge of pain=harm, it might be more useful to help them experience doing something different and help them explore and generate their own conclusions from the experience.

I think both occupational therapists and clinicians who provide opportunities for movements and experiences (such as massage therapists, physiotherapists, osteopaths, chiropractors, myotherapists etc) are in an ideal position to guide people through new experiences – and then help them explore those new experiences. Rather than telling people what to think or believe (especially amongst those folks who are unconvinced by “book learning”!) we’re in a good position to help them work out what’s going through their minds – and what it feels like to do something differently. Instead of convincing, we can help people ponder for themselves. This is the essence of graded exposure: going from “OMG I can’t do that!” to “Oh yeah, I can master this”. It’s the difference between reading about how to ride a bicycle – and actually getting on a bike to learn to ride.

I agree that cognitive processes are really important in understanding a person’s experience of pain. I think, though, we’ve focused on overt thoughts to the detriment of trying to understand other aspects of cognition. We need to spend some more time exploring attention and distraction from pain; memories and how these influence pain; and to examine some of the implicit features of our understanding – and instead of approaching changes to thinking/understanding via the hammer of information dumping, maybe we can ponder the opportunities that arise from helping people experience something different and new.



Bonica, J. J. (1993). Evolution and current status of pain programs. Journal of Pharmaceutical Care in Pain & Symptom Control, 1(2), 31-44. doi:10.1300/J088v01n02_03

Crombez, G., Viane, I., Eccleston, C., Devulder, J., & Goubert, L. (2013). Attention to pain and fear of pain in patients with chronic pain. Journal of Behavioral Medicine, 36(4), 371-378.
Kolb, D. A. (2014). Experiential learning: Experience as the source of learning and development (2nd Ed), Pearson Education: New Jersey.
Lanzetta, J. T., Cartwright-Smith, J., & Eleck, R. E. (1976). Effects of nonverbal dissimulation on emotional experience and autonomic arousal. Journal of Personality and Social Psychology, 33(3), 354.

Moseley, G. L., & Butler, D. S. (2015). Fifteen years of explaining pain: The past, present, and future. Journal of Pain, 16(9), 807-813. doi:10.1016/j.jpain.2015.05.005


Teamwork: Gaps or overlaps?

For many years now, interprofessional/multidisciplinary teams have been considered the best model for delivering pain management. This stems from studies conducted right back as far as J J Bonica in 1944 (Bonica, 1993), and originally referred to teams consisting of several medical specialties. Bonica later initiated a multidisciplinary/interdisciplinary pain programme in 1960, including 20 people from 14 medical specialties “and other health professions”. In 1977, Bonica and Butler classified pain programmes into five groups – major comprehensive multidisciplinary programmes – more than six disciplines and involved in education and research; comprehensive multidisciplinary – four to six disciplines and involved in education and research; small multidisciplinary – 2 or 3 disciplines; syndrome-oriented specialising in single diagnoses; and modality-oriented using a single treatment. There were, at the time, 327 facilities around the world – including New Zealand (The Auckland Regional Pain Service).

Bonica didn’t comment on the team structure of these facilities, nor on the mix of “other health professions” involved. There has been a significant reduction in the numbers of comprehensive pain management centres, particularly in North America since the 1990’s. Fragmented, unidimensional treatment seems to be far more common than integrated multidimensional approaches.

Why might teamwork and structure of teams be important in pain management?

I like this discussion of why interprofessional/interdisciplinary teams might be more effective in pain management than multidisciplinary: “Multidisciplinary teams are unable to develop a cohesive care plan as each team member uses his or her own expertise to develop individual care goals. In contrast, each team member in an interdisciplinary team build on each other’s expertise to achieve common, shared goals. Therefore, it is crucial to indicate that multidisciplinary teams work in a team; whereas, interdisciplinary teams engage in teamwork.”

The argument for interprofessional teams in pain management is that by drawing on a common model of pain, each profession can align their treatments to meet the person’s goals, using a common framework, language and broad principles. But, and it’s a big but, this model depends on mutual trust, respect and time spent together developing a common understanding of each team member’s contributions. This is not something in which many health professionals have much training. For a good discussion of ways to foster good dynamics, Youngwerth and Twaddle’s 2011 paper is a nice place to start.

Why write about this now?

I was prompted to write about this because of a set of questions I was posed by a group of clinicians from another profession. We ostensibly work in a team, under the ACC Pain Contracts which specify a “multidisciplinary” approach. The questions, however, reflected both a lack of knowledge about pain management group programmes, and a lack of respect for the clinical skills provided by the people who deliver the programme I’ve developed. And it’s not the first instance of such behaviour.

I rarely criticise New Zealand healthcare policy, at least not on the pages of this blog. In this instance, though, I think it’s time to point out some of the issues that are present in the way pain contracts are being delivered since late 2016.

For those who’re not aware, ACC is NZ’s only personal injury insurer, owned by the country, with no-fault, 24 hour cover. That means anyone who has an accidental injury in NZ firstly can’t sue, and secondly has their treatment and rehabilitation paid for. Like most personal injury insurance companies, ACC’s main problem is the burden of long-term claims where often the main issue preventing return to work and case closure is persistent pain. As a result, pain services have been provided under ACC rehabilitation policy under a “provider-funder split” model since 2000.

ACC contracts providers to deliver pain management services. These services were to involve a number of designated professions, and these professionals were to be at least two years post-graduation, and to have completed postgraduate education in pain and pain management. And no, I don’t think a weekend course counts as “postgraduate education”. Unfortunately, the remuneration under these contracts is incredibly low. Remuneration rates are pre-determined by ACC, so that occupational therapy and physiotherapy are given one hourly rate, psychologists have a higher rate, and medical practitioners have the highest rate of all. There’s no variation in rates to fund experienced clinicians, so everyone gets the same amount irrespective of skill level. There is little to no allowance for team meetings, and there’s no allowance for screening or reporting included in the funding for the group programme I’ve developed.

Aside from the low funding, there are other concerns for me. There has been no auditing of the providers delivering these services. As a result, large businesses naturally try to maximise profit, employing entry-level clinicians for the contracts. Incredibly challenging for new graduates who have had limited exposure to persistent pain and pain management, and often apply acute pain management principles to chronic conditions. And that risks prolonging disability and exacerbating distress of people needing help.

Secondly, because these are new contracts, with quite different requirements from earlier iterations, groups have had to recruit a great many clinicians. Some of those clinicians presumed, I think, that their professional qualification is sufficient to work with people who have persistent pain. Even if their training had no pain content. ACC considers professional registration to be quite sufficient to practice in this area. While some of these clinicians are very experienced – pain management is not simple, and it is specialised. I have heard of practitioners continuing to use gate control theory as their primary “pain education”. While it’s an advance on being told you have “somatic disorder”, it doesn’t exactly reflect modern pain concepts. Again, using outdated information risks prolonging disability and exacerbating distress in a group of vulnerable people.

