Science in practice

Why I don’t trust my clinical reasoning: and why this matters


“See someone experienced” I hear people with pain say. “They’ll know what’s wrong with you.”

Well, based on the research I’ve read, I wouldn’t be so sure. In fact, I’m certain my own clinical reasoning is biased, prone to errors that I don’t notice, and influenced by factors that most clinicians would be horrified to think they, too, were influenced by.

Let me give you a few to ponder:

I’m interested in women and pain – and there’s a lot of evidence showing that women’s pain doesn’t get the same kind of diagnostic and management attention as men. Now part of this is due to the inherent bias in research where experimental studies often rely on male rats, mice and undergraduates because they don’t have those pesky hormonal fluctuations each month. Even volunteering to take part in a pain study has been found to be biased – people who volunteer have been shown to be more risk-taking and more extraverted (Skinner, 1982) – though to be fair this is an old study!

But contextual factors such as gender, distress and even the supposed diagnosis do influence judgements about pain intensity (Bernardes & Lima, 2011) including potentially life-threatening chest pain (Keogh, Hamid, Hamid & Ellery, 2004). Gender bias has been identified in a large literature review of gender bias in healthcare and gendered norms towards people with chronic pain (Samulowitz, Gremyr, Eriksson & Hensing, 2018).

And if you have the misfortune to be judged to have low trustworthiness and you’re a woman, you’re more likely to be thought to have less pain and to be exaggerating your pain (Schafer, Prkachin, Kaseweter & Williams, 2016). Beware if you’re overweight and a woman because you’ll be likely judged as having less intense pain, the pain will be judged as less interfering, more exaggerated and less related to “medical” factors – women’s pain in particular is likely to be judged as “psychological” and given psychological therapy rather than other treatments (Miller, Allison, Trost, De Ruddere, Wheelis, Goubert & Hirsch, 2018).

The weird thing is that the clinicians involved in these studies were oblivious to their bias. And let’s not even go there with people of colour or so-called “minority” groups such as LGBTQI.

So as clinicians our initial impressions of a person can lead us astray – and I haven’t even started with the contribution experience has on clinical reasoning. Let me go there then!

Something that cognitive psychologists have explored for some years now, is the type of thinking that we draw on for clinical reasoning. System one is “fast reasoning” – where we rapidly, instinctively and emotionally make decisions on the fly. Kahneman (1982) first described these two processes and noted that fast thinking gets better with rehearsal and are helpful especially for skilled clinicians needing to make decisions in pressured contexts, and draw on “pattern recognition” – or to be precise, draw on deviation from a recognised pattern (Preisz, 2019). System two is “slow reasoning” where decisions are made in a considered way, are not influenced by emotional state, and can be thought of as “rational.” Slow thinking is most useful where the situation is complex, where decisions need to weigh multiple pieces of information, where the situation might be novel, or where, for persistent pain in particular, there are multiple disease processes occurring.

OK, so we should choose system two, right? Not so fast! System one is hard to switch from – it’s what underpins “intuition” or “hunches” – and it gets more entrenched the more experienced we are. According to Preisz (2019), system one “seeks to form a coherent, plausible story by relying on association, memories, pattern matching and assumption.”

Why is system one thinking not so great? Well, we’re human. We’re human in the way we respond to any reasoning situation – we anchor on the first and most “plausible” ideas, and these might be unrelated to the actual presentation we see. For example, if we’ve been reading a journal article on a new treatment and its indications, it’s amazing how many people will present with those exact same indications in the next week! This is availability bias or anchoring bias. We’re also inclined to believe our own patients and judgements are different from “those people” – especially “those people” who might respond best to clinical guidelines. This means that even in the face of clear-cut research showing the lack of effects of knee arthroscopy (Brignardello-Petersen, Guyatt, Buchbinder, Poolman et al, 2017) an orthopaedic surgeon I know argued that “we choose our patients very carefully” – essentially arguing that his patients are different, and this approach is the best one.

If experienced clinicians find it hard to “unstick” from old practice, or move quickly to “intuitive” reasoning (even if it’s called “pattern recognition”), and if we all find it hard to recognise when we’re biased, or even that we are biased, what on earth should we do? All us old hands should retire maybe? All follow algorithms and not use “clinical judgement”? Take the “human” out of clinical management and use AI?

Some of these things might work. There is evidence that algorithms and AI can offer effective and (perhaps) less biased diagnosis and management than our unaided human brain (Kadhim, 2018) but there are also studies showing that direct comparisons between decision aids and clinical judgement are rarely made, and those that have been carried out don’t show superior results (Schriger, Elder, & Cooper, 2017). But watch this space: AI is a rapidly developing area and I predict greater use of this over time.

The risk with decision aids is – garbage in, garbage out. If we look at existing research we can see that male, pale and potentially stale dominates: this doesn’t bode well for people of colour, for women, for the unique and idiosyncratic combination of diseases a person can have, or for untangling the impact of disease on the person – in other words, disability and illness.

So, to summarise. We are all biased, and it’s best to acknowledge this to ourselves upfront and personal. We can then turn to strategies that may reduce the biases. For me, the one I turn to most often is a case formulation, using information gathered from a semi-structured interview and a standard set of questionnaires. These have been developed a priori so my biases in information gathering are limited. By taking time to follow a case formulation, my thinking is slowed to that more deliberative system two. At least some of the biases I know I’m prone to are mitigated.

And yet, I know I am biased. That’s why I use a supervision relationship to help me identify those biases, to be challenged and to reflect.

Bernardes, S. F., & Lima, M. L. (2011, Dec). A contextual approach on sex-related biases in pain judgements: The moderator effects of evidence of pathology and patients’ distress cues on nurses’ judgements of chronic low-back pain. Psychology & Health, 26(12), 1642-1658.

Brignardello-Petersen, R., Guyatt, G. H., Buchbinder, R., Poolman, R. W., Schandelmaier, S., Chang, Y., Sadeghirad, B., Evaniew, N., & Vandvik, P. O. (2017, May 11). Knee arthroscopy versus conservative management in patients with degenerative knee disease: a systematic review. BMJ Open, 7(5), e016114. https://doi.org/10.1136/bmjopen-2017-016114

Kadhim, M. A. (2018). FNDSB: A fuzzy-neuro decision support system for back pain diagnosis. Cognitive Systems Research, 52, 691-700. https://doi.org/10.1016/j.cogsys.2018.08.021

Kahneman, D., Slovic, S. P., Slovic, P., & Tversky, A. (1982). Judgment under uncertainty: Heuristics and biases. Cambridge university press.

Keogh, E., Hamid, R., Hamid, S., & Ellery, D. (2004). Investigating the effect of anxiety sensitivity, gender and negative interpretative bias on the perception of chest pain. Pain, 111(1-2), 209-217.

Miller, M. M., Allison, A., Trost, Z., De Ruddere, L., Wheelis, T., Goubert, L., & Hirsh, A. T. (2018, Jan). Differential Effect of Patient Weight on Pain-Related Judgements About Male and Female Chronic Low Back Pain Patients. J Pain, 19(1), 57-66. https://doi.org/10.1016/j.jpain.2017.09.001

Preisz, A. (2019, Jun). Fast and slow thinking; and the problem of conflating clinical reasoning and ethical deliberation in acute decision-making. Journal of Paediatric Child Health, 55(6), 621-624. https://doi.org/10.1111/jpc.14447

Samulowitz, A., Gremyr, I., Eriksson, E., & Hensing, G. (2018). “Brave Men” and “Emotional Women”: A Theory-Guided Literature Review on Gender Bias in Health Care and Gendered Norms towards Patients with Chronic Pain. Pain Research and Management, 2018.

