Assessing problems with sleep and pain – ii

Last week I wrote about my approach to assessing sleep problems in those with persistent pain. As an ex-insomniac I’ve spent a while learning about sleep so I can understand what’s going on, and why sleep can be such a problem. In this week’s post I want to dig a little deeper into what’s going on with poor sleep, as well as some of the unique features of sleep in people experiencing persistent pain.

Having reviewed the five main areas that are fundamental (and can/should be assessed by anyone working with people who experience persistent pain), the next area I want to look at with people is mood. There are two primary psychopathological contributors to poor sleep: the first we’ve dealt with last week (Question 4 – what’s going through your mind…) which is by far and away the most common initiator and maintainer of insomnia, and it doesn’t even need to be a diagnosable anxiety disorder! The second, you’ll probably have guessed, is depression.

Depression is common in people with both rotten sleep and ongoing pain (Boakye, Olechowski, Rashiq, Verrier, Kerr, Witmans et al, 2016), and there are some suggestions that pain and depression may be related and similar neurobiological processes may be involved for both (increased limbic activity being one of them). In depression, there is increased activity in the HPA Axis, reduced BDNF (brain-derived neurotrophic factor), and reduced 5HT with increased pro-inflammatory cytokines . In persistent pain, there may be activity in the HPA Axis, there is certainly reduced BDNF except in the spinal cord, and reduced 5HT, along with increased pro-inflammatory cytokines. And in sleep disturbances there is also increased activity in the HPA Axis, redced BDNF, reduced 5HT and guess what… increased pro-inflammatory cytokines. And all three interact with one another so that if you happen to be depressed, you’re more likely to experience pain that goes on, and your sleep will also reduce your mood and increase your pain. And the reverse. All very messy indeed!.

What this means is that assessing for low mood and the impact on sleep is important – if someone’s describing waking well before they usually do, in the wee small hours (anywhere from 3 – 5am if they usually wake at 7.00am) I’m ready to screen for low mood. To be honest I always assess for that anyway! Depression is also associated with low motivation and loss of “get up and go” so this is likely to interact with poor sleep, creating a very tired person.

There are three other very important aspects of sleep I like to assess for: sleep apnoea, where someone stops breathing for seconds to minutes at a time, often snorting awake, and this may be associated with snoring and daytime sleepiness. Often the person won’t be aware of their sleep apnoea, so it can be helpful for a bed-partner to let you know whether this is a feature of your patient’s sleep.

The next are a group of movement disorders of sleep, many of which are associated with the third area I assess, which are medications.

Movement disorders of sleep include restless leg syndrome – that feeling of absolutely having to move the legs, usually at night, and relieved by getting up to walk around, but in doing so, making it difficult to sleep. Another is periodic limb movement disorder of sleep, which can be every 5 – 30 seconds of leg twitching all night long, and in some cases, whole body twitching though this is less frequent and less rhythmic. This latter problem may not be noticed by the person – but their bed-mate will know about it! – and this problem may be associated with both sleep apnoea and restless leg, AND some doses of antidepressants. Another common contributor to these problems is low iron levels – worth checking both iron and medications!

Finally with medications, I like to understand not only what the person is taking, but also when they’re taking them. Several points are important here: some medications are usually sedating such as tricyclic antidepressants but in some people nortriptyline can paradoxically increase alertness! If that’s the case, timing the dose is really important and should be discussed with either the prescribing doctor, or a clinical pharmacist. Opioids depress respiration (ie slow breathing down) so can be problematic if the person has sleep apnoea AND is taking opioids, the drive to inhale may be less, causing more frequent and deeper periods without breathing normally. For restless legs and periodic limb movement disorder, some antidepressants (venlafaxine is one of them) in high doses can cause the twitching and once the dose is reduced, this fades away, at least a bit.  There is a very small amount of research suggesting that NSAIDs can influence sleep quality in some people also.

The effects of poor sleep are many: anything from micro-sleeps during the day (problematic while driving or operating machinery!), to more irritability, sluggish responses, less concentration and more difficulty solving problems. Pain is associated with more frequent micro-wakenings during the night (Bjurstrom & Irwin, 2016) but findings with respect to whether deep sleep, REM sleep or light sleep were consistently more affected weren’t clear.

Having completed my assessment, more or less, I can also use a few pen and paper measures: Wolff’s Morning Questions (Wolff, 1974), Kryger’s Subjective Measurements (1991), Pittsburgh Sleep Quality Index (Bysse, Reynolds, Monk et al, 1989) and the Sleep Disturbance Questionnaire (Domino, Blair,& Bridges, 1984) are all useful. Speaking to the partner is an excellent idea because I don’t know about you but I never snore but my partner swears I do! Who do you believe?!

