risk factors

Migraine – it still hurts

A quick link to a CME offered by MEDSCAPE CME.  Unfortunately, it is primarily medically-oriented, but this slide show by Dr Sheena Aurora looks at epidemiology, burden and risk factors for migraine.  One in four people in a household in the US will have migraine! That’s a lot of people, and they have a higher risk of cardiovascular disease…

Recovering from a wrist or ankle fracture: pain-related fear, catastrophising and pain influences outcome

I have no idea how many wrist and ankle fractures occur every year, but I can bet it’s not a small number by any imagination. For most of us, I’m guessing we’d expect to have a fracture, wield a wonderfully-autographed cast, get it removed and go on our merry way – but after reading this article, and having seen some very sad people over the years, perhaps my expectations of speedy return to normal might be over-inflated!

This paper by Linton and colleagues from Orebro, Sweden, is a novel one in that most of our knowledge about chronic pain comes from observing people recovering from low back pain. After all, low back pain is one of the most common pains, it’s the one that produces the most long-term work disability, and treatments for it eat up health budgets like little else. BUT, in a fracture, we have a nice clear-cut incident that begins the process of recovery, a well-known pathophysiology, and a well-understood recovery process from a biophysical perspective, so it provides us with a great opportunity to see what the risk factors are for longterm pain and disability.


I’ll just briefly describe the methodology here – 79 participants were sequentially recruited to the study if they had a distal fracture, were aged between 18 – 70 years, without any other fractures and not experiencing dementia. Nine dropped out and these were older and female. Each participant completed a modified Fear Avoidance Beliefs Questionnaire, completing this in respect of their beliefs about pain before their injury; the Pain Catastrophising Scale, numeric rating scale for pain before the injury, expected recovery at six months, and actual recovery at six months. Mobility and strength were also tested a time one and six months later.

Pain and worry reduced over time, as expected. Pain level started at 4 (ranging between 0 – 8), and dropped at T2 and T3. The number of people reporting no pain or no worry increased from T1 to T3 from 4% to 46% and from 30% to 57%.
69% of patients thought they should be fully recovered within six months when they were first asked. 36% thought this at T2, but the percentage who actually thought they were fully recovered at T3 were only 29%. That means that although most people thought they would be recovered within 6 months, actually less than 1/3 actually were. I certainly didn’t expect that!

Mobility and strength ratings at T3 showed 45% were not fully recovered at six months if they had a wrist fracture, while 33% of people with ankle fractures were not fully recovered. And yes, they do describe the processes used to measure mobility and strength.

Looking at fear-avoidance and catastrophising, there was a bit of work to get the groups divided into high and low levels of fear.
The proportion of patients with low fear at T1 and T2 was 54%, while for catastrophising was 56%. 29% of the participants had a high fear and pain profile at T2. What this lead to, after looking at an odds ratio analysis examining the relationship between fear and catastrophising, was that higher fear-avoidance beliefs increased the risk for pain , but high fear wasn’t significantly related to future self-rated recovery, mobility or strength. For catastrophising, there was a relationship only for strength.

What can we conclude?
OK, apart from the small sample size, and the need to use the strength of the noninjured limb as a substitute for the fractured one at T1, and the authors suggest that taking more frequent repeated measures might have helped; we can conclude a couple of interesting things:

  1. People over-estimate how rapidly they will recover from distal fractures, and expect to be fully recovered within six months, when 2/3 of people won’t be
  2. Higher-than-average scores, or increasing scores on pain-related fear and catastrophising are linked with future pain and recovery levels.  This is similar to the repeated findings in back pain and other types of pain, so shouldn’t be unexpected.
  3. As catastrophising and pain-related fear drop over time, this could mask those who are at risk of failing to recover.

Linton and colleagues suggest that close monitoring of fear-avoidance beliefs and catastrophising over the first few weeks of recovery would help health professionals intervene with psychological input designed to reduce fear and normalise expectations.  In other words, reassure patients that it’s normal for distal fractures to take a lot longer than 6 months to fully settle down.

One problem with this for me is that I don’t know anywhere in New Zealand where fear-avoidance and catastrophising is measured in people who have distal fractures! Most orthopaedic departments are singularly lacking in psychologists, and provided the fracture is uniting, patient’s fears and worries rarely get a look-in!  I can’t speak for GP’s who might have some ongoing input with a person who has a distal fracture, but I’d expect that unless the person attends and asks for help, their worries and fears won’t be identified.  This then means the physiotherapist who might be involved to help improve function (and maybe the occupational therapist) are the ones who may be left trying to address the patient’s fears.

Do we need to introduce another layer of psychological input delivered by psychologists for this group of people?