Schafer, G., Prkachin, K. M., Kaseweter, K. A., & Williams, A. C. (2016, Aug). Health care providers’ judgments in chronic pain: the influence of gender and trustworthiness. Pain, 157(8), 1618-1625. https://doi.org/10.1097/j.pain.0000000000000536

Schriger, D. L., Elder, J. W., & Cooper, R. J. (2017, Sep). Structured Clinical Decision Aids Are Seldom Compared With Subjective Physician Judgment, and Are Seldom Superior. Ann Emerg Med, 70(3), 338-344 e333. https://doi.org/10.1016/j.annemergmed.2016.12.004

Skinner, N. F. (1982, 1982/12/01). Personality characteristics of volunteers for painful experiments. Bulletin of the Psychonomic Society, 20(6), 299-300. https://doi.org/10.3758/BF03330107

Knowledge gaps for working together


Whenever we work with someone living with pain, we form a team. A team, by definition, is “a distinguishable set of two or more people who interact dynamically, interdependently, and adaptively towards a common and valued goal/objective/mission” (Salas et al., 1992). So while many clinicians work outside an interprofessional team, they are always working in a team consisting of at least the person with pain, and themselves.

There’s a good deal of research on teamwork, and a heap of references in pain management literature on the benefits and, indeed, the need, to work in a team for best outcomes (both in terms of effects for the person and in terms of cost-effectiveness). Gilliam and colleagues (2018) demonstrate that long-term outcomes are retained by participants attending an interdisciplinary pain rehabilitation programme, while Guildford and colleaguees (2018) also showed reductions in analgesic use during an interdisciplinary pain management programme. It’s not new news folks!

Teamwork is well-investigated in health, particularly interprofessional/interdisciplinary teamwork. Much of this research, however, is focused on nursing and medicine interactions, with rather less attention paid to allied health and nursing/medicine teamwork. This matters because while nursing and medicine are moving away from the old medical model, the professions probably represent the two most similar in terms of clinical models. And this matters because one thing that’s found to be important for good teamwork in health is having a shared mental model (for example – from operation room – Wilson, 2019).

All good so far – nothing new here, move along, right?

Hold it right there, folks.

You see, when we work together in a team, particularly for people with persistent pain, we often generate a heap of new information about the person we hope to help. In New Zealand, the person will have completed the ePPOC set of questionnaires, then there will probably have been some physical performance testing, maybe some basic ROM, and muscle testing, perhaps some daily life functioning tasks, certainly some more psychological questionnaires, if the person sees a medical practitioner, there will be the obligatory bloods, urine, perhaps imaging – you know what I mean! A heap of information that each clinician deems necessary and I haven’t yet gone into each clinician’s desire to “hear the story from the beginning again!”

What’s lacking in our research on teamwork in persistent pain is discussion about how we assemble this information so that we move from a multidisciplinary team – Multidisciplinary teams involve people from different health disciplines working alongside one another while using clinical models drawn from their own professional discipline (Körner, 2010) – to an interprofessional/interdisciplinary team – Interdisciplinary teams also involve people from different health disciplines working alongside one another but meet regularly to collaborate on treatment goals and priorities (Ruan & Kaye, 2016). There is limited hierarchy and considerable communication, cooperation and often overlap between team members (Körner, 2010).

Not only a lack of a shared mental model (because we all think our model is The Best), we also lack an understanding of team processes. How do we develop an effective way to communicate, to cooperate, to deal with conflict in an open and creative way, to coordinate our work so things happen at the right time, to be coached so that the team-as-a-whole moves in the same direction and new people coming to the team feel part of the culture? Not forgetting that teams work in an ever-changing context, and team membership changes over time, while the overall team culture is something that emerges from a team collective (Salas, et al., 2015).

Are pain rehabilitation teams different from teams working in older person’s health, or palliative care, or as part of a primary health team?

I suspect so, but I can’t find good research detailing how our pain teams are different. It’s like a black box of mystery (a bit like interprofessional pain management programmes – one murky black box out of which a person pops!)

I’m left with this feeling that because teams in pain management and rehabilitation have become scarce in most part of the US, and that this is where all the research funding lives, there’s not very much that we actually know. We don’t know who holds the positions of power – is it the medical practitioner? the psychologist? the physiotherapist? the occupational therapist? Who makes the call as to when it’s time to work with the person to move from pain reduction to living well alongside pain? Are the team members actually using a common model or are they really working in parallel? And how can a team be maintained over time – I’ve had the privilege of working in a very close-knit and effective team for some years, but I’ve seen that team become smaller, fragmented, more multidisciplinary than interprofessional, with limited attention to processes of induction, developing effective conflict management, and really becoming weakened.

There is one conclusion I can draw from the mountains of material I’ve been learning and it’s this: it’s impossible to put a bunch of clinicians together and call them a team without putting effort in to develop those processes I’ve listed above. And when was the last time you attended a CPD session on “how to work in a team?”

Gilliam, W. P., Craner, J. R., Cunningham, J. L., Evans, M. M., Luedtke, C. A., Morrison, E. J., Sperry, J. A., & Loukianova, L. L. (2018). Longitudinal Treatment Outcomes for an Interdisciplinary Pain Rehabilitation Program: Comparisons of Subjective and Objective Outcomes on the Basis of Opioid Use Status. J Pain, 19(6), 678-689. https://doi.org/10.1016/j.jpain.2018.02.010

Guildford, B. J., Daly-Eichenhardt, A., Hill, B., Sanderson, K., & McCracken, L. M. (2018). Analgesic reduction during an interdisciplinary pain management programme: treatment effects and processes of change. Br J Pain, 12(2), 72-86. https://doi.org/10.1177/2049463717734016

Körner, M. (2010). Interprofessional teamwork in medical rehabilitation: a comparison of multidisciplinary and interdisciplinary team approach. Clinical Rehabilitation, 24(8), 745-755. https://doi.org/10.1177/0269215510367538

Ruan, X., & Kaye, A. D. (2016). A Call for Saving Interdisciplinary Pain Management. J Orthop Sports Phys Ther, 46(12), 1021-1023. https://doi.org/10.2519/jospt.2016.0611

Salas, E., Dickinson, T. L., Converse, S. A., & Tannenbaum, S. I. (1992). Toward an understanding of team performance and training. In Teams: Their training and performance. (pp. 3-29). Ablex Publishing.

Salas, E., Shuffler, M. L., Thayer, A. L., Bedwell, W. L., & Lazzara, E. H. (2015). Understanding and Improving Teamwork in Organizations: A Scientifically Based Practical Guide. Human Resource Management, 54(4), 599-622. https://doi.org/10.1002/hrm.21628

Wilson, A. (2019). Creating and applying shared mental models in the operating room. Journal of Perioperative Nursing, 32(3), 33.

The stigma of being a woman in pain


Women, it is often thought, must be much tougher than men when it comes to dealing with pain – after all, don’t women have babies without anaesthetic? Don’t men faint at the sight of a needle?

Ummmm, not quite so fast. Now before I begin, in this post I’m referring to cis-gender females, and in the experiments, participants were selected on the basis that they believed that negative gender discrimination was a thing. And as I write this post, I want to be clear that sometimes we have to begin with a very simplified model before research can be conducted on a much more messy cohort – and that this doesn’t negate the incredibly harmful and known effects of gender discrimination, and trans/inter/queer experiences. I can only hope that by starting this kind of research, as a community we’ll begin to understand the terrible impact that stigma has on people.

This paper investigated whether stigma related to one’s identity influenced the perception of nociceptive stimulation. It’s written off the back of earlier research showing that when people are excluded socially, their experience of nociceptive stimulation was greater (ie people didn’t need as much stimulation for it to be perceived as painful) (Eisenberger, Jarcho, Lieberman & Naliboff, 2006). Other studies have shown that people with low back pain who perceive themselves as stigmatised reported greater pain intensity, and that stigmatisation is the main source of social consequences for this group of people (Zhang, Barreto & Doyle, 2020).