Teams to deliver pain contracts were often assembled in haste. Processes of induction, continuing education, developing a common clinical model, knowledge of other professionals’ contributions have all suffered as a result. Multidisciplinary practice is the norm – as one person I know used to put it, it’s “serial monotherapy”. Decision-making processes haven’t been developed, and integrating a clinical model common to all – and therefore abolishing a hierarchical structure – has just not happened. Instead a hierarchical, patch-protecting, and disjointed model where professionals are pitted against one another to gain some kind of dominance is emerging. A far cry from a mutually-respectful, integrated, non-hierarchical interprofessional team environment that research suggests is best for delivering pain management (Gatchel, McGeary, McGeary & Lippe, 2014).

When high value, low cost treatments for persistent pain are under-funded, and when costly yet ineffective treatments such as surgery continue being delivered, it’s the people who most need help who are harmed. I suppose what’s even more concerning is that despite 1 in 5 NZers living with pain lasting more than three months, and ACC claimants representing a small proportion of those living with pain, there is no New Zealand strategy for chronic pain management. People on ACC are, in most ways, rather lucky despite the failings of this contracting system.

The pain contracts could have represented an opportunity for innovation and an expansion of understanding between professions, what has happened instead is a tendency to deliver formulaic, ritualised programmes with gaps and overlaps, as a result of underfunding, poor quality control and both ignorance and power play in some instances.

We used to be world leaders in pain management. We have failed to capitalise on our headstart.  We should do better. We must do better for people living with pain.


Bonica, J. J. (1993). Evolution and current status of pain programs. Journal of Pharmaceutical Care in Pain & Symptom Control, 1(2), 31-44. doi:10.1300/J088v01n02_03

Gatchel, R. J., McGeary, D. D., McGeary, C. A., & Lippe, B. (2014). Interdisciplinary chronic pain management: past, present, and future. American Psychologist, 69(2), 119.

Youngwerth, J., & Twaddle, M. (2011). Cultures of interdisciplinary teams: How to foster good dynamics. Journal of Palliative Medicine, 14(5), 650-654.

When philosophy and evidence collide: is an occupation-focused approach suitable in pain management?

I have often described myself as a renegade occupational therapist: I like statistics, I think experimental research is a good way to test hypotheses, I don’t make moccasins (though I occasionally wear them!), I’m happy reading research and figuring out how I can apply findings into my clinical practice.

Occupational therapy is a profession that continues to evolve. The origins of occupational therapy lie back in the “moral” model of treatment for mental illness when advocates found that giving people things to do helped them become well (mind you, some of the reasons for admission to a “mental asylum” were things like “wandering womb”, novel reading, laziness and “female disease” read it here on Snopes). As time passed, occupational therapy was a way to “occupy” troops recovering from war wounds, and later, tuberculosis. At various points, occupational therapists have tried to enclose practice within prevailing models: anatomical, biomechanical, neurological. And then the scope broadens and the profession returns to “occupation” and all it means. Out of this latest movement, and informing occupational therapy practice today is the idea of “occupational science” – this is the “basic science” examining the factors that underpin occupational therapy practice (Yerxa, 1990).

Unlike most “basic sciences”, occupational science draws on areas of knowledge including anthropology, sociology and political science; all social sciences that bring their own philosophical biases to understanding social phenomena. Occupational science is about “what people do in daily life” – those routines, rituals, practices, customs and daily doings that support us in our roles, shape our place in the social world, and help us form an understanding of who we are in the world. Things like how we go about getting up, the way we serve a meal, the way we dress ourselves, how we go from one place to another, the hobbies and fun things we do – all fundamental building blocks of daily life. Occupational therapy, therefore, informed by occupational science, is focused on helping people participate in daily life as fully and equitably as possible, irrespective of health status, gender, ethnicity, religious belief, age and so on.

With a focus on not only helping people participate in occupations, but also using occupation as therapy, it’s not surprising to find a plurality of approaches to treatment. I have seen art used to help people with persistent pain represent the impact of pain on their sense of self – and to celebrate changes that have happened as a result of pain management. I have seen gardening used to help people become stronger, more confident to move and to reconnect with a hobby they had given up because of pain. I have seen people begin new hobbies (geocaching anyone?) as part of occupational therapy. I have used excursions to the local shopping mall to help people regain confidence and reduce their fear of crowded places where they might get bumped. Graded exposure is also an approach occupational therapists use to help people generalise their emerging skills to approach feared movements instead of avoiding them.

What I hope I don’t see is a return to a compensatory model for persistent pain. You know what I mean here: using gadgets or aids to “make life easier” when a person is dealing with persistent pain. Things like a special long-handled tool so people can pick something up from the floor – fine in a short-term situation like immediately post hip arthroplasty, but not so much when the problem is longstanding fear and avoidance. A special vacuum-cleaner so the person doesn’t have to bend – it’s so much easier yes, but it doesn’t address the underlying problem which can be remedied.

Why is a compensatory model not so good for persistent pain management? Well, because in most instances, though not all, the reason a person isn’t doing a movement when they’re sore is not because they cannot – but because that movement increases or might increase pain, and no-one really wants to increase pain, yeah? By providing a gadget of some sort, or even working through a way to avoid that movement, occupational therapists who use this sort of approach are ignoring the strong evidence that this reinforces avoidance as a strategy for managing pain, doesn’t address the underlying fear, and risks prolonging and actually reinforcing ongoing disability. This approach is harmful.

Helping people do things that might hurt isn’t a very popular idea for some clinicians and a lot of people living with persistent pain. It feels at first glance, like a really nasty thing to do to someone. BUT graded exposure is an effective, occupationally-focused treatment for fear of movement and fear of pain (Lopez-de-Uralde-Villaneuva, Munos-Garcia, Gil-Martinez, Pardo-Montero, Munoz-Plata et al, 2016). Used within an acceptance and commitment therapy model, graded exposure becomes “committed action” that’s aligned to values – and engaging in valued occupations is exactly what occupational therapy is all about.

Of course, not everyone enjoys this kind of work. That’s OK – because there are others who DO enjoy doing it! And it’s all in the way that it’s done – a framework of values, commitment, mindfulness and, that’s right, “chat therapy” – which some occupational therapists believe is right outside their scope of practice.

Now unless someone works in a vacuum, via some sort of mind-to-mind process, I cannot think of any therapist who doesn’t communicate with the person they’re working with. Humans communicate effortlessly and continuously. And “chat therapy” is about communicating – communicating skillfully, carefully selecting what to respond to and how, and focusing on clinical reasoning. Of course, if that’s ALL the treatment is about, then it’s not occupational therapy, but when it’s used in the aid of helping someone participate more fully in valued occupations using CBT, ACT, DBT or indeed motivational interviewing is one of the approaches occupational therapists can employ both within an occupation as therapy and occupation as outcome model.