People experiencing insomnia are not very reliable when describing their own sleep habits – we’re terrible at noticing when we’re actually asleep or awake in those early stages of sleep, so we typically think we’ve slept less than we actually have. We also do a whole lot of things to avoid not sleeping – and these can actually prolong and extend our sleeplessness!

We’ll discuss what to do about the factors you may have identified in your sleep assessment in next week’s instalment, but you can rest assured it’s not crucial for you to do anything yourself about some things. For example, if someone has sleep apnoea, referring for a sleep study is important, but not something YOU need to do! But please make sure a referral is suggested to someone who can make it happen. Similarly with medications and sleep movement disorders, it’s not something you should tackle on your own – please discuss managing these with a specialist sleep consultant, psychiatrist, or the person’s own GP. Mood problems – treat as you would any time you find someone with a mood problem.

Next week – off to the Land of Nod: A roadmap?!


Boakye, P. A., Olechowski, C., Rashiq, S., Verrier, M. J., Kerr, B., Witmans, M., . . . Dick, B. D. (2016). A critical review of neurobiological factors involved in the interactions between chronic pain, depression, and sleep disruption. The Clinical Journal of Pain, 32(4), 327-336.

Buysse DJ, Reynolds CF 3rd, Monk TH, et al. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res 1989; 28(2):193–213.

Domino G, Blair G, Bridges A. Subjective assessment of sleep by Sleep Questionnaire. Percept Mot Skills 1984;59(1):163–70.

Kryger MH, Steljes D, Pouliot Z, et al. Subjective versus objective evaluation of hypnotic efficacy: experience with zolpidem. Sleep 1991;14(5):399–407.

Moul DE, Hall M, Pilkonis PA, et al. Self-report measures of insomnia in adults: rationales, choices, and needs. Sleep Medicine Reviews, 2004;8(3):177–98.

Wolff BB. Evaluation of hypnotics in outpatients with insomnia using a questionnaire and a self-rating technique. Clin Pharmacol Ther 1974;15(2):130–40.


Pain – or disability?

One of the fundamental distinctions we need to make when working with people who experience pain is to understand the difference between experiencing pain – and the behaviour or actions or responses we make to this experience. This is crucial because we can never know “what it is like” to experience pain – and all we have to rely on as external observers is what we see the person doing. Differentiating between the various dimensions associated with our experience of pain makes it far easier to address each part in the distinct ways needed.

Let me explain. We know the current definition of pain – an unpleasant sensory and emotional experience associated with actual or potential tissue damage, or described in such terms (IASP, 1979). One of the key points of this definition was to remove the need for nociception as a requirement for pain to be present. So when we unpack what we understand about pain, the first step is to recognise that it’s an experience. Something we can never share with another person – just like we can’t share joy, the taste of a great craft beer, or what a lover’s touch is like.

We therefore have an inexact relationship between two concepts: nociception, or the biological mechanisms at play until the point at which we are conscious of pain; and pain, or the experience of what it is like to have an unpleasant sensory and emotional experience associated with actual or potential tissue damage, or described in such terms.

But wait, there’s more. Given that this experience is a subjective, internal and personal experience – how do we know when someone is experiencing pain?

When I ask students, their answer is “oh you can see grimacing, wincing, or they tell you” – and it’s true! But let’s notice something: they’re all behaviours. Things people do, either involuntarily or voluntarily, to signal that they’re sore. Behaviours or actions are not a direct indication of “what it is like” to experience pain. Like nociception and pain, there is an inexact relationship between what someone does when they’re experiencing pain – and their resultant behaviour. So we now have three somewhat overlapping concepts: nociception (biology), pain (experience), and behaviours (actions). They overlap because there is no direct 1:1 relationship between these concepts – although in some cases it may seem like there is.

What else influences our pain experience? If you’ve been paying attention to my blog these last few weeks, you’ll know that thoughts or beliefs and emotions also influence both our experience of pain and our behaviour. For example, if we know that the pain we’re experiencing is for good (such as post-surgical pain after hip replacement), we tend to be more forgiving, or at least more willing to experience it than if someone attacked us with a scalpel down some back alley! We have plenty of evidence that simply knowing the supposed cause, and something about the biology of pain, can help people to feel a little differently about it (emotions), and to move differently (behaviour) (Moseley & Butler, 2015; Tegner, Frederiksen, Esbensen & Juhl, in press), while emotions in both experimental and clinical studies have been shown to strongly affect pain intensity – and subsequent behavioural responses (Orenius, Raij, Nuortimo, Naatanen, Lipsanen, & Karlsson, 2017). Once again, the relationships are not exact – which is almost always the case when we’re studying complex systems!