These researchers (Zhang, Zhang, Li, Hu, Kong & Su, 2021) conducted two experiments to test the hypothesis that stigmatised women would experience greater pain intensity with nociceptive stimulation.
The first experiment used tonic cold pain (cold pressor test) in participants who had already been selected because they believed they had been stigmatised as a woman, asked them to immerse their hand in icy cold water (1 degree C) for as long as they could (to a maximum of 3 minutes), then take part in a mock online job interview. Some of the participants were told that was the end of the study; another group were told they were successful in the interview; and a third group were told that “woman are generally not suitable candidates for these kinds of jobs”; and the final group were simply told “you didn’t get the job” with no reason given. The latter three groups then underwent another cold pressor test as before. And finally they were all debriefed.

The researchers found that those who were told “women are generally not suitable for this kind of work” did feel more stigmatised than the others, and not only reported more sensitivity to cold (threshold) but also showed lower tolerance to the pain experienced in the cold pressor test.

The second experiment involved women who were selected as above. This group of women were shown images downloaded from Google – one set was of content showing devaluing of women, while another set were control or neutral images. The authors used a heat stimulation this time, and randomly showed either neutral or stigmatising images just before the heat was applied. Participants rated the pain after each stimulation.

The results of this experiment showed that when participants were shown the stigmatising content, they reported higher pain intensity from the same nociceptive stimulation. In other words – stigma-inducing images led to these women reporting more pain when given the same amount of heat stimulus.

Not content with this, the researchers conducted a third experiment, this time examining nociceptive-evoked brain responses. They used the same experimental design as for the second experiment, but instead of self-reporting, participants had EEG signals recorded during each heat stimulation.

The results of this experiment once again showed that when participants were shown stigmatising images, they rated their pain experience more highly, and that this was reflected in the EEG results they obtained. N1 amplitude and P2 latency in time and LEP magnitude in the time-frequency domain were influenced by the stigmatising cues.

What does this all mean?

Well, for one thing it’s nice to see research being conducted in women (there’s a bit of a bias against women being involved in basic science pain research because of that pesky old hormone thing – see Samuloitz, Gremyr, Eriksson & Hensing (2018) for more). And for a study to have positive findings.

I’m particularly interested in the brain responses – simply by manipulating the sense of stigma, the same nociceptive stimulation was processed differently. Now this isn’t the same as saying “psychological factors cause pain” because this study is not looking at that – nociceptive stimulation was included – but the same nociceptive stimulation was prioritised in parts of the brain usually active in emotional responses, while P2 is an area involved in the “advanced stage of perceptual processing” was activated sooner in the stigmatised manipulation than in the control condition. The authors argue that because stigma is a threat to sense of self, and because this sense of threat can lead to vigilance about potentially stigmatising cues, greater attentional processing is allocated to threat information, and this in turn, enhances the experience of pain. The greater N1 amplitude demonstrate that attention was drawn to stigmatising material and then influenced the subsequent nociceptive information.

Let’s take a moment to consider the implications of this. Many women have reported their feelings of being devalued both because of their gender as well as their reports of pain. Women may be told “there’s no cure for being a woman” and given inadequate pain relief for period pain (true story). Women do report more pain, are more likely to develop persistent pain, and seek help for pain more readily than men. The latter can be seen as a bad thing – shouldn’t we just “cope”?

Implicit attitudes towards women remain throughout our society, despite the efforts of Kate Sheppard who was one of the women who worked so hard to enable women to vote (in New Zealand, in 19 September 1893). People with pain are also often stigmatised. My post last week is intriguing in that I pointed out that we cannot determine who is, or isn’t, “faking”. It’s the only post I’ve had with nearly 40 votes, but a total score of 2/5. It’s unplatable to some to think that a subjective experience is just that – subjective, not able to be measured, and for clinicians, that we need to accept what a person says without judgement. Stigma is judgement – let’s not do it.

N. I. Eisenberger, J. M. Jarcho, M. D. Lieberman, and B. D. Naliboff, (2006)“An experimental study of shared sensitivity to physical pain and social rejection,” Pain, 126(1), pp. 132– 138.

Samulowitz, A., Gremyr, I., Eriksson, E., & Hensing, G. (2018). “Brave men” and “emotional women”: A theory-guided literature review on gender bias in health care and gendered norms towards patients with chronic pain. Pain Research and Management, 2018.

Waugh, O. C., Byrne, D. G., & Nicholas, M. K. (2014). Internalized stigma in people living with chronic pain. The Journal of Pain, 15(5), 550-e1.

M. Zhang, M. Barreto, and D. Doyle, (2020) “Stigma-based rejection experiences affect trust in others,” Social Psychological and Personality Science, 11(3), pp. 308–316, 2020.

Zhang, M., Zhang, Y., Li, Z., Hu, L., Kong, Y., & Su, J. (2021). Sexism-Related Stigma Affects Pain Perception. Neural Plasticity, 2021, 1-11. https://doi.org/10.1155/2021/6612456

Why do people with pain report differently on questionnaires than they do in physical performance testing?


One of the topics thrown around by people who don’t have an up-to-date understanding of pain is why people say one thing on a questionnaire, for example, what they can and can’t do, and perform quite differently when asked to do the same task in a clinic. It’s a good question, on the face of it: after all, people should know what they can and can’t do, and be consistent. If there is a difference, well obviously the physical performance test is far more objective than self-report – the therapist is right there watching, so there’s no room for doubt about which measure is The Most Accurate.

The main reason, according to these clinicians, for someone doing differently in the clinic compared with self-reporting, has to be because they’re biased. The person wants to misrepresent what they can and can’t do. Of course.

Superficially, and if you’re not knowledgeable about pain, behaviour, context and human interaction, you could be forgiven for accepting the idea that what you see in clinic is consistent with what’s being done in every day life. The physical movements are pretty much the same and the person is just being asked to do something they do all the time.

BUT – and it’s an enormous exception – humans are not robots. Not body bits that move when they’re pulled like a puppet on a string. People are meaning making, interpreting, social creatures with rapidly responding body systems that represent contexts in relation to memories, predictions and current demands.

I wrote a talk recently on some research that made my heart sing a bit. As an occupational therapist, my profession has long recognised that doing activities (occupations) that hold meaning is quite a different thing from doing a-contextual, meaningless movements. This is why occupational therapists are known to ask about what matters to you, and to use meaningful activities/occupations both as therapy and as outcome (Hitch & Pepin, 2021). The research I referred to was a proposal for an “ecologically grounded functional practice” within rehabilitation (Vaz, Silva, Mancini, Carello & Kinsella-Shaw, 2017). In this paper, the authors point out that “improvements at one level of functioning might not transfer immediately to other levels” and by this they mean that elbow flexion/extension improvements may not transfer into a person being able to feed themselves. They also pointed out that when people perform well-rehearsed activities in the context of goal pursuit – such as getting dressed, ready for work; catching a ball as part of a fast-moving game of netball; hammering a nail – the movements are not just about motor control, they’re about goal-directed behaviour in a context with an interaction between the person, the environment, any tools, the purpose of the activity and so on.

For example, if I want to eat soup, I not only need to have sufficient elbow flexion/extension, I also need to know where the soup bowl is (tried eating soup while lying down?), the texture of the soup (is it thick, thin, lumpy?), the heat of the soup (hot, cold, spicy) and even the social context – I might be OK slurping when I’m on my own, but I’m less inclined to slurp when in polite company. The way in which I carry out the flexion/extension will be very different with each contextual variation.

OK. So context matters, and both the what and why of movement will influence how I go about my performance.
What else?

Well, with a painful condition and especially when I’m not confident I can do it without increasing my pain, I’m much more likely to attempt a difficult movement task in the presence of someone who can monitor what I’m doing. Firstly that person might stop me if they think I’m doing something harmful (professional liability insurance offers some protection!). Secondly, it’s a lot harder to say “no” to someone who is right there in the room! This is called “demand characteristics” and has been associated with problems of the rubber hand illusion (Lush, Vazier & Holcombe, 2020). If someone expects you to do something, you’ll probably do it – because we social creatures don’t like to offend, because the person may inadvertently signal the response they want (see link).