I firmly believe that occupational therapists should follow an evidence base for their work. While I openly acknowledge the paucity of occupational therapy-specific research in persistent pain, particularly using occupation as therapy, there is plenty of research (carried out by other professions) to support approaches occupational therapists can adopt. After all, we already use developmental models, neurological models, sociological models, anthropological ones and yes, psychological ones. And that’s without venturing into the biomechanical ones! So it’s not an unfamiliar clinical reasoning strategy.

What makes occupational therapy practice in pain management absolutely unique are two things: a complete focus on reducing disability through enabling occupation, and a commitment to bringing skills developed “in clinic” outside into the daily lives and world of the people we are privileged to work with. What we should not do is focus on short-term outcomes like reducing (avoiding) bending with some new technique, while being ignorant of other occupational approaches. We are a fortunate profession because all of what we do is biopsychosocial, let’s not forget it.


López-de-Uralde-Villanueva, I., Muñoz-García, D., Gil-Martínez, A., Pardo-Montero, J., Muñoz-Plata, R., Angulo-Díaz-Parreño, S., . . . La Touche, R. (2016). A systematic review and meta-analysis on the effectiveness of graded activity and graded exposure for chronic nonspecific low back pain. Pain Medicine, 17(1), 172-188. doi:10.1111/pme.12882

Yerxa, E. J. (1990). An introduction to occupational science, a foundation for occupational therapy in the 21st century. Occup Ther Health Care, 6(4), 1-17. doi:10.1080/J003v06n04_04

Clinical reasoning in pain – emotions

The current definition of pain includes the words “unpleasant sensory and emotional experience” so we would be surprised if we encountered a person with pain who wasn’t feeling some sort of negative emotion, am I right? Yet… when we look at common pain assessments used for low back pain, items about emotions or worries are almost always included as indicators of negative outcomes (for example, STarTBack – Worrying thoughts have been going through my mind a lot of the time, I feel that my back pain is terrible and it’s never going to get any better, In general I have not enjoyed all the things I used to enjoy). And while the screening questionnaires have been validated, particularly for predictive validity (ie higher scores obtained on these measures are associated with poorer outcomes), I wonder how much we know, or think we know, about the relationship between emotions and pain. Perhaps its time for a quick review…

Firstly, let’s define emotions (seems easy!) “Emotions are multicomponent phenomena; (2) emotions are two-step processes involving emotion elicitation mechanisms that produce emotional responses; (3) emotions have relevant objects; and (4) emotions have a brief duration.” (Sander, 2013). There are thought to be six evolutionarily shaped basic emotions such as joy, fear, anger, sadness, disgust, and surprise (Ekman, 1992); but as usual there are complications to this because emotions are also examined in terms of their valence – negative or positive – and arousal (similar to intensity, but in terms of how much our physiology gets excited).

There are two main brain areas involved in processing both pain and unpleasant stimuli in general are the amygdala and the prefrontal cortex. These areas don’t exclusively deal with pain but with stimuli that are especially salient to people (remember last week’s post?), and researchers are still arguing over whether particular areas are responsible for certain emotions, or whether “emotions emerge when people make meaning out of sensory input from the body and from the world using knowledge of prior experience” based on basic psychological operations that are not specific to emotions (Lindquist et al., 2012, p. 129) . I’m quoting from an excellent book “The neuroscience of pain, stress and emotions” by Al, M. Absi, M.A. Flaten, and M. Rogers.

Now researchers have, for years, been interested in the effects of emotions on pain – there is an enormous body of literature but luckily some good reviews – see Bushnell et al., 2013; Roy, 2015 ;  Wiech and Tracey, 2009. What this research shows, essentially, is that pain is reduced by positive emotions, and increased by negative emotions. Now we need to be somewhat cautious about over-interpreting these results because they’re mainly conducted in experimental designs with acute experimental pain – people are shown pictures that elicit certain emotions, then poked or zapped, and asked to rate their pain (and their emotions, usually). It’s thought that the way these emotions influence pain is via our descending inhibitory pathways. Now the situation with real people experiencing pain that is not experimentally administered is probably slightly different – a lot more salient, a lot more worrying, and far less controlled. Nevertheless it’s worth knowing that when you’re feeling down, you’re likely to rate your pain more highly. If the emotion-eliciting stimuli are particularly arousing (ie they’re REALLY interesting) then the effect on pain ratings is greater. Experimenters also found pain reduces responses to pleasant stimuli, but there isn’t such a strong relationship with negative stimuli.

The valence (positiveness or negativeness – if that’s a word LOL) activates motivational systems either pleasant = appetitive, or unpleasant = defensive. Arousal or alertness gives us a clue as to how much motivation we have to either move towards or away from the stimulus. The degree of arousal affects our pain experience – so the more negative and angry we are, we rate our pain more highly; while the happier and jollier we are, we rate our pain as less intense. BUT, as for most things in pain, it’s complex – so once we get more than moderately angry/alert/aroused, the less we experience pain. The diagram below shows this kind of relationship – from the same book I quoted above (it’s worth getting!).

Does this mean we should freak people out so they experience less pain? Don’t be dumb! Being that alert is really exhausting. But what this diagram can explain is why some people, when they’re first attending therapy and are asked to do something out of the ordinary and just so slightly threatening (like lifting weights, or jumping on a treadmill) might report higher pain intensity – because we’ve caught them at the moderate arousal level where pain is facilitated.

Clinically, what this information means is that if we’re hoping to improve someone’s pain via pleasant or positive emotions, we’d better make sure they’re fairly high energy/arousing – a hilarious comedy perhaps – because lower intensity pleasure doesn’t affect pain much.

We should, at all costs, avoid eliciting fear and worry, or anger in the people we treat – because this increases pain intensity. This means giving people time to get used to our setting, what we’re asking them to do, and the intensity of whatever activity we’re going to do with them. In graded exposure, we should give people skills in mindfulness well before we begin doing the exposure component – because it’s likely to evoke higher than usual pain intensity if they can’t “be with” the increased anxiety that emerges during this kind of treatment.

And finally, if someone is experiencing anger, depression, sadness or anxiety – this is a normal psychological reaction integral to our experience of pain. It’s not necessarily pathological – though it probably increases the pain intensity the person reports.

I think we could promote far more scheduling pleasurable experiences as a routine part of therapy. What makes people smile, feel joy, have a good belly laugh? When was the last time they watched a comedy or joked with their family? Therapy can be fun, just see my friend Alice Hortop’s work on comedy as therapy (!



Ekman, P.  (1992). An argument for basic emotions. Cognition and Emotion, 6, 169–200.