I think not – I do think maybe occupaitonal therapists and physiotherapists could be mindful of the natural healing process, the time it takes to recover, and be aware of how fear-avoidance and catastrophising interact to produce disability – and start to structure graded activities to increase the  confidence a person can have in their own ability to move.  And perhaps encouraging gradual return to functional activities rather than ‘exercises’ per se could be a better approach.  No gadgets please OT’s!  Learning to use the limb, attending to fear and catastrophising and addressing these quickly are well within the clinical skills of occupational therapists, physiotherapists, nurses and GP’s – so watch, listen and respond, even in these acute and so-called ‘simple’ injuries.

Linton, S., Buer, N., Samuelsson, L., & Harms-Ringdahl, K. (2010). Pain-related fear, catastrophizing and pain in the recovery from a fracture Scandinavian Journal of Pain, 1 (1), 38-42 DOI: 10.1016/j.sjpain.2009.09.004

The ABCDEFGHI of persisting pain

There are some things in life that need repeating – the ABC’s are one of them. In this case I mean the ABC’s of persistent pain, and I’m really referring to the risk factors for chronic disability associated with pain.

Diane Jacobs, of Sherwood Physiotherapy, has summarised some of the discussion a few of us were having on SomaSimple about the so-called ‘Yellow Flags’ or psychosocial factors that are known to be associated with disability from pain. It’s available here and is a great resource freely available on the internet. Just don’t copy it without appropriate reference to her, and don’t alter it without consulting Diane or myself. The original ‘Psychosocial yellow flags’ document is available here, integrated with the NZ Acute Low Back Pain Guide.

It’s designed to be given to a patient to go through, preferably with a clinician, to help the person identify factors that might be problematic in their recovery. It can be used as a discussion pointer, and it might be an eye-opener for other clinicians too!

Health literacy: if you can’t read it, you don’t get it


For a long time I’ve been a fan of ‘plainspeak’ or eliminating gobbledegook.  I know my writing is often quite complex, but I try to write as if I’m talking directly to someone – and yes, I do use all those complicated words in real life!

But after working in pain management for a while certain themes keep coming up, and one of them is the large number of people who don’t read and write well.  I value the concept of multiple intelligences, and some people are simply ‘practical’ and hands-on, and don’t like books much – but to actually not be able to read and write well enough to understand common labels and warnings and instructions, that is an incredible handicap.

I have just been reading a paper by Don Nutbeam called ‘The evolving concept of health literacy’ in which he discusses the concept of poor health literacy as a risk factor for poor health status, and conversely, the idea that by improving health literacy it can become an asset to enable people to take control over their health and the factors that shape their health.

Poor health literacy is associated with lower socioeconomic status, reduced use of health resources and information, poorer adherence to treatments and higher drop-out rates.  It’s easy to see why both direct and indirect effects of not being able to make sense of written information might have an effect on health.

Nutbeam’s paper makes the point that decisions about health are made in a context, part of which is the ability of the individual to make their way through the health care system.  This system inevitably involves negotiating paperwork, being aware of pre-requisites for care, being able to clearly state what is needed – and to the right person. More than that, it’s being able to retain the complex information that a health care decision is based on, and maintain that working knowledge to understand why certain actions need to be taken at certain times.  It’s clearly not just about an individual being able to read – its about an interaction between health care systems and individuals, and it includes, according to Nutbeam, three critical points at which literacy directly influences health – access (who do I see? where do I go?), the interaction between the person and the provider (how does the provider ensure the person understands and retains the information?) and self-care (how does the person continue to take best care without directly drawing on the ‘system’).

The alternative view of health literacy as an asset suggests that by improving literacy, it becomes integral to empowering the person to carry out self care – and to receive appropriately tailored information so they can engage in activities that increase wellbeing.

That’s a big ask for a health care system that often does view health as ‘the absense of disease’.

What can you and I do?  Check with our patients whether they can read the information we provide – write simply in bullet point, avoid jargon, make sure patients have time to absorb information before making a decision, and in the end – listening, listening, listening.

D NUTBEAM (2008). The evolving concept of health literacy☆ Social Science & Medicine DOI: 10.1016/j.socscimed.2008.09.050

Screening for the risk of chronic disability


While not quite a Philosopher’s Stone, finding a method to identify ‘at risk’ people has been something clinicians have been searching for over a long time. It has to be brief, easy to use, accurate and acceptable to the people using it.

We have known in the field of chronic pain that some of the most important risk factors are not biophysical (ie reduced range of movement, changes on X-ray), but are psychosocial – but of the myriad factors out there, which ones?!