There are other reasons people don’t report the same level of disability on a questionnaire and in physical performance testing: they don’t measure the same things, people forget (especially if they haven’t tried in a while), the day of physical performance testing could be a bad day (or a good day), in physical performance testing the person is usually asked to do it maybe once or twice – in daily life that same activity might be carried out many times across a day, week, month. The environment in a clinical testing environment is typically well-lit, the space around the person is clear, the noise level is usually reasonably low, the floor surface is flat and usually hard lino and free of rugs or pets, there’s minimal distraction, the only thing the person has to think of is this one movement – they’re not picking up the washing off the floor while rushing to put it in the washing machine before dashing out the door to pick the kids up from school.

Even the questions are different – “does pain interfere with…?” is a different question from “can you step up onto this step using a hand rail?”

And don’t let me even start on the meaning of performance either way – for example, if the person is really keen on getting knee surgery, might “good” performance mean they, without even knowing it, alter how they do a movement? What if the person is apprehensive about how the results of this testing might affect their rehabilitation and return to work – again without even knowing it, might this not have some influence on how the person performs?

Testing and measurement is a core skill and research area in psychology. Dang, King & Inzlicht (2020) offer some really good insights into the reasons responses differ between self-report and performance, and to be fair, they don’t even consider the influence of pain and physical capability as I have above. Pain-related performance is a specialty area of its own, nonetheless we can still draw from their paper because many of the problems they recount are absolutely part of pain and disability self-report and physical performance.

They describe the reliability paradox (that reliability = variance between individuals divided by variance between individuals + error variance) – in other words, we need low levels of between-person variability so that any experimental manipulation is maximised. But in real life, we almost always exhibit variability in our performance – so the reliability of two measures limits the correlations that can be observed between then, with lower reliability leading to weaker observed correlations.

The authors also describe the very different response processes involved in self report and performance – as I mentioned above, self-report measures ask people to reflect on what they do in real life in many different contexts that are unstructured. Performance measures take a snapshot based on performance in specific and highly structured situations. Self-report measures capture a person’s perception of their capabilities whereas physical performance reflects the observations of someone else. And performance assessments generally tap into peak performance, not daily performance – tapping into some of the discrepancies we see between “can do” and “will do” (competence-performance discrepancy).

So, when you read arguments on social media from well-known physiotherapists suggesting that the person who reports a difference between what they perceive they can do, and what they have done in a physical performance test is “biased”: know that we have absolutely NO WAY to determine “bias”, “malingering”, “faking bad”, “faking good” – and that there are many well-understood reasons for the difference in performance. Read this paper for more on why we can’t detect “malingering” in people with pain: Tuck, N. L., Johnson, M. H., & Bean, D. J. (2019, Feb). You’d Better Believe It: The Conceptual and Practical Challenges of Assessing Malingering in Patients With Chronic Pain. J Pain, 20(2), 133-145. https://doi.org/10.1016/j.jpain.2018.07.002

Dang, J., King, K. M., & Inzlicht, M. (2020). Why are self-report and behavioral measures weakly correlated?. Trends in cognitive sciences, 24(4), 267-269.

Hitch, D & Pepin, G. (2021) Doing, being, becoming and belonging at the heart of occupational therapy: An analysis of theoretical ways of knowing, Scandinavian Journal of Occupational Therapy, 28:1, 13-25, DOI: 10.1080/11038128.2020.1726454

Lush, P., Vazire, S., & Holcombe, A. (2020). Demand characteristics confound the rubber hand illusion. Collabra: Psychology, 6(1).

https://methods.sagepub.com/reference/the-sage-encyclopedia-of-communication-research-methods/i4252.xml

Not all pain is the same


When I started working in the field of persistent pain, many of the approaches used were based on the idea that every pain was the same. Oh yes, of course we had neuropathic pain and inflammatory pain, but our treatments tended to approach each person as if they were pretty similar. We later refined that approach and started to look at people in groups. In the service I worked in, we used the Westhaven-Yale Multidimensional Pain Inventory which generates three main psychologically-based profiles – and for a long time this was a very useful way of establishing who needed the three-week residential programme, and who would do well with a briefer outpatient programme.

Well things change over time, and we’ve become more aware of what Clifford Woolf describes as a “mechanism-based” classification approach (Woolf, 2004). In this approach, clinicians try to establish the dominant mechanistic group in which a person’s pain might be classified, then suit the treatment to that mechanism. This means clinicians diagnose inflammatory pain, neuropathic pain, and nociplastic pain – and use what looks like the best combination of medications to suit the mechanisms. For example, for neuropathic pain it’s more likely people will be given gabapentin/pregabalin and a tricyclic antidepressant in combination than an opioid.

There’s a problem, though – in fact, TWO problems I can see.

Methods for identifying pain mechanistic groups

The first problem is that we don’t have wonderful methods for establishing the main mechanistic groups.

In fact, in a recent very large and thorough review of methods used to discriminate between each category, the authors found that “few methods have been validated for discrimination between pain mechanism categories”, and although there was “general convergence” between methods, there was also “some disagreement” (not that this is unfamiliar to anyone who reads research!) Shraim, Masse-Alarie, & Hodges, 2021).

What was interesting, albeit not too unexpected, was the overlap of findings between categories because people present with mixed types of pain; and that many of the studies attempted to only discriminate between two of the groups, rather than more. Having said this, the authors identified five groups of method used to help clinicians and researchers distinguish between pain mechanisms: clinical examination; quantitative sensory testing; imaging; diagnostic and laboratory tests; and questionnaires asking participants to describe their experiences.

Now I know that research studies aim to be a little more rigorous than clinical practice, but that should sound an alarm when we begin looking at what we need to do in clinical practice. “Subjective” pain examination included aggravating and easing factors, pain location and pain characteristics (can anyone tell me when pain is NOT subjective?). It also included psychological factors (although my radar went off at this – more of this later!). Physical examination (is this supposed to be ‘objective’ pain examination?) included general clinical assessment, general neurological testing, nerve provocation testing and neurodynamics, clinical bedside somatosensory function testing, movement and functional testing. Quantitative sensory testing had no greater degree of sensitive, specificity and reliability than physical examination and “subjective” history, and laboratory testing was pretty poor despite superficially looking more “accurate.” A similar state exists for questionnaires – oh lordy!

So these authors found 200 methods that could be used to determine which pain fits into a specific box, but overall the results are pretty underwhelming for clinicians wanting a direction for their approach. It’s not helped that the current “gold standard” used is – wait for it – clinician-based diagnosis.

Where are we left? Well, I think we’re not that far away from where we were in the 1990’s and early 2000s. We really don’t have a clear way to distinguish between the various mechanisms, and many people likely present with pain that includes more than mechanism. However – these authors provide a table summarising the commonly used, and possibly most likely approaches to diagnose pain mechanisms, and this is useful for those of us who want the “best guess” for now.

Problem two

At the beginning of this piece, I said there were two problems with using mechanistic descriptors. You can see the problems with reliability, specificity and so on – and the lack of agreement in the research and likely “mixed” presentations we will see in clinic – from my comments above (do read the whole article, though, it’s well worth it).

The second problem is that these descriptors, even when accurate and reliable, don’t tell us anything about the person experiencing pain. Unless, and until we have effective treatments for each of these mechanisms, we are inevitably running experiments to see what might work for this person in front of us. And this means we find less utility in diagnosis than we would if we drew on a case formulation approach.