Flaten, M. A. (2016). The neuroscience of pain, stress, and emotion : Psychological and clinical implications. In Al, M. Absi, M. A. Flaten, & M. Rogers (Eds.), Neuroscience of Pain, Stress, and Emotion: Amsterdam, Netherlands : Elsevier.

K.A. Lindquist, T.D. Wager, H. Kober, E. Bliss-Moreau, L.F. Barrett, (2012). The brain basis of emotion: a meta-analytic review. Behavioral and Brain Sciences, 35 (03),  121–143



Roy, M. (2015). Cerebral and spinal modulation of pain by emotions and attention. Pain, Emotion and Cognition, 35–52.


Sander, D. (2013). Models of emotion: the affective neuroscience approach. in J.L. Armony, P. Vuilleumier (Eds.), The Cambridge handbook of human affective neuroscience, Cambridge University Press, Cambridge, pp. 5–56




Clinical reasoning and why models of low back pain need to be integrated

Clinical reasoning has been defined as “the process by which a therapist interacts with a patient, collecting information, generating and testing hypotheses, and determining optimal diagnosis and treatment based on the information obtained.” (thanks to The model or lens through which we do these processes naturally has a major influence on our relationship with the person, the information we think is relevant, the hypotheses we develop, and ultimately the problems we identify and how we treat them. No arguments so far, yes?

So when we come to thinking about pain, particularly where a “diagnosis” can’t be readily established – or where the treatment doesn’t directly address a proposed causal factor – clinical reasoning should be led by some sort of model, but how explicit is our model, really? And, what’s more, how well does the research support our model, and the relationships between variables?

I’m thinking about my approach as an occupational therapist where my interest in assessment is to identify why this person is presenting in this way at this time, and what might be maintaining their current predicament; and my aim is to identify what can be done to reduce distress and disability, while promoting participation in daily occupations (activities, things that need to be done or the person wants to do). For many years now I’ve used a cognitive behavioural model first developed by Dr Tim Sharp who has now moved into Positive Psychology. His reformulation of the cognitive behavioural model works from the “experience” of pain through to responses to that experience, but incorporates some of the cyclical interactions between constructs. The model doesn’t include inputs to the “experience” component from the nociceptive system – but it could.

Many other models exist. Some of them are quite recent – the STarT Back Tool, for example, provides a very simplified screening approach to low back pain that some people have identified as a clinical reasoning model. Another is by Tousignant-Laflamme, Martel, Joshi & Cook (2017), and is a model aimed at pulling all the various approaches together – and does so with a beautifully-coloured diagram.


You knew there would be one! What I think these two models omit is to generate some relationships between the constructs, particularly the psychological ones. You see, while it’s a cyclical interaction, there are some relationships that we can identify.  And over the next few weeks I’ll be writing about some of the known associations, just to begin to build a picture of the relationships we can assess before we begin generating hypotheses.

For example, we know that the nervous system, and in particular our mind/brain, is never inactive and is therefore never a completely blank slate just waiting for information to come into it, but we also know there are relationships between the intensity/salience/novelty of a stimulus that attract attention, and that this competes with whatever cognitive set we have operating at the time (Legrain, Van Damme, Eccleston, Davis, Seminowicz & Crombez, 2009). So one relationship we need to assess is current contexts (and there are always many), and the times when a person is more or less aware of their pain.

Now, what increases the salience of a stimulus? For humans it’s all about meaning. We attribute meaning to even random patterns (ever seen dragons and horses in the clouds?!), so it’s unsurprising that as we experience something (or watch someone else experiencing something) we make meaning of it. And we generate meanings by relating concepts to other concepts – for a really good introduction to a very geeky subject, head here to read about relational frame theory. Relational frame theory is used to explain how we generate language and meanings by relating events with one another (The Bronnie translation! – for an easier version go here). Wicksell and Vowles (2015) describe this, and I’m going to quote it in full:

As described by relational frame theory, the theoretical framework underlying ACT, stimulus functions are continuously acquired via direct experiences, but also through their relations with other stimuli [5]. This implies that a behavioral response is not due to just one stimuli but rather the relational network of stimuli. Pain as an interoceptive stimulus is associated with a large number of other stimuli, and the actions taken depend on the psychological function(s) of that relational network of stimuli. A seemingly trivial situation may therefore elicit very strong reactions due to the associations being made: a relatively modest pain sensation from the neck trigger thoughts like “pain in the neck is bad,” which in turn are related to ideas such as “it may be a fragile disk,” and “something is terribly wrong,” that eventually lead to fatalistic conclusions like “I will end up in a wheelchair.” Thus, even if the initial stimulus is modest, it may activate a relational network of stimuli with very aversive psychological functions.

In other words, we develop these networks of meaning from the time we’re little until we die, and these mean any experience (situation, context, stimulus, event, action) holds meaning unique and particular to the individual. And these networks of meaning are constructed effortlessly and usually without any overt awareness. Each event/experience (yeah and the rest) then has further influence on how we experience any subsequent event/experience. So if you’ve learned that back pain is a Very Bad Thing, and you’ve done so since you were a kid because your Mother had back pain and took herself to the doctor and then stopped playing with you, you may have a very strong network of relationships built between low back pain, resting, healthcare, abandonment, sadness, anger, loneliness, fear, mother, father, pills, treatment – and the this goes on.

So when we’re beginning to construct a clinical reasoning model for something like low back pain we cannot exclude the “what does it mean” relationship. Every time someone experiences “ouch!” they’re processing a network of associations and relationships and behaviours that go on to influence their response to that experience – and affect attention to it and subsequent response to it.

Over 1000 words and I’ve not even started on emotions and pain!

Take home message: Even if we think we’re not addressing “psychological” stuff – we ARE. Omitting the “what does it mean to you?” and failing to factor that in to our clinical reasoning and subsequent treatment means we’re walking uphill on a scree slope. Oh, and telling someone they’re safe does not change those associations, especially if they’re longstanding. There’s more needed.


Legrain V, Damme SV, Eccleston C, Davis KD, Seminowicz DA, & Crombez G (2009). A neurocognitive model of attention to pain: behavioral and neuroimaging evidence. Pain, 144 (3), 230-2 PMID: 19376654

Sharp, T. J. (2001). Chronic pain: A reformulation of the cognitive-behavioural model. Behaviour Research and Therapy, 39(7), 787-800. doi:

Tousignant-Laflamme, Y., Martel, M. O., Joshi, A. B., & Cook, C. E. (2017). Rehabilitation management of low back pain – it’s time to pull it all together! Journal of Pain Research, 10, 2373-2385. doi:10.2147/JPR.S146485

Wicksell, R. K., & Vowles, K. E. (2015). The role and function of acceptance and commitment therapy and behavioral flexibility in pain management. Pain Management, 5(5), 319-322. doi:10.2217/pmt.15.32

Do pain management programmes really influence “doing” in daily life?