The Orebro Musculoskeletal Screening Tool is a brief, 25-item self report questionnaire originally developed as a tool for screening ‘at risk’ patients. It has been used widely in New Zealand since the first publication of the Acute Low Back Pain guidelines and screening for ‘yellow flags’ in 1997. It’s also been used in several studies looking at the predictive validity of the questionnaire for work loss and ongoing disability (e.g. Hurley, Dusoir, McDonough, Moore, Baxter,2001).

The study referred to in this post is a further examination of the measure over a longer term than the Hurley et al study – this one is over 3 years. I won’t go through all the methodology of this study, apart from mentioning that the inclusion criteria were people aged between (more…)

Sleep – pain – sleep – pain – sleep

Today a client and I were discussing sleep. She said to me
‘Why is it that I can’t get off to sleep because I’m so sore, then I have a bad pain day, I’m really tired,
and I still can’t go off to sleep, so the next day I have a worse pain day.’

We’ve known anecdotally and in cross-sectional studies that people who have pain often have poor sleep, and when they have poor sleep, their pain gets worse… It seems to work in both directions!
And now, with a well-designed naturalistic study by Edwards and colleagues, we have some good data on the nature of this relationship.

The study was carried out in a national sample of nearly 1000 people who were not patients, but were recruited as part of another study of ‘midlife’. Most of the people were in their mid-40’s, roughly 50:50 male/female, mainly white (90%) and married (81%). For eight consecutive evenings, participants were telephoned, and asked about their sleep and pain ratings. Other information was also gathered, but what is of interest to us as clinicians, is the relationship between sleep and pain and sleep and pain over consecutive nights.

Now, as usual I’m not going anywhere near discussing the mathematics involved in structural equation modelling. One day I’ll blog about it, but not today. If you want to go into more detail yourself – go and read the article yourself. However, the findings were surprisingly clear: sleep predicted subsequent sleep, pain predicted subsequent pain, sleep predicted subsequent pain, pain predicted subsequent sleep.

The authors themselves state “significant effects of previous sleep on next-day pain, and of previous pain on subsequent sleep were found. Interestingly, the relationship was somewhat stronger for the prospective association of sleep with subsequent pain (Beta = .08, Zvalue = 7.9, p < .0001) compared to the prospective association of pain with later sleep (Beta = .04, Zvalue = 3.1, p = .002).”

What this means is that poor sleep predicted increased pain better than high pain predicting poor sleep, but there was a relationship between the two.

In terms of analysing the effect of yesterday’s pain on today’s pain – well, again the findings were quite interesting. The previous days’ pain was more predictive of today’s pain than age, BMI, comorbidities, an emotional disorder or using prescription medication.

Again, quoting the authors of this study, “The results of the present study suggest a significant curvilinear prospective association of sleep duration with subsequent daily pain report; individuals sleeping for less than 6 h, or for 9 h or more, reported more frequent pain complaints the following day. Evaluating the extremes of sleep duration, sleeping for three hours or less was associated with an 81% increase in pain frequency relative to sleeping 6–9 h, and sleeping for more than 11 h was associated with a 137% increase in pain frequency.”

Now that is pretty useful information. Remember that this is a normal, nonclinical population – these people don’t (yet) have persistent pain for which they are seeking treatment. It suggests that attending to poor sleep may be useful to

(1) screen for individuals who may be at risk of developing persistent problems associated with pain

(2) that poor sleep may be part of a group of symptoms that may need to be considered together when we develop new analgesics

(3) day-to-day fluctuations in sleep may be important to consider when evaluating patients in pain. If the person has had poor sleep for several days prior to an appointment, their pain report may be higher than ‘typical’ for them – and the sleep disturbance could be due to factors such as anxiety about the appointment, creating a vicious cycle.

(4) management of poor sleep may need to be the focus of non-pharmacologic pain management to help people cope better with their problem.

Remembering the usual proviso’s for non-experimental methodologies (and that there is maths and statistics involved!), and as the authors acknowledge, multidimensional measures of both sleep and pain would more accurately measure quality of sleep and pain rather than merely quantity, and most importantly, this study involved self report of sleep – which is notoriously unreliable – it is still clear that this study provides us with some food for thought.

And especially given that our clinical population so often complains of poor sleep – perhaps developing an increased awareness of non-pharmacological sleep management would be a good idea.