Because thoughts, beliefs and emotions have both impacts on nociceptive processes and on pain experience and behaviours, I’ve depicted them as overlapping (if there was a way to show this in 3-D believe me, I would!).

But wait, there’s more!

We know context makes an enormous difference to a person’s experience of pain AND the behaviours they take in response to their pain. While contextual factors don’t directly influence nociception, these factors do influence thoughts and beliefs, emotions, and behaviour. For example, we know that in adolescents with pain, parental responses influence the amount of treatment seeking (Stone, Bruehl, Smith, Garber & Walker, 2018); and that spouses or partners of people living with pain can affect both pain intensity and behaviour because of the way they interact (Burns, Post, Smith, Porter, Buvanendran, Fras & Keefe, 2018). We also know that in different communities, responses to pain can differ: people who pursue body suspension (being pierced and suspended by hooks) are supported by those around them to “hang in there” (no pun intended!). Factors such as legislation make a difference to pursuing treatment, while treatment itself can perpetuate disability and may even increase attention to pain.

Why bother explaining all this?

The implications of understanding these associations are quite profound. Firstly, nociception is a small but important contributor to our pain experience. Most pain starts with a nociceptive stimulus, even if it ultimately ends up less influential than cortical ‘interpretive’ processes. Secondly, the experience we have of pain is something we can’t share – and thirdly the only way we can begin to infer that another person has pain is via their behaviours, or what they do. This means pain measures like the visual analogue scale, FACES scale, numeric rating scale are not direct measures but are used by people to give a message about their pain. All behaviour is influenced by both our thoughts/beliefs and emotions and contextual factors including who is nearby, past responses they’ve made to our messages, what’s normal or expected in various contexts, and the purpose we believe our behaviour will serve. And of course, many of the influences and behaviours we do are not things we’re consciously aware of because we’ve been doing them since we were born.

So when I think about what we might do to help someone with their pain, I firstly acknowledge that I can’t directly influence someone’s own experience. I’m working to influence what they do about their pain, their relationship to their pain, their beliefs and understanding, their emotions and how they communicate this to other people around them. And to me, the first step is being ready to hear what people believe about their pain. Only after I’ve successfully conveyed this to the person can I ever begin to come alongside them to help them change what they do.


Burns, J. W., Post, K. M., Smith, D. A., Porter, L. S., Buvanendran, A., Fras, A. M., & Keefe, F. J. (2018). Spouse criticism and hostility during marital interaction: effects on pain intensity and behaviors among individuals with chronic low back pain. Pain, 159(1), 25-32.

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

Orenius, T. I., Raij, T. T., Nuortimo, A., Näätänen, P., Lipsanen, J., & Karlsson, H. (2017). The interaction of emotion and pain in the insula and secondary somatosensory cortex. Neuroscience, 349, 185-194.

Porreca, F., & Navratilova, E. (2017). Reward, motivation, and emotion of pain and its relief. Pain, 158, S43-S49.

Stone, A. L., Bruehl, S., Smith, C. A., Garber, J., & Walker, L. S. (2018). Social learning pathways in the relation between parental chronic pain and daily pain severity and functional impairment in adolescents with functional abdominal pain. Pain, 159(2), 298-305.

Tegner, H., Frederiksen, P., Esbensen, B. A., & Juhl, C. (2018). Neurophysiological pain-education for patients with chronic low back pain-a systematic review and meta-analysis. The Clinical Journal of Pain.

When it hurts – but it’s important to keep doing

To date, despite years of research and billions of dollars, there is no satisfactory way to reduce pain in all people. In fact, our pain reduction treatments for many forms of persistent pain are pretty poor whether we look at pharmaceuticals, surgery, psychological treatments or even exercise. What this means is there are a lot of disillusioned and frustrated people in our communities – yet life carries on, and people do keep doing!

In an effort to understand what might help people who don’t “find a cure”, researchers and clinicians have been looking at mediators. Mediators are factors that explain a relationship between two variables. In the study I’m examining today, the predictor is pain intensity, and the criterion variable is participating in valued life activities (the things we want or need to do). The research question was whether self-efficacy and/or pain acceptance mediated engaging in valued life activities.