Disability and coping are two of the most important targets in persistent pain management, along with distress and pain intensity (the latter perhaps being the most difficult to influence). A question, however, is whether existing measures of disability truly capture the activities or occupations that people most value. For example, if house cleaning is just not my thing, even if my functioning improved over the course of a programme, would I choose to spend that new capability on vacuum-cleaning or out there in the garden?!

The Westhaven-Yale Multidimensional Pain Inventory (MPI)(Kerns, Turk & Rudy, 1985)  is a true workhorse of pain management measurement: It’s a measure that provides, via the subscales, an estimate of the complex inter-relationships involved in pain-related disability and distress. Three profiles of coping approach have been derived from the MPI indicating “Adaptive Coping” – the person is likely to continue to improve and engage in living well if they carry on as they are; “Interpersonally Distressed” – the person is experiencing difficulties with relationships and is feeling unsupported by others around them; and “Dysfunctional” – the person is struggling with their pain, and both disability and distress can feel overwhelming.

But the disability components of the MPI don’t really indicate the kinds of activities that people might be passionate about – or those that are relevant to their lives. The items are about general activities people “typically” do – in a North American setting. I know they’ve been validated in many different populations but they still represent a general idea of activity and function rather than the unique and idiosyncratic occupations that people do. (What's the difference? An occupation is the unique way I perform the important tasks of my day - the way I dash around the house while simultaneously cleaning my teeth, getting the dogs outside, making sure the curtains are opened and the appliances off just before I head out the door! We could call this "getting ready for work" but the way I do it, and the way you do it is probably very different!)

Why is that important? Well, because people value different occupations differently, and because values drive motivation, a broad measure of disability may not capture the true value of a pain management programme. And this is why the Canadian Occupational Performance Measure (COPM) (Law, Baptiste, McColl, Opzoomer, Polatajko & Pollock, 1990) is often used by occupational therapists to establish exactly what it is a person wants to be able to do. Goal Attainment Scaling is another option.

The study

In this study by Persson, Eklund, Lexell & Rivano-Fischer (2017), the aims were to assess longterm changes in MPI profiles over the course of a year; to look at associations between “Adaptive Coper” profiles at follow-up and improvements in occupational performance; and to look at factors at baseline that might be used to predict changes over time. Participants were, as usual, attending a tertiary pain management centre at a University hospital. They took part in a 21 day programme (oh the luxury of time!), and were seen by a team working together (oh the luxury of a co-located team!) including a psychologist, physician, physiotherapist, occupational therapist and social worker (oh for interprofessional teamwork!). The measures taken at pre and post were the MPI, COPM, and background demographic details.

Now here’s where the analysis gets pretty technical: dropout rates and pre-treatment differences amongst the MPI profiles were analysed using Matt-Whitney U, Kruskal-Wallis, chi-squared and independent t-test. Cross-tabs were used to describe the distribution of the MPI profiles. McNemar’s test was used to assess changes over time on each of the four profiles, and ultimately the participants were grouped into four profiles: never AC (Adaptive coper); losing AC; gaining AC, and stable AC, and these groupings were used to analyse associations between changes in MPI profile and changes on COPM from baseline to follow-up. And this, folks, is why occupational therapists NEED to learn statistical analyses! Because they went on to use univariate and multivariate logistic regressions to see whether demographic factors would predict these profiles at follow-up, incorporating satisfaction and performance scores on COPM.

What did they find?

More participants started the programme with a profile of dsyfunctional or interpersonally distressed than adaptive coper – and at both discharge and follow-up the adaptive coping group increased to be the largest subgroup. 20% of the participants ended up in the “gaining AC” group, 11% the “stable AC” group, 6% formed the “losing AC” group, and 63% were never in the adaptive coping group at all. Not so terrific in terms of broad outcome measures, but pretty consistent with other programmes around the world.

On the more individualised outcome measure of the COPM, people in the never AC group showed least clinically relevant improvement, both the stable AC and gaining AC groups showed more improvement on occupational performance and satisfaction with performance than the never AC group, while the stable AC group improved more than the losing AC group in terms of occupational performance changes.

Long-term changes showed those who started as AC were more likely to stay that way over time, and baseline scores for occupational performance were significantly associated with being AC at follow-up – in other words, people who felt OK about what they could do in terms of occupation, tended to become more capable and less disabled over time as measured by the MPI. More importantly, those people who had higher baseline scores on occupational satisfaction were 1.3 times more likely to shift to the AC group over time. “Clinically relevant improvements on performance and satisfaction with performance at 1-year follow-up, in occupations prioritized by the participants, were associated with having an AC profile at follow-up.” (p. 6).

Confidence in doing important occupations breeds confidence in other parts of living with persistent pain.

What does this all mean anyway?

Well, importantly, people who shifted from one of the other profiles to adaptive coping over a year showed clinically relevant improvements in occupational performance and satisfaction with performance – the authors suggest these findings show improvements on and satisfaction with “doing and performing” important occupations. Not being classified into adaptive coping at any time was associated with least improvement in occupational performance and satisfaction. There is an association between “doing/performing” and “coping/adapting” that hasn’t really been studied. Perhaps starting with things people enjoy instead of beating them over the head with things they’ve never been interested in could improve pain management outcomes. Pain management programmes DO influence doing in real life – if real life doing is used as therapy!

The authors also point out that “30–47% of participants who did not show an AC profile at follow-up still showed clinically relevant improvements on occupational performance and on
satisfaction with occupational performance.” (p. 8) This suggests that the MPI coping profiles only represent one part of the outcomes that are important to individuals: occupation being the individualised daily doings that people value very highly, and pointing to the need to explore individualised outcome measurement in more studies. Given that patient-centred pain management is based on individual goals, using only generic measures is likely to give misleading outcomes about treatment efficacy.


Kerns IVRD,TurkDC, Rudy TE. (1985) West Haven-Yale Multidimensional Pain Inventory (WHYMPI). Pain. 23:345–56

Law M, Baptiste S, McColl M, Opzoomer A, Polatajko H, Pollock N. (1990). The Canadian Occupational Performance Measure: an outcome measure for occupational therapy. Canadian Journal of Occupational Therapy. 57(2), pp82–7.

Persson, E., Eklund, M., Lexell, J., & Rivano-Fischer, M. (2017). Psychosocial coping profiles after pain rehabilitation: Associations with occupational performance and patient characteristics. Disability and Rehabilitation: An International, Multidisciplinary Journal, 39(3), 251-260.

Thinking the worst – and willingness to do things despite pain

Catastrophising, perhaps more than any other psychological construct, has received pretty negative press from people living with pain. It’s a construct that represents a tendency to “think the worst” when experiencing pain, and I can understand why people who are in the middle of a strong pain bout might reject any idea that their minds might be playing tricks on them. It’s hard to stand back from the immediacy of “OMG that really HURTS” especially when, habitually, many people who have pain try so hard to pretend that “yes everything is really all right”. At the same time, the evidence base for the contribution that habitually “thinking the worst” has on actually increasing the report of pain intensity, increasing difficulty coping, making it harder to access effective ways around the pain, and on the impact pain has on doing important things in life is strong (Quartana, Campbell & Edwards, 2009).