What’s the difference? Diagnosis allows us to group “like with like” – on the basis of similar underlying mechanisms. We can then treat those mechanisms, and voila! the person recovers! It works well with fractures, with infectious diseases, and even with diabetes. It doesn’t work as well when we don’t have treatments we can use on the mechanisms. For example, although we can diagnose many neurological disorders, for so many of them we have very poor treatments. This means people live with their disease – and this is where a diagnosis falls down. It does not tell us HOW this person will experience their disease. Diagnosis doesn’t explain illness, disability, functional status, or participation.

And, because all of us are unique, this means that a one-size-fits-all approach to persistent pain (or even an algorithm, subtype, or subgroup) isn’t likely to offer clinicians or the person with pain a useful path towards well-being.

You’ll remember I said my radar went off with the psychological assessments included in the methods used to identify a pain mechanism. My reason is this: pain is a stressor. Even a paper-cut captures my attention (albeit just a little until I use a hand sanitiser!), my heart rate goes up a bit, I’m alerted to the experience and want to get away from it. Now imagine if that pain continued. Maybe variably, maybe constantly, maybe intermittently. And imagine if I couldn’t get a good understanding of what’s going on. And perhaps I was being questioned by my clinicians – and maybe even stigmatised. “What do you mean, you have pain we can’t diagnose, we can’t image, we can’t treat?” I’m guessing by now, perhaps some months after my pain started, I’d be feeling a bit irritated, perhaps a bit low in mood, my sleep might not be great, I might find it hard to do what matters to me because I’m not sure if I’m doing myself some harm.

What we don’t know in many studies of pain “predictors” is whether they are cause or effect. There is undoubtedly an association between various measures of pain-related anxiety, avoidance, low mood, thinking the worst. What we do not know is whether this was present before the pain came on – or whether it came afterwards.

So, to my mind, using psychological factors as part of diagnosis risks labelling people and what are probably normal responses to abnormal experiences. Let’s not do that.

Where am I left after reading this paper? I’m glad someone set about doing this review. I think it offers a good summary of the state of play, and identifies some of the current problems with a mechanistic approach. We need to get consensus on definitions, we need far better methods, we need to stop using the word “subjective pain examination” (because ALL pain is subjective), and we need to leave psychological factors out of diagnosis until we can clearly identify which came first.

Shraim, M. A., Masse-Alarie, H., & Hodges, P. W. (2021, Apr 1). Methods to discriminate between mechanism-based categories of pain experienced in the musculoskeletal system: a systematic review. Pain, 162(4), 1007-1037. https://doi.org/10.1097/j.pain.0000000000002113

Woolf CJ. Pain: moving from symptom control toward mechanism-specific pharmacologic management. Ann Intern Med 2004;140:441–51.

Family and friends matter


I’m going back to my series on behavioural approaches to pain management (it’s a slow process!). For the first two go here and here. Now I want to talk about the impact of family and friends on people living with pain.

The people we live with are so influential on what we do and believe about pain. It’s our parents who first taught us the relationship between the word “pain” and the experience we know as pain. It’s our parents and family who responded when we cried, who kissed it better (or not), who told us to “harden up” (or not), who took us to the doctor (or not), who showed us, through their own behaviour, how to “do pain.”

There’s a good deal of research investigating the impact of friends and family on pain behaviour (remember the distinction I make between pain-the-experience and pain behaviour or what we do when we’re sore? click). For instance, a systematic review by Snippen, de Vries, van der Burg-Vermeulen, Hagedoorn and Brouwer (2019) looked at people with chronic diseases, and the attitudes and beliefs of significant others. They found that “positive and encouraging attitudes regarding work participation, encouragement and motivating behaviour and open communication with patients” were facilitators for work participation while “positive attitudes towards sickness absence and advise, encouragement or pressure to refrain from work” were barriers to returning to work.

In another study, Burns, Post, Smith, Porter and colleagues (2019) observed spouse dyads behaviour after arguing then the person with pain undergoing a pain induction task. Spouses that believed that the patient’s pain was a mystery were significantly more likely to be perceived by the patient as giving critical/invalidating responses toward the patient during the discussion; while spouse perceptions that the patient’s pain was a mystery were related to internal and negative attributions spouses made while observing patients display pain behaviors during the structured pain behavior task (p. 1176).

In another study, this one a daily diary study with people living with osteoarthritis in their knee, found that on days when the person with pain reported more thinking the worst, their spouses were more unhappy during the day. And on the days when the partner was more irritated with the person living with pain, that person reported more thinking the worst the next morning. The link? The people with pain who were thinking the worst were also more grumpy through the day, and this was rubbing off on their partner. (Martier, Zhaoyang, Marini, Nah & Darnell, 2019).

Makes sense, doesn’t it? That when we see our loved one demonstrate that they’re sore, and they’re grumpy – and if we’re not sure they’re for real – we might be less supportive as partners than if we think their pain is for real. And over time the pattern of being sympathetic might wear thin – in fact, Chris Main (psychologist) describes a pattern of initial solicitous behaviour (the “there, there dear, I’ll fetch you a cup of tea”), then resentment (“surely you’ve recovered now?”), then anger and punitive behaviour (ignoring the person, getting irritated with them), but then feeling guilty about this (“OMG I know, it’s not your fault and I’ve been so mean”), returning to being solicitous – until the next time the partner feels fed up.

What does this mean for a behavioural approach?

Well, it’s not surprising that if one of the partners thinks the other “should be well now”, they’re likely to be unsympathetic as we begin changing the person’s behaviour. Often we’re attempting to help someone be consistent with their daily activities, and this can often begin by reducing how much should be attempted so the person can “do no more on a good day, and do no less on a bad day.”

And if the partner is really worried about the person with pain, and afraid that doing more is going to increase pain and prolong disability, it’s also not surprising that the partner is likely to be worried about us asking the person to do things differently (especially exercise!).

And don’t forget that during this time, both partners are probably trying to keep some semblance of normal going. They still have the usual household tasks to get done, to pay the bills, to get the kids to and from school, to keep in touch with extended family and friends and so on.

It’s stressful. And we add to the burden when we ask the person to do something different, whether this be doing exercises, using a mindfulness or relaxation technique, perhaps go to various appointments all around town…and if we don’t include the impact of what we expect on the partner, we’re possibly not going to have “the team” on board with the rehabilitation programme.

The very best option is to ask the person’s partner to come in to at least one of our treatment sessions, so we can spend some time talking about what we’re asking the person with pain to do, and getting an indication from the partner about their willingness to follow the programme. The next best option is to write the programme down, and include “things family can do to help” – listing the kinds of things family and friends can do (and what they should avoid doing).

You see, people we see for help never live in a vacuum. They always have a context of friends, family, home, responsibilities, expectations from them, expectations for the work we do. Forgetting about this and expecting a good result fails to recognise the embedded nature of life. Contextual factors are important, no person is an island.

Burns, J. W., Post, K. M., Smith, D. A., Porter, L. S., Buvanendran, A., Fras, A. M., & Keefe, F. J. (2019). Spouse and patient beliefs and perceptions about chronic pain: effects on couple interactions and patient pain behavior. The Journal of Pain, 20(10), 1176-1186.

Martire, L. M., Zhaoyang, R., Marini, C. M., Nah, S., & Darnall, B. D. (2019). Daily and bidirectional linkages between pain catastrophizing and spouse responses. Pain, 160(12), 2841.

Snippen, N. C., de Vries, H. J., van der Burg-Vermeulen, S. J., Hagedoorn, M., & Brouwer, S. (2019). Influence of significant others on work participation of individuals with chronic diseases: a systematic review. BMJ Open, 9(1), e021742. doi: 10.1136/bmjopen-2018-021742

Modifying pain behaviour (2)


Two concepts that receive limited attention in the allied health literature are nomothetic and idiographic approaches. I’m discussing these concepts here because when we’re considering pain behaviour, I think we can focus much more on “generic” (nomothetic) concepts than we do idiographic ones – and yet we say we’re about the unique person in front of us.