Ahlstrand, Vaz, Falkmer, Thyberg and Bjork (2017) used a cross-sectional study to explore relationships between the variables above in a group of people with rheumatoid arthritis (RA), drawn from three rheumatology registers in South East Sweden. Participants were required to have confirmed RA; be between 18 – 80 years; have had RA for four years or more; and have data included in the quality register – a total of 737 people agreed to take part (from a total of 1277 meeting entry criteria).

The researchers used the Swedish versions of Health Assessment Questionnaire (Wolfe, 1989) to establish degree of difficulty in daily activities, as well as the Valued Life Activities scale (Katz, Morris & Yellin, 2006); the Arthritis Self-Efficacy Scale (Lorig, Chastain, Ung, Shoor & Holman, 1989); and the Chronic Pain Acceptance Questionnaire (Wicksell, Olsson & Melin, 2009).
The statistical analyses included Chi-square tests of independence to identify significant differences in categorical factors due to gender, and steps were taken to establish whether there were gender differences for pain acceptance, self-efficacy and valued life activities. Pearson correlations were used to explore the relationships between acceptance, self efficacy and the valued life activities summary score, and then univariate regressions were undertaken to test each individual factor (eg pain, pain acceptance and self efficacy on valued life activities). Then, only the significant contributors in univariate analyses where entered into the hierarchical linear regression models. The tests were to establish whether self-efficacy would predict valued life activities after acceptance and pain scores were considered.

Finally, structural equation modelling was used to examine the contribution and influence of pain, activity engagement and self-efficacy on difficulties performing valued life activities. A note here: The authors used the structure of the ICF model to name the constructs in their structural equation model.

What did they find?

The people who responded to this survey tended to be less active than those who were on the registers but didn’t respond, so we need to keep this in mind when we interpret their results. They found that women reported slightly more pain than men, but there were no differences between men and women on all measures except that men scored more highly on the symptom control subscale of the self-efficacy measure. A point to note here is that, unlike the Pain Self Efficacy Questionaire, this measure includes attempts to reduce or control pain and/or disability, so it’s a slightly different construct from the PSEQ which measures confidence to engage in doing things despite the pain.

In terms of pain, pain acceptance, and arthritis self-efficacy, there were low to moderate associations between these and engaging in valued life activities. In fact, all pain acceptance and self-efficacy constructs measured in this study were associated with performing valued life activities. In other words, when people are confident, and willing to do things and engage in activities despite pain, the more valued activities they actually do. In fact, one of the more striking findings was a negative relationship between activity engagement and performing valued life activities – those with lower activity engagement scores reported great difficult engaging in what was important to them (not especially surprising given that both scales are about doing what’s important and getting on with life).

Now for the really geeky model: structural equation modeling found a rather complex relationship between all the variables – so complex I’m going to include the diagram.

What does it show? Well, there’s a relationship between pain intensity and valued activity engagement – the more pain, the less people do what’s important. BUT this is mediated by “personal factors” (remember the ICF labels). These personal factors are the pain acceptance activity engagement, self-efficacy for pain and self-efficacy for symptoms. Interestingly, pain willingness, the other subscale on the pain acceptance scale, wasn’t correlated.

Or is it surprising? To my mind there are some interesting conceptual issues with this study. Firstly, in a group that is self-selected and represents slightly more disability than those who didn’t respond, it’s not surprising that pain intensity and disability were correlated. This is something we see often pre-treatment in chronic pain settings. It’s also no surprise to me that the Arthritis self-efficacy scales were associated with valued activities, and with activity engagement – the arthritis self-efficacy scales ask “How certain are you that you can decrease your pain quite a bit?”; “How certain are you that you can that you can make a small-to moderate reduction in your arthritis pain by using methods other than taking extra medication?” amongst other questions. These suggest that pain reduction is a primary aim in arthritis management. The Chronic Pain Acceptance Questionnaire, however, is a very different beast. The Activity Engagement scale is about doing things that are valued (similar to the Valued Life Activity scale), while the  Willingness scale is about being willing to live life again despite pain – for example “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.”.

While the authors argue that this study shows the value of self efficacy, stating “Active management promotes a sense of confidence, or self-efficacy, for dealing with pain that is associated with improved participation in daily activities and wellbeing.” I think the Arthritis Self-Efficacy Scale’s focus on controlling pain and other symptoms is incompatible with the constructs implied in the CPAQ. The ACT (Acceptance and Commitment Therapy) approach to pain is, as I’ve mentioned many times, a focus on engaging in valued activities irrespective of pain intensity – a more achievable goal for many than becoming confident to reduce pain as the ASES measures.