What then, could counter this tendency to feel like a possum in the headlights in the face of strong pain? In the study I’m discussing today, willingness to experience pain without trying to avoid or control that experience, aka “acceptance”, is examined, along with catastrophising and measures of disability. Craner, Sperry, Koball, Morrison and Gilliam (2017) recruited 249 adults who were seeking treatment at an interdisciplinary pain rehabilitation programme (at tertiary level), and examined a range of important variables pre and post treatment.  Participants in the programme were on average 50 years old, mainly married, and white (not a term we’d ever use in New Zealand!). They’d had pain for an average of 10.5 years, and slightly less than half were using opioids at the time of entry to the programme.

Occupational therapists administered the Canadian Occupational Performance Measure, an occupational therapist-administered, semi-structured interview designed to assess a person’s performance and satisfaction with their daily activities (Law, Baptiste, McColl, Opzoomer, Polatajko & Pollock, 1990). The performance scale was used in this study, along with the Chronic Pain Acceptance Questionnaire (one of my favourites – McCracken, Vowles & Eccelston, 2004); the Pain Catastrophising Scale (Sullivan, Bishop & Pivik, 1995), The Patient Health Questionnaire-9 (Kroenke, Spitzer & Williams, 2001); and The Westhaven-Yale Multidimensional Pain Inventory (Kerns, Turk & Rudy, 1985).

Now here’s where the fun begins, because there is some serious statistical analysis going on! Hierarchical multiple regression analyses is not for the faint-hearted – read the info about this approach by clicking the link. Essentially, it is a way to show if variables of your interest explain a statistically significant amount of variance in your Dependent Variable (DV) after accounting for all other variables. Or, in this study, what is the relationship between pain catastrophising, acceptance and pain severity – while controlling for age, gender, opioids use, and pain duration. The final step was to enter a calculation of the interaction between catastrophising and acceptance, and to enter this into the equation as the final step. A significant interaction suggests one of these two moderates the other – and this is ultimately captured by testing the slopes of the graphs. Complex? Yes – but a good way to analyse these complex relationships.


Unsurprisingly, pain catastrophising and acceptance do correlate – negatively. What this means is that the more a person thinks the worst about their pain, the less willing they are to do things that will increase their pain, or to do things while their pain is elevated. Makes sense, on the surface, but wait there’s more!

Pain catastrophizing was significantly (ps < .01) and positively correlated with greater perceived pain intensity, pain interference, distress due to pain, and depression – and negatively correlated with occupational therapist-rated functioning. Further analysis found that only pain catastrophising (not acceptance) was associated with pain severity, while both catastrophising and acceptance predicted negative effect (mood) using the WHYMPI, but when the analysis used the PHQ-9, both pain catastrophising and pain acceptance uniquely predicted depressive symptoms.  When pain interference was used as the dependent variable, pain acceptance uniquely predicted the amount of interference participants experienced, rather than catastrophising. The final analysis was using the performance subscale of the COPM, finding that pain acceptance was a predictor, while catastrophising was not.

What does all this actually mean?

Firstly, I found it interesting that values weren’t used as part of this investigation, because when people do daily activities, they do those they place value on, for some reason. For example, if we value other people’s opinions, we’re likely to dress up a bit, do the housework and maybe bake something if we have people come to visit. This study didn’t incorporate contexts of activity – the why question. I think that’s a limitation, however, examining values is not super easy, however it’s worth keeping this limitation in mind when thinking about the results.

The results suggest that when someone is willing to do something even if it increases pain, or while pain is elevated, this has an effect on their performance, disability, the interference they experience from pain, and their mood.

The results also suggest that catastrophising, while an important predictor of pain-related outcomes, is moderated by acceptance.

My question now is – what helps someone to be willing to do things even when their pain is high? if we analyse the CPAQ items, we find things like “I am getting on with the business of living no matter what my level of pain is.”;  “It’s not necessary for me to control my pain in order to handle my life well.”; and “My life is going well, even though I have chronic pain.”. These are important areas for clinicians to address during treatment. They’re about life – rather than pain. They’re about what makes life worth living. They’re about who are you, what does your life stand for, what makes you YOU, and what can you do despite pain. And these are important aspects of pain treatment: given none of us can claim a 100% success rate for pain reduction. Life is more than the absence of pain.



Craner, J. R., Sperry, J. A., Koball, A. M., Morrison, E. J., & Gilliam, W. P. (2017). Unique contributions of acceptance and catastrophizing on chronic pain adaptation. International Journal of Behavioral Medicine, 24(4), 542-551.

Kerns IVRD,TurkDC, Rudy TE. (1985) West Haven-Yale Multidimensional Pain Inventory (WHYMPI). Pain. 23:345–56.

Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. Journal of General Internal Medicine. 16(9), 606-13.

Law M, Baptiste S, McColl M, Opzoomer A, Polatajko H, Pollock N. (1990). The Canadian Occupational Performance Measure: an outcome measure for occupational therapy. Canadian Journal of Occupational Therapy. 57(2), pp82–7.

McCracken LM, Vowles KE, Eccleston C. (2004). Acceptance of chronic pain: component analysis and a revised assessment method. Pain. 107(1–2), pp159–66.

Quartana PJ, Campbell CM, Edwards RR. (2009) Pain catastrophizing: a critical review. Expert Reviews in Neurotherapy, 9, pp 745–58.

SullivanMLJ, Bishop SR, Pivik J. (1995). The Pain Catastrophizing Scale: development and validation. Psychological Assessment. 7:524–32.

One way of using a biopsychosocial framework in pain management – vi

I could write about a BPS (biopsychosocial) model in every single post, but it’s time for me to explore other things happening in the pain management world, so this is my last post in this series for a while. But it’s a doozy! And thanks to Eric Bowman for sharing an incredibly relevant paper just in time for this post…

One of the problems in pain management is that there are so many assessments carried out by the professionals seeing a person – but very little discussed about pulling this information together to create an overall picture of the person we’re seeing. And it’s this aspect I want to look at today.

My view is that a BPS approach provides us with an orientation towards the multiple factors involved in why this person is presenting in this way at this time (and what is maintaining their presentation), and by integrating the factors involved, we’re able to establish a way to reduce both distress and disability. A BPS approach is like a large-scale framework, and then, based on scientific studies that postulate mechanisms thought to be involved, a clinician or team can generate some useful hypotheses through abductive reasoning, begin testing these – and then arrive at a plausible set of explanations for the person’s situation. By doing so, multiple different options for treatment can be integrated so the person can begin to find their way out of the complex mess that pain and disability can bring.