Firstly, this site offers a good summary of the difference between nomothetic and idiographic – click

Essentially, nomothetic approaches focus on underlying generalities, perhaps traits, and are a solid part of the science of measurement in psychology. Given that much of our allied health measurement practice is based on psychological theories (such as using aggregated or grouped data to search for differences in means between two groups), it’s not surprising that we’ve tended to reach for a self-report measure when we want to understand what a person thinks and does when they’re sore. Think of the Oxford Knee Score, or the Oswestry Disability Index, for examples!

Here’s an item from the Oswestry Disability Index (Fairbank, Couper, Davies et al, 1980)

Section 5 – Sitting
I can sit in any chair as long as I like.
I can sit in my favorite chair as long as I like.
Pain prevents me from sitting for more than 1 hour.
Pain prevents me from sitting for more than ½ hour.
Pain prevents me from sitting for more than 10
minutes.
Pain prevents me from sitting at all.

When a person reads these items, they’re asked to indicate the answer that best fits their experience, but left unanswered are these points: what time of day? what kind of chair? what is the person doing in the chair? who is around that person? why is the person sitting for a long time? what is it about the pain that stops the person from sitting? what do they think is going on?

While the measure itself is based on rigorous methodology, has excellent psychometric properties and so on – it doesn’t investigate important dimensions that we need as clinicians to help this person perhaps alter their sitting tolerance.

Alternative measurement approaches are available: item response theory is one (click) and multi-level modelling is another (click) – but the former still considers latent traits (ie can we identify a general underlying response that underlies all the variability we see in the data), and multi-level modelling also assumes that the respondents still belong to a general population who will demonstrate similarities around the variable in question.

The problem is that people don’t always follow the rules. Here’s an example:

A woman I saw once had low back pain, and was very afraid to bend forward. She was particularly worried about bending down in the shower to wash her lower legs, and when she saw me she avoided putting her handbag on the floor because this would mean she’d need to bend down to pick it up.

To get around this concern, she’d learned to sit on the floor of her shower to wash her lower legs, used pull-on shoes with elastic laces, or court shoes for work, and she’d put socks and pantihose on while sitting on the floor.

At the same time, she was comfortable sitting for around an hour, was able to stand as a customer service person for an eight hour day, and was happy driving – but not happy about reaching into the back of her car (it was a two-door) because it meant she was bending.

For this woman, her score on the Oswestry was below 20% or considered to be “minimal disability” – and yet she was almost turning herself inside out to be able to do what mattered to her.

An idiographic approach to her situations looks a little more deeply at the function of behaviour in context. If we take a look at the amount of spine flexion within her activities of daily living, we can see that sitting on the floor to wash her legs, and to pull shoes and socks on involves just as much movement as if she was bending down. What was different? Well, she was really afraid she’d slip in the shower and land in an undignified heap on the floor, needing to be rescued – while being naked! She said she’d been told that she shouldn’t bend because she had a disc prolapse and she’d seen one of those spine models with the bright red disc bulge and thought this was going to be much worse if she bent over. She was very concerned about appearances as she worked in a customer service role, so developing a way to still get dressed while avoiding bending forward was really important to her – but it took her much longer to do, much more effort to do it, and she remained quite certain that this red jelly would ooze from her disc if she bent forward.

In a behavioural approach to pain management, it’s important to understand the antecedents and consequences of a behaviour, so we can understand what elicits the behaviour, and what consequences occur to maintain it. In this woman’s case, any context where she might need to lean forward – such as making her bed, picking clothes up from the floor, putting shoes and socks on from standing, picking her handbag up, reaching into the back of her car to fetch something – elicited a thought (image) for her of her disc oozing out. Combined with her interpretation of the advice not to bend when she first sought help, her response was one of fear – and one thing we learn very early on as humans is that we should avoid things that generate fear.

The consequences of her avoiding forward flexion were many: her fears weren’t allayed except in the moment, and she remained highly concerned about the disc bulge; she felt relieved in the moment as she avoided doing the movements she thought would harm her. This is negative reinforcement – fear (negative experience) is reduced (withdrawn) because she avoided the movement (relief – I’ve avoided a disaster!). She also avoided doing many things she’d enjoyed – like playing tennis (bending down to pick up a ball? No way!), picking her clothes up from the floor (she had a home helper do this, and do her washing), she’d changed the shoes she wore to avoid having to bend down to tie laces, and she sat on the floor of her shower to avoid having to bend down to wash her legs.

When we started to work on helping her move on with life, it was really important to understand the unique combination of context and function of her strategies for avoiding bending. Just telling her that her discs wouldn’t bulge out wouldn’t alter those powerful images in her mind! We can’t unlearn an association once we’ve learned it. And she’d been practicing this association between an image of disc bulge oozing and bending – and all the activities where we bend, and all the associations she’d made between jelly wobbling (because the disc is basically jelly, right?), and all the other things she knew about jelly – it’s not strong, it can smear over things, it wobbles, it can melt…. My approach was to help her experience doing without the dire consequences, starting from simple and moving to more challenging over time. More on this next week!

As clinicians, our words matter, as do the images and models we have in our clinics. We also must be mindful that the people we try to help will bring their history and the unique associations they’ve made between things they’ve been told, metaphors they’ve heard and the values that matter to them. Respecting all those vitally important and idiosyncratic aspects of being human is integral to a behavioural approach to pain rehabilitation. Let’s not put people into algorithms or groups or boxes, because if we take the time to learn about their uniqueness we can create more powerful – and fun! approaches to helping them live their lives again.

Fairbank J, Couper J, Davies J, et al. The Oswestry low back pain questionnaire.
Physiotherapy 1980;66:271–3.

…the “so what” question and why it matters to take a break from work


At the conclusion of each of the courses I teach at University of Otago, I ask students the “so what” question. So what that we learned about neurobiology? So what that we discussed social constructs and how they shape pain behaviour? So what that we learn that thoughts and beliefs influence our pain experience? What does it all mean when we’re sitting with a person experiencing pain?

This last week I’ve been on a brief trip to the West Coast of the South Island of New Zealand Aotearoa. It is a wild and isolated part of our country. So wild that in parts the annual rainfall is over 6,000mm (see the map below!), and the wind blows so that the trees grow almost horizontally. For two days there was no power (and thus no internet, no cellphone cover!) and the gravel road to our campsite was closed until 7.00pm while the power lines were being replaced… I won’t talk about the sandflies and mosquitoes – the size of helicopters!! Well perhaps I exaggerate…

Taking a break from talking pain brings me to my “so what” question. Why do I spend my time trying to help people, especially clinicians, learn about pain? Why am I so focused on bringing a narrative that says “we can’t reduce or remove all pain” and at the same time “it’s possible to live well with pain”? What is my “so what”?

Stepping back from the crabby discourse I see so often on social media – like whether hands on or hands off is preferable, whether pain is sensation or perception, whether exercise should be this or that – I think my purpose is to remind everyone, and especially clinicians, that when we’re working with someone who has weird pain that hangs around our job is to find out what this person’s main concern is. And to remember that irrespective of how much we help someone change their pain, ultimately, they will go on to live their own life. Not ours. Theirs.

It struck me from time to time as I swatted sandflies (helicopter sized ones, of course), that many of us work within inflexible processes and systems that demand we identify goals after only just meeting a person. It struck me that the people who develop policy and who get involved in establishing processes are not engaged in public discourse, at least, not in social media where so many of “us” hang out. I pondered how it is that the collective weight of allied health – numbering far more than our medical colleagues – has not yet shifted our conversations about best ways to help people with pain away from symptom reduction, despite our lack of success when it comes to pain. How we continually fixate on “if the pain goes, the person will go back to normal”. How we tout exercise as The Cure despite such small effect sizes on pain intensity, quality and disability. And for exercise, we could substitute needles, manual therapy, taping, medications…. How we want simple recipes, algorithms that sort people into “responders” and “non-responders” while failing to acknowledge that so far we haven’t achieved this and besides these approaches assume that everyone wants the same outcome.