To their credit, the authors also indicate that men and women who continue to experience pain despite optimal medical treatment might benefit from strategies to increase their confidence to manage their own symptoms – but that a focus on pain control instead of participation despite pain is probably unhelpful. They go on to say that “by focusing on pain aceptance and activity engagement despite pain, self-management strategies may change the focus from pain control to a more flexible engagement in valued activities.” I couldn’t agree more – and I wish they’d used the Pain Self Efficacy Questionnaire instead of the ASES in this study. Maybe we need more discussion about appropriate measures in rheumatology research.


Ahlstrand, I., Vaz, S., Falkmer, T., Thyberg, I., & Björk, M. (2017). Self-efficacy and pain acceptance as mediators of the relationship between pain and performance of valued life activities in women and men with rheumatoid arthritis. Clinical Rehabilitation, 31(6), 824-834. doi:10.1177/0269215516646166

Katz PP, Morris A and Yelin EH. (2006). Prevalence and predictors of disability in valued life activities among individuals with rheumatoid arthritis. Annals of Rheumatology Diseases. 65: 763–769.

Lorig K, Chastain RL, Ung E, Shoor S and Holman HR. (1989). Development and evaluation of a scale to measure perceived self-efficacy in people with arthritis. Arthritis & Rheumatism, 32(1): 37–44.

Wicksell RK, Olsson GL and Melin L. (2009). The Chronic Pain Acceptance Questionnaire (CPAQ)-further validation including a confirmatory factor analysis and a comparison with the Tampa Scale of Kinesiophobia. European Journal of Pain, 13: 760–768.

Wolfe F. (1989). A brief clinical health assessment instrument: CLINHAQ. Arthritis & Rheumatism,  32 (suppl): S9

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 – v

Theories are an important part of scientific development. Theories are essentially a collection of propositions or hypotheses that build a picture of what is in order to predict or control or somehow explain what’s going on. The extent to which a theory’s predictions represent what actually happens, given a set of circumstances, allows us to place more or less faith in the adequacy (or perhaps accuracy) of that theory. The problem with social theory is that there are so many complex interactions between variables that it’s very hard to generate hypotheses that represent what actually goes on in the world – so we end up with skinny theory that explains very little, and in turn this allows naysayers to argue “oh but it isn’t so”.

A biopsychosocial framework is one of those messy, complex theoretical models of “the way people are” that beg for people to argue against it. “It’s too complex”, “it’s too broad”, “it’s too reductionist”, “it’s not clinically useful” – all points against this way of viewing people. Yet, after years of using this model, I still find myself unable to find an alternative way of attempting to understand my two clinical questions: why is this person presenting in this way at this time (and what is maintaining their situation), and what can be done to reduce distress and disability?

Social theories are not something many health professionals are introduced to during their undergraduate training. We’re not trained to understand topics like structure of societies, organisations, groups and everyday lives and how they come about. We don’t typically get trained to think about power and who defines what is normal and abnormal, or who generates names for things – classifications, taxonomies, diagnoses. We rarely get to unpack the hidden discourse of who holds power in healthcare delivery, policy development – even social spending on health.

The people I typically see, living with persistent pain, are often from what posh folks call “the wrong side of the tracks”. Many people don’t have good employment histories. They may not have savings, they may live off a benefit. They are often not well-educated, having left school to do manual work. Their daily routines might be chaotic, and the idea of “keeping fit” or “eating well” doesn’t occur to them because their lives are about getting through the day, loving the family they have, and maybe looking towards a tomorrow where things might be different.

In pain management, we’ve not really spent much time examining the kinds of social relationships or social structures in which the people who really struggle with managing pain come from.  I’m not sure I’ve read very much research exploring, for example, whether people who have two jobs and live on a minimum wage experience greater difficulty developing skills in pacing their activities. I’ve not heard much from the people who live in this way expressing their understanding of what contributes to their distress and disability. I don’t see much about how uncertainty of employment pushes people into unsuitable work – while work is good for most people, what about those minimum wage jobs with unsavoury work environments, precarious employment tenure, cold, wet, smelly and physically demanding jobs with little prospect for the future? I don’t see very much about the effect of someone living on the bare bones of their threadbare trews going to see a medical specialist dressed immaculately in a bespoke suit and silk tie, with the handmade shoes and a language of healthcare that is incomprehensible to anyone other than another similarly clad specialist.

For a sociopsychobiological model of pain (yes, that’s a word, and no I haven’t got it backwards – see this) to gain traction, I think it’s timely to ponder the way our communities view persistent pain. Communities include our own healthcare communities – the manual therapy, physical therapy, occupational therapy, nursing, medical enclaves that use special language and dress in certain ways to demonstrate that we know our stuff. And we need to take a minute to understand the communities the people we hope to help come from.

