What’s missing from pain assessment?

I am stumped. Here is a common clinical quandary – one of the most important outcomes from pain management is to increase participation in daily activity (occupation), yet the measures we use just don’t quite cut it for me. Here’s a couple of examples: in the IMMPACT recommendations from 2008 (in the paper I’ve cited below and from the one I cited earlier this week) two ‘disability’ measures are suggested as a way to assess disability.  One is the ‘Interference’ subscale from the Multidimensional Pain Inventory, and the other is the Pain Disability Index.  Both of these have sound psychometric properties, have been used for a long time in pain management research and clinical use – but both of them have problems.

The problems are about floor and ceiling effects – so they don’t measure very high and very low levels of disability, both are not especially sensitive to small amounts of change so a person needs to make quite significant changes for these measures to capture them, and most importantly to me – they give each domain of functioning equal weighting.  As well – neither of these measure participation in our social world particularly well – what about being an effective parent or being able to go shopping or go to a restaurant or catch a bus?  And how were these measures developed?  Were people with pain asked to generate these domains or were they derived from what clinicians and researchers believe is important.

As I mentioned in my previous post on what people with pain think is important, I don’t believe the ‘client-focused’ aspects of these measures is especially well-developed.  In other words, the measures that are currently available don’t really represent individual’s ability to do what matters in their lives.  And of course these measures are somewhat culturally-bound.

This area of health measurement is something that needs more attention – how to accurately measure, across patient groups and diagnoses, the degree to which pain interferes with doing things.  There’s a tussle between focusing on the components: can the person move each limb, walk, bend, reach and so on; and the larger ‘whole’ of being able to get around the community, have good relationships, manage home life, go to work.

I don’t know what the answer is, but I am looking!  I think there is merit in working towards a reasonably short questionnaire (because seriously, we can give people SO many to complete!) that involves people ranking the importance of an activity as well as their perception of how much pain interferes with being able to do it.

Now before the occupational therapists remind me of COPM (the Canadian Occupational Performance Measure) I need to remind them that this measure takes time and is very individual.  This isn’t quite as helpful in either a clinical or research setting as a pen and paper tool because it does require an occupational therapist to administer it, and I’m not entirely certain of the inter-rater reliability of it.  To me, a measure that needs a clinician to facilitate answering it introduces another set of variables that can’t be easily controlled for, making analysis quite difficult especially in a clinical setting rather than a research setting.  And I do live in a real world where there are three occupational therapists in the Pain Management Centre, and all with quite different ways of eliciting information from people.  Not to mention that one of the reasons I’m currently looking at these measures is the need to find something that can be introduced to another Pain Management Centre so we can compare results!

Whew! Got that off my chest!

Where to from here?

I’ve a suspicion that it might be possible to work with the findings of the IMMPACT group of Turk, Dworkin, Revicki et al. (2008). Maybe if we attended to the concerns about function and participation that people with chronic pain want addressed, and started to develop a measure based on their needs, we might find a way to really monitor useful outcomes.

I think it will take some time – and I hope that people with a strong interest in participation and occupation (ie my occupational therapy colleagues) will get involved in this, along with the physiotherapists and psychologists.

The measure needs to incorporate how well the person with pain is able to get involved in the community, as well as his or her individual activities.  We need to move beyond ‘can you walk?’ and look much more at ‘can you walk where you want to go?’, and we need to recognise that not all domains of functioning are equal.

How this is done is a methodological issue with loads of challenges – but it’s so important that I’m concerned that it hasn’t been looked at before.  What a challenge for us as clinicians!


Dworkin, R., Turk, D., Farrar, J., Haythornthwaite, J., Jensen, M., Katz, N., Kerns, R., Stucki, G., Allen, R., & Bellamy, N. (2005). Core outcome measures for chronic pain clinical trials: IMMPACT recommendations Pain, 113 (1-2), 9-19 DOI: 10.1016/j.pain.2004.09.012



  1. Excellent post. You’re thinking right up my alley. I’m a physio researcher at the University of Western Ontario, and along with colleagues from psychology, epidemiology and occupational therapy we’ve been conducting patient interviews and focus groups in different countries trying to get at what makes a person ‘whole’ again after an injury / pain experience. From this information we’ve constructed a prototype tool (currently in alpha testing), that coincidentally also includes an importance indicator for each of the currently 20 items. Keep an eye out as we move forward (last name: Walton), working alongside people in pain has been a great experience and we’re pretty excited about the potential for this tool.

    1. Music to my ears Dave! I really like the way you’re thinking – no sneak previews? It would be great to ‘road test’ something like this in our Pain Management Centre. cheers Bronnie

  2. There has got to be a cultural aspect I suspect. As soon as you move from ‘can you walk?’ to something more valued, you need to place it within a context. Perhaps a NZ specific assessment would be the first point – eg present recognisable scenarios, and establish the persons confidence to cope with them. Social barbecues, sports days, beach visits, lengthy car journeys all present the individual with physical challenges that raise issues around importance and confidence.
    Just a thought.