Taking a break from work offers me a chance to refresh my perspective. My pain, it must be said, doesn’t take a break. And that, folks, is the reality for so many people in our communities. Because persistent pain persists. When we’re at work, and when we’re on holiday. When we’re trying to sleep, and when we’re busy with family. And we all come from what was our normal lifestyle. And some clinicians think that if only we would – understand pain neurobiology, pace, exercise, eat right, use mindfulness, check our thinking and get rid of maladaptive beliefs… then life would be fine. But would that life be what I want? Would it look like my life? Would I be able to be ME inside that regimen of all those things?

Clinicians, we can often omit to ask “what’s your main concern about your pain?” And we often forget to find out what that person values in their life. Our goal setting turns out to be OUR goals, often based on pain reduction – or focused on achieving X, Y, Z. Doing this means attention is paid to the end point – but then the process of getting there is left out. And life is a process (OK a journey) not a goal (OK a destination).

As I approach my teaching this year, and my interactions online, I want to emphasise respecting the autonomy and strengths people living with pain bring with them. That a person’s life and choices are theirs to make – and if we try to change people, we’ll fail. We can invite people to experiment with, play with, test, try out different ways of being, but unless we understand a person’s values and work with them, we’re probably not going get more than superficial compliance. Let’s be respectful and honour the complexity of each individual we encounter – and let’s not treat them as part of an algorithm.

Modifying pain behaviour (1)


In my post last week I talked about pain behaviour and why pain behaviours are often a good treatment target in pain rehabilitation. I also talked about pain intensity rating scales and how, because rating scales are a form of communication, the numbers we obtain from them aren’t a true measure of pain: they reflect what the person wants to communicate about their pain to someone at that time and in that context.

This week I want to discuss modifying pain behaviour, and believe me, we are all in the business of modifying behaviour even if we think we’re doing something completely different!

Ethics

One of the issues about modifying behaviour is addressed right at the beginning of Fordyce’s chapter on “Techniques of behavioral analysis and behavior change” and this is the ethical issue of informed consent. It’s important because behaviour change using behaviour modification techniques can operate without the person’s awareness (and does so All The Time). As clinicians, though, we have an obligation to ensure we obtain informed consent from our patient/client before we embark on any treatment. Of course, you and I know that this doesn’t happen in the way that I’d like to see it! When I’m a patient, I’d like to have my options laid out in front of me, with the pro’s and con’s over both short and long term clearly explained. Then I can choose the option that I prefer. But actually, most of the time I’ve received treatment from any clinician, I’ve been given little or no information about alternatives – it’s been assumed that I’ll go along with what the clinician has chosen for me. How’s that for informed consent?

Back to behaviour change. Fordyce clearly details the approach he prefers which is clear discussion with the person about what is proposed – that “well” behaviour will be reinforced via social interaction and “praise”, and “unwell” behaviour will either be ignored or redirected.

Behaviour change done badly

Where I’ve seen behaviour modification done badly is where the clinician fails to indicate to the person that this is the approach being taken (ie no informed consent), where this is applied to all people irrespective of their treatment goals and without discriminating the types of behaviours to be modified, and where it’s applied without empathy or compassion. The kind of “one size fits all” approach. More about this in a minute.

Fordyce points out that “almost every behaviour change problem can be analysed into one or a combination of these three possibilities: 1) Some behaviour is not occurring often enough and needs to be increased or strengthened; 2) some behaviour is occurring too frequently and needs to be diminished in frequency or strength or eliminated; and 3) there is behaviour missing from the person’s repertoire that is needed and that therefore must be learned or acquired.”

Behavioural analysis (lite – more to come in another post!)

So we can work out which behaviours to focus on, as clinicians we need to do some behavioural analysis. This is often best carried out by observing the person – best in his or her natural environment because the contextual cues are present there – but at a pinch, in a clinic setting. I like video for analysing behaviour, particularly something like limping or guarding or compensatory movements, but larger repertoires of behaviour can be self-reported. For example, if someone recognises that they’re resting more often than they want (especially useful if the person values returning to work), then the person can time how long they rest for and work to reduce that time. Fitness trackers or movement trackers can be great for monitoring this. Other options include asking the person’s family about the particular behaviours they notice as indicators that the person is having trouble with their pain: people around the person with pain often know what’s happening well before the person has said anything!

Now this raises my earlier point about lacking empathy or compassion. It doesn’t feel normal to ignore someone who is wincing, looking “pained” or talking about how much they hurt. And this is why, I think, many clinicians don’t enjoy using behaviour modification in a deliberate way – it either feels unsympathetic, so we avoid it, or we do a 180 turn and we apply “ignore all pain behaviour” indiscriminately. Fordyce definitely did NOT suggest this!

Being human in behaviour change

So, how do we approach a person who is distressed? Do we ignore them or comfort them or what? In true time-honoured tradition, I’m going to say “It depends.”

First, we need to analyse the function of the distress in this context, and in the context of our treatment goals. Remember informed consent! We need to clearly articulate and obtain agreement for our behavioural target, and if someone is distressed and this isn’t our target, then we need to respond in an empathic and supportive way. If we’ve observed, however, that the person we’re working with is often distressed as we begin a new activity, perhaps one that pulls the person towards doing something unfamiliar or a bit scary, then we might have a conversation with the person about what we’ve seen, and with agreement, begin to modify our response.

When I describe “function” of distress in this context, I mean “what does the distress elicit from us, and for the person?” – what are the consequences of that distress for the person? If we reduce our expectations from the person, or the person avoids doing the new activity, then we can probably identify that the distress is functioning to reduce the demands we’re putting on the person. Our behaviour as a clinician is being modified by the behaviour of the person – and probably unwittingly. Reducing demands reduces anxiety, a bit, and it may be anxiety about doing that movement (or experiencing pain as a result of doing that movement) that’s eliciting distress. I wouldn’t say being distressed in this context is deliberate – but it’s functioning to draw us away from maintaining the treatment goals we developed with the person.

So what can we do? In this instance, we might remind the person of our agreement to stick to our plan of activity, we can acknowledge that they’re feeling anxious (that’s probably why we’re doing this activity in the first place!), we can reassure the person that we trust that they can do this (boosting self-efficacy via verbal encouragement), and we can maintain our treatment goal.

That’s hard!

Yep. Using this approach is not for the faint-hearted. It means we need to be observant, to always be thinking not just about the form of behaviour we’re seeing, but about its function. We need to monitor our own behaviour (verbal, facial expressions, subtle body shifts, all the non-verbal “tells” we make), and we need to change our own responses to what the person does. And often we find this self-awareness difficult to do. Most of our responses are “automatic” or habitual, and behaviour modification means we need to interrupt our habitual responses so we can help our patient/client do what matters to them.

For a brilliant description of Fordcye’s approach as applied in a case study, Fordyce, Shelton & Dundore (1982) is a great example of how a seriously disabled person was helped via this approach. Remember, this was carried out with the person’s full consent! Chapter 4 of Fordyce’s Behavioral Methods for Chronic Pain and Illness gives the best blow-by-blow description of how to go about this. And for a rebuttal to some of the criticisms of a behavioural approach to pain management, Fordyce, Roberts and Sternbach (1985) offer some very helpful points. That paper also offers some of the best analyses of pain behaviour and why it’s needed as part of pain rehabilitation.

Fordyce, W. E., Shelton, J. L., & Dundore, D. E. (1982). The modification of avoidance learning pain behaviors. Journal of behavioral medicine, 5(4), 405-414.

Fordyce, W. E., Roberts, A. H., & Sternbach, R. A. (1985). The behavioral management of chronic pain: a response to critics. Pain, 22(2), 113-125.

Pain behaviour: what is it and what do we do about it?