At the stroke of a keyboard, the labels we give to someone – fibromyalgia, “degenerative changes”, “pre-existing condition”, “depression” – alter the treatment that person receives within healthcare. No question about it – if a person is receiving accident compensation (in NZ it’s ACC) and someone gives that kind of label to them, they’re going to the bottom of the health queue. The vagaries of our system mean that person doesn’t receive work-related rehab, they’re disentitled from ACC, no more weekly compensation, and oh yes they now go through the dehumanising process of attending the “Ministry for Social Development”.

I’m not arguing against the way our ACC legislation is written. And I’m not certain that receiving compensation is always a good thing. What I am pointing out is that when health professionals view the person in front of them as “other” – beneficiary, ACC claimant, pain patient – we are issuing a social declaration. And that means we’re exerting a degree of power over them and their lives. The labels we give have power. And this has a significant impact on the way that person views their pain, and the treatment they may receive.

I think until we begin to include, extend, and invite people living with pain to co-investigate their experience and to contribute to our health professional education (including scientific meetings), we’ll carry on thinking of ourselves as somehow superior to, and certainly more powerful than, the people we hope to treat. Hats off to Rajam Roose for developing the San Diego Pain Summit where this year she’s included a patient panel to give an insight into what it means to hear “your pain is just an output of your brain”. Can we have more please.

What can we do to reduce distress and disability? One thing we can do is begin a conversation about persistent pain being something that anyone can experience. It’s just that people without resources end up dealing with not only pain but also lack of power to change the way it’s treated.

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.


Getting persistent pain and disability confused

As I read blogs and tweets and posts on social media, and even peer reviewed papers in journals, I often read that what we’re trying to do in sub-acute pain management is to prevent chronic pain from developing (note, when I talk about pain that goes on beyond healing, more than three months, or has no useful function, I may use the term “chronic” or I may use the more recent term “persistent” – they mean the same thing, except persistent has perhaps less baggage…).

I want to take aim at that focus – to prevent pain from persisting – and think carefully about it. Let’s take a 56 year old woman with a painful knee, a knee that’s been diagnosed as having osteoarthritis (OA). Now, although we have surgical management for OA (a knee replacement – uni-compartment or even a total knee replacement), in most cases surgeons are not enthusiastic about doing a knee replacement on a younger person, particularly someone who is active (plays netball, golf, runs, gardens). So if a knee replacement is not a thing – yet – what do we do? Most of us will know about the value of remaining active and fit, losing weight and maintaining good range of movement (see here for the NICE guidelines, 2017). We know that these things will maintain function – but they won’t stop cartilage deterioration (much, if at all), and they won’t stop the pain. No matter what we do – even medications are not always especially helpful – pain is likely to persist. Does that mean we’ve failed? Reading some of these blogs, it certainly seems it does.

Let’s take back pain – most of us will know back pain occurs periodically throughout life, from the time we’re teens, through to old age. In some people a single bout of back pain happens and then they’re fully recovered and never bothered again, but for many of us, we’ll be troubled with repeated bouts throughout our lives. And still others will have one bout than just never ends (Axen & Leboeuf-Yde, 2013; Vasseljen, Woodhouse, Bjorngaard, & Leivseth, 2013).  This is despite our best efforts to prevent the onset of low back pain, and to treat it effectively – pretty much all our treatments provide a small amount of help but only exercise has been shown to prevent a new bout after the first one (Choi, Verbeek, Wai-San Tam & Jiang, 2010) – and even then the evidence was “moderate” and only at one year.

So… when we begin to examine claims that by treating musculoskeletal problems early we can prevent pain from becoming chronic or ongoing, I think we need to stop and pause before letting the blood rush to our head.

If we can’t prevent pain from hanging around, what can we do? What is the aim of all this treatment?

Well, let’s take a quick look at the Global Burden of Disease (Hoy, March, Brooks, Blyth, Woolf, Bain et al, 2014). In this piece of work, “Out of all 291 conditions studied in the Global Burden of Disease 2010 Study, LBP ranked highest in terms of disability (YLDs), and sixth in terms of overall burden (DALYs). The global point prevalence of LBP was 9.4% (95% CI 9.0 to 9.8). DALYs increased from 58.2 million (M) (95% CI 39.9M to 78.1M) in 1990 to 83.0M (95% CI 56.6M to 111.9M) in 2010. Prevalence and burden increased with age.” [emphasis mine].