    1. That would make it difficult to compare patient groups across countries, but maybe would give a better indication of the ‘interference’ pain is posing in an individual’s life. I hadn’t considered that option before – but well worth considering. Thanks!

  3. What is it we really want to know?
    What patients are able to do ( = health) or how they actually experience their abilities and possibilities (= quality-of-life). Asking about the (dis)ability to walk a specific distance or to the third floor, participate in a barbecue or sport is about health and about what you’re able to do, asking to what extend you experience a problem with your abilities is asking about how you see your situation. The difference is of great importance. Many persons, who have disabilities like being bound to a wheelchair, report good quality of life even though they are not able to walk to the third floor, walk a mile or even get out of that wheelchair, because they have adapted themselves to their situation and recalibrated their values. Just like “pain” is a subjective experience so is the way its consequences affects the person, or to put it another way “the patient must be their own control value”
    Questionnaires like SF-36 and EuroQol asks for specific abilities, other questionnaires like Multidimensional Pain Inventory, Pain Disability Index and Brief Pain Inventory how pain interacts with these specific abilities.
    In my opinion our primary task is to help to improve the quality of life for the individual patient based on their situation and values not based on a more general estimate of what is “good”. When we developed DoloTest we were very clear to ask “To what extent do you experience problems..”

    1. Interesting question, though I don’t really have any answers. As a psychologist who is interested in pain and measurement though, it does seem like a question which i would like to play a part in answering.

      I suspect that the best way forward is to examine the mismatch between what people want to do, versus what they can do. I suspect a discrepancy measure like that might be useful in assessing how we get people to where they want to be, which is presumably the aim of all of this.

      Have you considered implicit measures? The IAT (Implicit Association Test) is widely used, and has been used for pain in one study that I am aware of – http://tiny.cc/e4nqc

      It may perhaps be a good place to start. Please note that there is some controversy about the IAT, but it does appear to have some behavioural predictive power.

      1. It’s great to know that I’m not the only one who asks interesting questions and doesn’t have clear answers! I like the idea of a discrepancy measure – covering domains of ‘occupation’ or function that people with pain consider are important. I hadn’t thought of implicit measures, and I’m not entirely sure how that would work for a measure of disability like this, but it’s always good to consider things from a completely different standpoint **rushes off to go exploring**
        thanks for taking the time to comment – and hopefully you’ll visit again, with the answer!

    2. I think what we are trying to identify is how much pain interferes with being able to do what people want – and it’s possibly a combination of QoL AND function. The purpose behind wanting to measure this is to demonstrate to purchasers and evaluators of various pain management programmes that the input has an effect that is meaningful to individuals. It’s also useful to know whether the intervention reduces dependence on things like health services, home care services, time/energy of relatives etc. and it’s this latter aspect that QoL may not tap into entirely well. There are some people for whom accessing care from another is a core process for their way of managing their life (to reduce distress, increase a sense of belonging etc) – but this can disadvantage the person from whom care is being accessed, and it doesn’t lend itself to the focus of most self management interventions, which is to increase self efficacy and internal locus of control.
      So to me it’s a vexed question still – to what extent does pain interfere with what a person really wants?
      Thanks for your thoughts – and yes I still have more reading to do on your DoloTest!

  4. Have you considered using the Australian Therapy Outcome Measures for OT (AusTOMs)? They are therapist rated and very quick to use. When using this tool, I choose the domain that best fits the client’s goal and rate it on 4 scales: “Impairment, Activity, Participation and Distress”. Hopefully I’ve got this right. I just wanted to bring this measure into the discussion, because I think it has a lot of potential for use in clinical settings. In research you can consider more time consuming tools. In the pain clinic where I work, the AusTOMs for OT are working fairly well to show functional gains. We also use the Brief Pain Inventory. I like that tool too, but some clients find it hard to complete.

  5. I recently did my masters research comparing the measures recommended by the IMMPACT group to the International Classification of Functioning (ICF) framework and to the Core sets for chronic widespread pain put out by the ICF and found the IMMPACT measures to be lacking very much in activities and participation and environmental factors. i have made an evaluation checklist from the core set items and although it is long this information can be done by any dicipline and should be understood and transferred amongst different disciplines and it tells about the patient at a given point in time so if any measures are looked at at a later date then you can note the changes. It is lengthy however a lot of info does not apply to each patient as it offers many questions for any chronic pain patient. The ICF is valid and reliable. I teach it to my students and it would be nice to have more people using it. It is administered by the clinician

    1. Hi Lesley, did you publish this study? I think I recall reading a paper recently that identified the gaps in current psychometrics for pain, and yes, social participation is missing!
      Thanks for your comments.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.