The “mechanisms” involved range from the biological (yes, all that cellular, genetic, biomechanical, muscle/nerve/brain research that some people think is omitted from a BPS approach IS included!), to the psychological (all the attention, emotion, behavioural, cognitive material that has possibly become the hallmark of a BPS approach), and eventually, to the social (interactions with family, friends, community, healthcare, people in the workplace, the way legislation is written, insurers, cultural factors and so on). That’s one mess of stuff to evaluate!

We do have a framework already for a BPS approach: the ICF (or International Classification of Functioning, Disability and Health) provides one way of viewing what’s going on, although I can empathise with those who argue that it doesn’t provide a way to integrate these domains. I think that’s OK because, in pain and disability at least, we have research into each one of these domains although the social is still the most under-developed.

Tousignant-Laflamme, Martel, Joshi & Cook (2017) provide an approach to help structure the initial domains to explore – and a way to direct where attention needs to be paid to address both pain and disability.

What I like about this model (and I urge you to read the whole paper, please!) is that it triages the level of complexity and therefore the intervention needed without dividing the problem into “physical” and “psychosocial”. This is important because any contributing factor could be The One to most strongly influence outcome – and often an integrated approach is needed, rather than thinking “oh but the biological needs to be addressed separately”.

Another feature I like about this model is the attention paid to both pain and disability.

Beginning from the centre, each of the items in the area “A” is something that is either pretty common, and/or easily modified. So, for example, someone with low back pain that’s eased by flexion, maybe has some osteoarthritis, is feeling a bit demoralised and worries the pain is going to continue, has a job that’s not readily modified (and they’re not keen on returning) might need a physiotherapist to help work through movement patterns, some good information about pain to allay their worries, an occupational therapist to help with returning to work and sleeping, and maybe some medication if it helps.

If that same person has progressed to become quite slow to move and deconditioned, they’re experiencing allodynia and hyperalgesia, they have a history of migraine and irritable bowel, their sleep is pretty rotten, and they’re avoiding movements that “might” hurt – and their employer is pretty unhappy about them returning to work – then they may need a much more assertive approach, perhaps an intensive pain management programme, a review by a psychiatrist or psychologist, and probably some occupational therapy intervention at work plus a graded exposure to activities so they gain confidence despite pain persisting. Maybe they need medications to quieten the nervous system, perhaps some help with family relationships, and definitely the whole team must be on board with the same model of healthcare.

Some aspects are, I think, missing from this model. I’d like to see more attention paid to family and friends, social and leisure activities, and the person’s own values – because we know that values can be used to help a person be more willing to engage in things that are challenging. And I think the model is entirely deficits-based meaning the strengths a person brings to his or her situation aren’t incorporated.  Of course, too, this model hasn’t been tested in practice – and there are lots of gaps in terms of the measures that can be used to assess each of these domains. But as a heuristic or a template, this model seems to be practical, relatively simple to understand – and might stop us continuing to sub-type back pain on the basis of either psychosocial risk factors or not.

Clinicians pondering this model might now be wondering how to assess each of these domains – the paper provides some useful ideas, and if the framework gains traction, I think many others will add their tuppence-worth to it. I’m curious now to see how people who experience low back pain might view an assessment and management plan based on this: would it be acceptable? Does it help explain some of the difficulties people face? Would it be useful to people living with pain so they can explore the factors that are getting in the way of recovery?

Tousignant-Laflamme, Y., Martel, M. O., Joshi, A. B., & Cook, C. E. (2017). Rehabilitation management of low back pain – it’s time to pull it all together! Journal of Pain Research, 10, 2373-2385. doi:10.2147/JPR.S146485

One way of using a biopsychosocial framework in pain management – iv

And yes! There’s more to this series of posts on how I use a biopsychosocial model in practice!

Today’s post is about moving from a conceptual model to a practical model, or how we can use research in our clinical reasoning.

A biopsychosocial model (BPSM) as envisaged by Engel was a framework for clinicians to think about why this person is presenting in this way at this time (and what may be maintaining their situation), as well as what could be done to reduce distress and disability. Engel wanted clinicians to go beyond disease processes, isolated from the people experiencing them, and to explore aspects of how the person coped with everyday challenges (including health), the factors that influenced their decision that their health problem was indeed a problem, and the context of seeking healthcare.  He wanted clinicians to be scientific about how they generated hypotheses which could then be tested in clinical practice, and ultimately confirm or disconfirm the contribution of that factor.

The “bio” aspect of pain (which is a contentious word – I’ll comment in a bit) involves disease processes, trauma, all the biological aspects prior to conscious awareness of the “ouch” we know as pain. Theoretical developments in this area include all the work being conducted in terms of understanding anatomy and physiology of the human body, from molecular study (information transmission from one neurone to another); detailed understanding of spinal cord mechanisms; of the role of glia; of inflammatory processes; of genetic and epigenetic changes; of relationships between blood flow to and from various parts of the brain; of biomechanics; of normal healing processes – and so on. There’s no lack of information being generated by researchers undertaking basic science about the biological mechanisms involved in our experience of pain. Because I typically see people with persistent pain that has been present for maybe 12 months or more (usually much longer than that), I rely on the work of my colleagues to make a good diagnosis. Most people have had more investigation than is probably helpful for them, and I think we can use Clifford Woolf’s broad mechanisms as a reasonable stance when considering an underlying mechanism involved in a person’s pain. Essentially he identifies four main mechanisms: nociceptive, inflammatory, neuropathic and what is now known as “nociplastic” (where the nociceptive system appears to have a problem with processing information).

Yes, we can argue that our current state of understanding is incomplete and there is more to learn, but by working from these basic mechanisms I think we can begin to work on the “bio” part of a biopsychosocial model with a degree of confidence. For my work, anyway, these mechanisms seem to provide a reasonable framework from which the “bio” part of management can begin.

But this is where many clinicians start – and stop. Directly treating, for example, inflammation, certainly provides a reduction in pain – for example, my partner who takes Humera for his ankylosing spondylitis. He no longer experiences inflammatory pain and as his CRP levels reduced, so too did his pain. We can see similar effects when someone has a grotty old hip joint replaced, which removes nociceptive input, ultimately leaving them with a shiny new and painfree hip (in most cases). But as my partner found out, having no pain doesn’t immediately change old habits.

His situation is a nice illustration of the interaction between a disease process which responded really well to a drug that eliminates inflammation, and his beliefs and behaviour which wasn’t changed. Let me explain – once his drug kicked in and he had no pain, he found it odd not to have to think about his pain when climbing hills. It took him about a month or two to fully return to hill climbing in the way he’d done before his anky spond started. That’s right – no pain for a month or two, but that long before he felt confident to go about his activities. And he’s not a man who worries much about his pain!