What this means is that although low back pain is not a fatal disease, that may well be the problem – people don’t die from low back pain, they live with disability all the days of their life. And worse, the burden of low back pain is increasing. And this is despite all the work we (you, me, the entire health system) is putting in.

If we can’t “get rid of” low back pain (and it looks like we don’t yet have the tools to do so), what are we trying to do?

Given our poor outcomes for completely curing low back pain, we need to aim to reduce the impact of pain on people’s lives.

And not just low back pain, but things like tennis elbow, frozen shoulder, neck pain, abdominal pain, pelvic pain, headache, migraine, osteoarthritis…

For a moment, let’s think about the effect on a person going through treatment, being promised that “pain education” will reduce their pain, that exercises will get rid of their pain, that gadget A or B will get rid of their pain, that treatment Y or Z will get rid of their pain. What do you think it feels like to be completely adherent about everything you’re being asked to do, but still feeling a failure because that pain does not go? Think of the language used by some of our colleagues – “failed back syndrome”? Who failed, exactly?

Before I get harangued for breathing the word that, ooops, our treatments don’t work very well, let me address the issue of “pain education” and pain intensity. Don’t forget that the only way we can know how much it hurts someone is by asking them. And our usual tool is that 0 – 10 scale, where 0 = no pain and 10 = most extreme pain imagined. Have you ever tried doing that on yourself? Seriously – how do you rate your own pain? Some of that pain rating is about how much we’re prepared to (capable of) putting up with. Some of that rating is about how bothered (fed up, distressed, frustrated) we are about our pain. Some of it is about “OMG I don’t know what this is and how long it’s going to go on for”.

What this means is that when someone gives an explanation it can –

  • make the experience less frightening,
  • less distressing,
  • more understandable,
  • less bothersome

and as a result, when we’re then asked for our pain intensity rating on that darned scale, we reduce the score we give our pain. It does not necessarily mean the pain has reduced in intensity – a pain scale is a means of communicating something about our experience, thus it’s a pain-associated behaviour with the purpose of communicating something. So if a person isn’t ‘convinced’ by our pain education, you know they’ll keep their score pretty high.

So, there are some people for whom we cannot reduce or get rid of their pain. It’s likely to persist. And it’s these people who can be viewed as “heartsink” patients, who hang around not getting better. Well, unless we begin looking at their experience and examine what they’re looking for (and believe me, it’s not pain reduction – it’s what pain reduction means they can do) we’re going to be stuck. And so will they. Let’s get it into our heads that pain reduction is not achievable for all, but reducing the impact of pain on life is something we can all help with. Let’s stop demonising the person who has to live with pain that doesn’t respond to all our ministrations and begin looking deeply at ourselves and why we avoid recognising that we can’t win ’em all. And let’s get on with the business of helping people do what’s important in their lives, irrespective of pain.



Axén, I., & Leboeuf-Yde, C. (2013). Trajectories of low back pain. Best Practice & Research Clinical Rheumatology, 27(5), 601-612. doi: http://dx.doi.org/10.1016/j.berh.2013.10.004

Choi, B. K. L., Verbeek, J. H., Wai-San Tam, W., & Jiang, J. Y. (2010). Exercises for prevention of recurrences of low-back pain. Occupational and Environmental Medicine, 67(11), 795-796. doi:10.1136/oem.2010.059873

Hoy, D., March, L., Brooks, P., Blyth, F., Woolf, A., Bain, C., . . . Buchbinder, R. (2014). The global burden of low back pain: Estimates from the global burden of disease 2010 study. Annals of the Rheumatic Diseases, 73(6), 968-974. doi:10.1136/annrheumdis-2013-204428

Vasseljen, O., Woodhouse, A., Bjorngaard, J.H., & Leivseth, L. (2013). Natural course of acute neck and low back pain in the general population: The HUNT study. Pain, 154(8), 1237-1244.

Knee pain – and central sensitisation

Last week I started to discuss central sensitisation indicators in people with osteoarthritic knees, based on a paper by Lluch, Nijs, Courtney, Rebbeck, Wylde & Baert, et al (2017). I’m going to continue with this topic this week, because with the rise of osteoarthritis in the general population and particularly the impact of an aging population, I think we will all need to think hard about how we conceptualise osteoarthritis, and what we do for management. While efforts within my own Department (CReaTE – tissue engineering) involve developing new ways to remodel knee-joint tissues, we know that it will be some years before this approach is widely available (human trials haven’t started yet), and given the relative lack of funding for joint replacements, I think developing effective assessment and rehabilitation for painful knees is a real area of development.