ABERNETHY, A. (2008). Pain and sleep: Establishing bi-directional association in a population-based sample☆. Pain, 137(1), 1-2. DOI: 10.1016/j.pain.2008.03.022

EDWARDS, R., ALMEIDA, D., KLICK, B., HAYTHORNTHWAITE, J., SMITH, M. (2008). Duration of sleep contributes to next-day pain report in the general population☆. Pain, 137(1), 202-207. DOI: 10.1016/j.pain.2008.01.025

Smoking and pain

Something I’ve noticed many times is the number of people experiencing chronic pain who also smoke. It used to be thought that people who smoke perhaps had poorer health behaviours which lead them to be less fit, less careful about eating well, and perhaps to having poorer responses to stressors generally. And I’ve heard it said that people who smoke have ‘addictive personalities’. Well it’s not quite a simple as either explanation – but it’s time to put some of the evidence forward about smoking, nicotine, and pain.

This is not an exhaustive review by any means, just a couple of papers that I thought might help when considering whether to suggest quitting smoking to people who have ongoing pain.
I ought to add that the studies I refer to don’t demonstrate causation – they are correlational studies, which mean that there is an association between smoking and chronic pain, but we don’t know whether the relationship is direct, or perhaps mediated through other things – and I’ve suggested some above.
Despite this, here are some papers that are quite persuasive.

The first is a large national study published in 2006, specifically looking at tobacco smoking and pain in the general population – quite different from our hospital patients, or those directly seeking treatment. Quoting directly from the abstract: Former and current heavy smokers had higher odds for greater numbers of pain locations and for moderate and intense pain than never smokers after adjustment for analgesic medicament use and behavior-related risk factors. Female former heavy smokers had an adjusted odds ratio (OR) of 1.6 (95% confidence interval, CI, 1.2-2.2) and male former heavy smokers had an adjusted OR of 1.4 (CI 1.1-1.8) for higher numbers of pain locations compared to never smoking women and men respectively (female current smokers: OR 1.4, CI 1.0-1.9; male current smokers: OR 1.3, CI 1.1-1.7).

The findings suggest that former and current heavy smokers are more likely to report more pain locations and more intense pain than never smokers.

Well, that’s quite high odds! And sadly, it doesn’t seem to make a lot of difference whether you’ve stopped smoking or continue to smoke – the odds are still high.

A second study worth considering is this one published in 2005, where people consulting a GP for the first time with acute low back pain were followed. Again quoting from the abstract: At 4 weeks and 3 months 76% of the patients had recovered. Mean pain intensity and mean disability scores dropped 58% and 68%, respectively, of initial levels during the 3 months. The proportion with sickness absence was 8% at 4 weeks and 6% at 3 months. Several sociodemographic, clinical, and psychological factors were of prognostic value. Compared with their respective reference categories, age above 45 years (odds ratio 4.4, 95% confidence interval 1.4-14.0), smoking (3.0, 1.1-8.5), two or more neurological signs (4.6, 1.4-14.9), a score of >90 on the psychosocial screening (3.1, 1.0-9.4), and high levels of distress (4.1, 1.3-12.8) were the best prognostic factors of nonrecovery at 3 months.

And a final study worth mentioning is a very recently published one by Ryall and colleagues, studying people attending primary care and physiotherapy for arm pain: Altogether, 313 (83%) of 375 subjects completed follow-up, including 53% with ‘continuing’ and 24% with ‘unremitting’ pain. ‘Continuing’ pain was predicted most strongly by male sex (OR 1.9, 95% CI 1.2-3.2) (this association was restricted largely to the elbow), higher frequency of pain in the past month at baseline (OR 2.5, 95% CI 1.1-5.6), chronic pain at sites outside the arm (ORs 1.6-2.4 for different sites) and current smoking (OR 3.3, 95% CI 1.6-6.6). There were also indications that mental health and fear-avoidance beliefs influenced prognosis. Predictors for the other two adverse outcomes were similar.

Curious – some very strong pointers that smoking increases the risk of persistent pain and poorer recovery from acute pain, yet I’m not sure that I’ve ever heard that in public health messages about smoking, and I’ve not yet heard of quitting smoking efforts targeting people with chronic pain. Time for action? I think so.
BTW I don’t know why I get smiley’s through my posts – any geeks out there able to tell me? Can’t see it in the HTML code, but then again, I’m not a technophile…

Grotle, M., Brox, J. I., Veierod, M. B., Glomsrod, B., Lonn, J. H., & Vollestad, N. K. (2005). Clinical course and prognostic factors in acute low back pain: patients consulting primary care for the first time. Spine, 30(8), 976-982

John, U., Hanke, M., Meyer, C., Volzke, H., Baumeister, S. E., & Alte, D. (2006). Tobacco smoking in relation to pain in a national general population survey. Preventive Medicine, 43(6), 477-481.

Ryall, C., Coggon, D., Peveler, R., Poole, J., & Palmer, K. T. (2007). A prospective cohort study of arm pain in primary care and physiotherapy–prognostic determinants. Rheumatology, 46(3), 508-515.