I’m re-reading Fordyce’s classic Behavioral Methods for Chronic Pain and Illness and once again I’m struck by how many of the concepts he introduced and systematically investigated are either mis-interpreted and ignored in our current approaches to helping people with persistent pain. Today I’ll explore just a tiny portion of what Fordyce described.

Pain behaviour refers to all the observable actions we do in relation to experiencing pain (NB some people include thoughts as well, but for today I’ll just focus on observable actions). There are roughly two groups of actions: those involuntary ones that we can call nocifensive responses that include reflex withdrawal underpinned by spinal reflexes but including brainstem circuits (see Barik, Hunter Thompson, Seltzer, Ghitani & Chesler, 2018); and those that are developed and shaped by learning (operant conditioning as well as social learning).

When I write about learning, I often have comments about this suggesting people have a choice about what they do, and that this learning must involve conscious awareness – the upshot of these comments is the idea that if we just tell someone that they’re doing something, information alone will be sufficient to change how often they’re doing it. Well, I don’t know about you, but if you’ve ever chewed your nails, changed your diet, decided to go on a social media diet, or do more exercise, you’ll know that there’s an enormous gap between knowing about and being able to follow through. So let me review some of the processes involved in learning and pain behaviour.

Pain behaviour probably has evolutionary significance. What we do when we’re sore acts as a signal to others, whether those actions are voluntary or involuntary. For example, while limping off-loads weight from the sore limb, it can also function to let other people know there’s something wrong. Groaning or sighing also lets people around us know that we’re not OK. Remembering that we’re a social species, being able to let others know that we need help – or not to do what we just did – means we’re more likely to receive attention, and also to warn others about potential danger. Of course, by eliciting help, we’re kinda obligated to help others when they do the same, which may be why when we see someone demonstrating prolonged pain behaviours we tend to feel annoyed: we might be asking ourselves “If they’re not going to reciprocate, why would I help? Dem’s the rules”

Now pain behaviour is also subject to learning principles. In other words, the specific behaviours we do develop in form and frequency depending on context. The underlying analysis goes like this: an antecedent is present (maybe it’s a particular person, location, or occasion), the behaviour occurs, then something in the environment/context occurs – and it’s this “something in the environment/context” that influences whether the behaviour is repeated, and/or the frequency of that behaviour. The easiest example of this is when you watch a three-year-old playing just a little distance from Mum and Dad. When she trips and falls, she’ll probably get up and brush herself down – and then you’ll see her look for Mum or Dad, and if they’re close enough, she’ll probably let out a bit of wail. In the context of Mum and Dad and her falling over, she’s learned that if she cries she’s likely to get a cuddle or some attention, and this is nice. In the absence of Mum and Dad, if she trips she’s less likely to cry because she’s not likely to get that cuddle. Clever huh?

So if that kind of learning occurs from the time we’re little, it’s easy to see how rapidly this pattern of behaving can become habitual, and when it’s habitual it’s unlikely to be something the child is aware she’s doing. Crying, or seeking attention, when we’re sore is something we’ve learned to do from an early age and while the form of that attention likely differs as we mature, the underlying mechanisms still apply (please don’t scream the place down when you go get your Covid vaccination! It’s OK for babies to cry, but not quite so socially appropriate for grown-ups to cry!).

How does the form of that behaviour change? It’s called “shaping” and it is something that occurs naturally through social learning, and it can also occur in a planned way. Take the example of the three-year-old falling and crying: crying is probably OK outdoors where there’s plenty of room and not too much attention being paid to the interactions between parents and child. Take that same behaviour indoors, perhaps in a supermarket or worse – a quiet waiting room – and it’s likely the parents will shush the child more quickly, and be a little more firm about any ongoing wailing. The context is different, the parents respond differently, and the child learns that it’s not OK to cry loudly where there are other people who might not approve. Over time children learn that in different contexts, different ways to attract attention are required. Clearly there’s more technical language we can use to describe this process, but for our purposes this is enough.

Why do we care about this?

Pain behaviour is normal. It’s something we all do. Mostly it functions in a positive way. We signal to others that we need help, we protect the sore body part, and gradually we recover and resume normal life. In some contexts, though, the tendency to continue doing pain behaviours outlives its welcome. In persistent pain it’s particularly problematic, but it’s also problematic in acute pain situations.

Let’s take the example of the dreaded pain rating scale. The 0 = no pain to 10 = most severe pain I can imagine scale. In the context of an emergency room, being asked to rate pain is a quick and very practical way for clinicians to decide how severe the presumed injury/tissue damage is, whether the person needs analgesia, and whether they’re responding to it. Give a number less than 3 or 4 and you’re probably not going to get a lot of pain relief. Give a number closer to 10, and you’ll get something. Give a number greater than 10 and you may get raised eyebrows. In an experiment by Herta Flor (Flor, Knost & Birbaumer, 2002), participants were given an electric shock and asked to rate their pain intensity (also nociceptive detection threshold (aka pain threshold) and pain tolerance). After they’d rated their pain over several trials, they were given one of two conditions: one in which they were given smiley faces and money when their rating was higher than their average rating for the previous trials, and one in which they were given a sad smiley when their rating was lower than their average. Flor and colleagues found that those people who had been given positive smiley faces for higher pain ratings rated their pain intensity significantly higher than those who had been given neutral or negative smileys.

This experiment doesn’t reflect changes in pain intensity. And this is a critical point to note! The stimuli were the same across both groups. What changed was the response offered to participants after they rated their pain. In other words, behaviour associated with experiencing pain and the resultant rewards given for higher ratings was reinforced.

This experiment, along with a large number of others, is one reason why I don’t like pain intensity measures being taken at every treatment session. Pain intensity ratings are behaviours subject to the contingencies that all behaviour is subject to – people learn what to do, and they do it. And they’re unaware of this process.

We often rely on pain intensity ratings in both experimental studies and clinical practice. Unfortunately, while a numeric rating scale or visual analogue scale are quick and dirty, they’re not like a pain thermometer. We just don’t have an objective measure of pain intensity. And we forget this.

Where am I going with this?

A couple of points. I don’t think we can always influence a person’s experience of their pain. This means that we’re often needing to influence what they do about it – because prolonged distress and disability is not good for anyone. Given the social nature of our species, and the involuntary nature of our response to another person’s distress, we’re inclined to try to reduce distress by offering comfort. Nothing wrong with that except where it gets in the way of the person beginning to do things for themselves. As clinicians we need to reinforce actions a person does to increase their capabilities. We also need to limit our reinforcement of illness behaviour, and we need to do this with the consent of the person – being open about why we’re doing this. Remember people learn this stuff without knowing they’re learning it! This means that as clinicians we must stop judging people and what they do in response to pain. Pain behaviour is learned over a loooong time, and it’s reinforced in so many places. People don’t do pain behaviour on purpose. So we can’t judge people as being “non-copers” or having “exaggerated illness behaviour” – we can just gently show the person what happens, why it happens, and what the effect of that pattern of behaving is having on their life.

The second point is that we can’t treat pain ratings as Truth with a T, and think that we’re getting a pure measure of pain intensity – because rating pain on a scale is a behaviour, and it’s influenced in exactly the same way as all behaviours are. This doesn’t mean ignoring someone’s pain intensity – it just means we need to listen to what the person is trying to communicate.

Barik, A., Thompson, J. H., Seltzer, M., Ghitani, N., & Chesler, A. T. (2018). A Brainstem-Spinal Circuit Controlling Nocifensive Behavior. Neuron, 100(6), 1491-1503 e1493. doi: 10.1016/j.neuron.2018.10.037

Flor, Herta, Knost, Bärbel, & Birbaumer, Niels. (2002). The role of operant conditioning in chronic pain: an experimental investigation. Pain, 95(1), 111-118. doi: https://doi.org/10.1016/S0304-3959(01)00385-2