So last week I discussed using simple measures such as >5 on a 0 – 10 VAS (NRS), pain drawings/maps showing radiating pain or widely distributed pain, the pattern of pain fluctuation (during activity, with an increase after activity), and using a couple of fairly simple questionnaires to help identify those most likely experiencing more than the “simple” OA pain we’ve learned about. And as always, identifying psychosocial factors which can lead to increased disability and distress is important.

Along with the clinical interview, we usually incorporate physical examination or physical performance testing. There are some indicators that might be useful such as inconsistent responses to our usual physical examination (ie testing increases pain even though some of them shouldn’t do so) – this should not be interpreted as a sign that the person is “faking bad” or exaggerating their experience. I can’t emphasise this enough! It’s possible that anxiety on the part of a person can wind the nervous system up – leading to what is usually non-nociceptive input being interpreted as nociceptive (Courtney, Kavchak, Lowry et al, 2010).

Another indicator is the presence of widespread hypersensitivity to mechanical stimuli – it’s a common finding in people who have central sensitisation and includes increased response to pressure and touch. You could, as a clinician, use a pressure algometer both close to the knee, and further away, to establish over-excitability of the nociceptive pathways. Interpreting findings using pressure algometry is not straightforward because there is overlap between those with OA and those without, but it’s possible to use norms from the general population (such as Nesiri, Scaramozzino, Andersen et al, 2011). It’s a bit of a challenge because of the overlap between the two populations, but can add to the clinical picture. Pain (allodynia) on light touch or being stroked with a cottonwool ball around the knee, is definitely a clue that something’s up.

Both thermal hyperalgesia and tactile hypoaesthesia (reduced sensitivity to von Frey fibre testing) have been associated with central sensitisation – if you don’t have formal testing apparatus, the back of a warmed teaspoon placed on the skin for 10 seconds should be experienced as hot but not painful in someone who isn’t tending to central sensitisation, and you can use cottonbuds (or cottonwool) to identify loss of sensation acuity, provided you do so in a systematic way (the authors suggest starting where it’s most painful and stimulating the skin in a wheel spoke pattern, gradually widening out).

Putting it all together

Any single test, on its own, is unlikely to be a good predictor of central sensitisation, but when combined with the information you obtain from the person, along with the relevant questionnaires, should begin to help develop a picture of who is likely to have a less-than-ideal response to planned trauma. What we do about reducing the potential for central sensitisation is still  begin hotly debated but we DO know that giving good information about pain mechanisms, and encouraging graded exposure and graded activity can be helpful. Given that exercise is a good approach for reducing the impact of osteoarthritis in the knee, for those with the additional burden of central sensitisation, I think swimming or hydrotherapy could also be helpful, as could mindfulness and even mindful movement like tai chi, yoga or xi gong.


People living with OA in their knees often spend many years having difficulty managing their pain before they are able to have surgery. From recent research in New Zealand, I don’t think many people are offered a pain “education” approach, and indeed, I’d bet there are a lot of people who don’t get referred for movement-based therapy either. Misunderstanding is rife in OA, with some people uncertain of the difference between osteoarthritis and rheumatoid arthritis, and others very worried that they’re going to “wear the joint out” if they exercise. While OA isn’t as sexy as low back pain, doesn’t have the economic cost of low back pain, and has a reasonable surgical option – it is still a significant problem for many people. Helping those people be more confident to move, helping reduce their uncertainty about the effect of movement on their joints, and giving them an opportunity to think differently about their knee pain would be a real step forward. Surgery, while helpful for many, is either not available or unsuccessful for others, and it’s time we attended to their needs as well.


Courtney CA, Kavchak AE, Lowry CD, et al. (2010). Interpreting joint pain: quantitative sensory testing in musculoskeletal management. Journal of Orthopaedic Sports Physical Therapy. 40:818–825.

Lluch Girbes E, Meeus M, Baert I, et al. (2015) Balancing “hands-on” with “hands-off” physical therapy interventions for the treatment of central sensitization pain in osteoarthritis. Manual Therapy. 20:349–352.

Lluch, E., Nijs, J., Courtney, C. A., Rebbeck, T., Wylde, V., Baert, I., . . . Skou, S. T. (2017). Clinical descriptors for the recognition of central sensitization pain in patients with knee osteoarthritis. Disability and Rehabilitation, 1-10. doi:10.1080/09638288.2017.1358770

Neziri AY, Scaramozzino P, Andersen OK, et al. (2011). Reference values of mechanical and thermal pain tests in a pain-free population. European Journal of Pain. 15:376–383.