To add some theory to this, his beliefs (that if he climbed hills a full speed he would inevitably end up with a very sore back) led to him having learned not to go a full pace (through both classical and operant conditioning). We could call this “pain-related fear and avoidance” – or “fear avoidance”. This is one theory that has been extensively researched, and we can integrate the hypotheses generated from this theory into our understanding of why my partner initially had some hesitation about climbing hills. Flowing on from this, we can consider treatments that have been found useful to address his hesitation.

The first treatment could be “explaining pain” to him. Now that wasn’t useful in this case because – oh yeah – his pain had gone! And although he knew his inflammatory pain wasn’t going to harm him (otherwise he’d never have been a high country fire fighter for 20 years despite his anky spond!), he didn’t like the after-effects of aggravating his pain. What helped was addressing his anxiety about the potential for a big flare-up – and this was primarily about beginning at a level that was just beyond his “normal” hill climb, and gradually progressing.

This superficially looks like “exercise” – but it’s exercise with a twist. My partner is as fit as a buck rat. His cardiovascular fitness was fine. Gradually increasing his hill walking wasn’t about increasing fitness – it was about helping him approach an activity that he was a tad concerned might flare his pain up, leading to a rotten night’s sleep (as it had in the past). In fact, this “treatment” was almost all about reducing avoidance by exposing him to things that increased his anxiety just a bit – enough for him to establish that the rotten sleep consequence didn’t happen.

So a biopsychosocial approach to his recovery involved the biological which quickly resolved his pain but left him with some concerns (reasonable ones I think) about pushing himself too hard. Addressing those concerns by taking a theory developed originally from phobia research, applying it to his situation and developing a treatment based on this theory, has led to his return to full participation. Using research-based information to address another part of “why is this person presenting in this way at this time, and what might be maintaining this situation” involves thinking beyond the disease process, and into understanding the problems the person identifies. It means thinking beyond a single discipline. It means reading widely and thinking creatively. That was a good part of Engel’s original proposition.


One way of using a biopsychosocial framework in pain management – iii

Before Christmas and the New Year break I was writing about how I use a biopsychosocial model in pain management – and I haven’t finished!

To review: The first post was about the context or the ideas behind Engel’s original model, and my two key clinical questions – why is this person presenting in this way at this time, and what can be done to reduce distress and disability?

The second post was about classical and operant conditioning and why these models are useful when we’re thinking about what a person does when they’re sore – and how their actions communicate to people around them. I also pointed out that many of these actions are not conscious, but have been learned and shaped from childhood, leading to a myriad of ways people express themselves when they’re in pain.

One of the criticisms of this approach to pain management is that “the model” isn’t scientific (therefore doesn’t lend itself to generating hypotheses that can be tested), and a second is that it’s “too fuzzy” and doesn’t specify what should be “in” and what should be “out” in clinical reasoning. I don’t agree with either of these statements and today I hope to present why.

Is a BPSM truly a “model”? What’s a model anyway? – one definition I’ve found is “In science, a model is a representation of an idea, an object or even a process or a system that is used to describe and explain phenomena that cannot be experienced directly.” In other words, they’re like a metaphor, bridging between something known and something abstract or unfamiliar. Models may be extremely detailed and mathematical, but may also be conceptual and broad. BPSM is probably the latter – a conceptual model from which we’ve developed some useful and testable hypotheses.

Engel himself described this as a scientific model, saying that it “enables the physician to extend application of the scientific method to aspects of everyday practice and patient care heretofore not deemed accessible to a scientific approach” (Engel, 1980, p.  535). He goes on to say that the doctor’s tasks are to find out what and how the patient is feeling, then to explanations (hypotheses) for the patient’s feelings and experiences, and then to test those hypotheses via clinical and laboratory studies (p. 536). Engel had some ideas about how a physician might generate hypotheses – based on his knowledge of general systems theory (von Bertalanffy, 1968). Engel appealed to von Bertalanffy’s idea that systems are a hierarchically arranged series of units, with the level of analysis dependent upon the complexity and unit of measure. In other words, molecular analysis is appropriate when cells and physiology are the unit of analysis, while the quality and influence of social connections are appropriate when looking at the influence of community and legislation on an individual’s behaviour.

The notion that a BPSM approach is “too fuzzy” and doesn’t provide structure is quite true: there’s not a lot of explanation as to how the various levels within a hierarchical system might interact. Interestingly, I think this problem is still relevant today!  While we know a lot about the brain, and a little about the mind – we don’t know how brain produces “mind”, although some philosophers and neuroscientists have taken a stab at it (thinking Andy Clark amongst others here). Similarly, although we know a lot about thoughts, beliefs and even relationships, we don’t know nearly as much about how thoughts and beliefs are adopted by a community, although Daniel Dennett has some thoughts about this.

So, it’s hardly surprising that when it comes to pain, we struggle to understand how biological processes, psychological processes and social ones interact to produce the experience of one person presenting for treatment at this time in this way – but this does not mean we should ignore what we DO know, which is that within each domain there is much to explore!

My preference is to draw on Brian Haig‘s idea of an Abductive Theory of Scientific Method, and in particular this paper on clinical reasoning, scientific method and abductive reasoning. The basic idea is that we recognise the existence of a phenomenon because either we’ve seen it before, or we’ve read about it. We distinguish between random events and a phenomenon because these appear to be consistent and repeated. Then we begin to generate some hypotheses to explain the presence of this phenomenon. Abduction is the process of studying what we see/observe/know (“facts”) and developing a theory to explain them (or generating a hypothesis). We then go about testing that hypothesis – and while we never truly confirm it, we can reject an alternative hypothesis tested against the first. As a result we arrive at what we can call a plausible explanation – something that “makes sense”, given what we’ve observed, and what we know about the world and how it works.

More about this geeky stuff next week. Meanwhile I think it’s worth pondering this: in “usual science” we somehow arrive at a hypothesis, and then set about testing it. No-one, it seems, knows where the original hypothesis comes from – and it’s rarely truly acknowledged. Researchers typically look for “gaps in the literature”, something that hasn’t been asked or answered yet, but what if you happen to be a clinician? I think clinicians routinely observe “interesting things” that, if we took some time to measure them, might be a phenomenon. Something that hasn’t yet been explained. I also think we have opportunities to be scientific about how we investigate what we see and do, if we’re prepared to be systematic and think about how we might control for confounds/bias. And I think those edges between levels within a hierarchy or between domains might be fruitful areas for clinicians to be exploring – bringing us to a practical application of Engel’s BPSM.

ENGEL, G. L. 1980. The clinical application of the biopsychosocial model. The American Journal of Psychiatry, 137, 535-544.

HAIG, B. D. 2008. Scientific method, abduction, and clinical reasoning. Journal of Clinical Psychology, 64, 1013-1018.