For most of my clinical working life the focus in pain management has been on factors that identify people who have a high risk of developing long-term disability associated with their pain. The tide is turning, though, and increasingly we’re seeing papers published that look instead at treatment provider attitudes, beliefs and behaviours as equally influential in prolonging disability.
The truth is, the health care consultation is an interaction between two people who both have some sort of agenda, and who are part of a community and are influenced by what is the ‘normal’ way to respond to someone’s pain problem. While it’s the person with pain who usually initiates a consultation, it’s the health provider who has a strong influence over what happens next.
One of the most interesting findings in recent years is that the ‘fear-avoidance’ beliefs of health providers strongly influence their advice to people with pain. If a provider has beliefs that moving an acutely painful back is ‘wrong’, then the advice they’ll give will probably be to use pain as a guide. This happens even after promulgation of acute low back pain guidelines. Similarly, if a provider doesn’t believe in ‘psychosocial factors’ then it’s not likely that he or she will identify them and manage them – even if they’re incredibly obvious.
How to measure the effectiveness of an intervention to change provider beliefs has been quite problematic. There have been several attempts, and these are briefly described in the paper by Bowey-Morris, Purcell-Jones and Watson (2010). The problem with these measures has been the unidimensional nature of the measurement tools – and what occurs in a consultation is not unidimensional at all, along with the need to be valid for providers of different professional background.
This study looks at the use of the Pain Attitudes and Beliefs Scale with GP’s, or primary care providers. The PABS was originally developed for physiotherapists, and aimed to identify whether a primarily biomedical view was held, or a biopsychosocial model. The aim of the study was to firstly evaluate the test-retest reliability of the tool, then in a second study, to establish whether it measured change after the practitioners were given a two-hour presentation on the ‘modern management of acute low back pain’, including material about the biopsychosocial approaches.
The presentation on managing acute low back pain covered a review of 4 main topics: trends in disability, including the role of iatrogenic factors; the role of work in maintaining health; current guidelines on diagnostic triage and treatment for LBP; and the introduction of a new service initiative—a LBP triage clinic. (I wish this could happen here in NZ!)
Briefly, the methodology was to identify a group of GP’s, give them the PABS at time one, give it to them again shortly afterward. The presentations were given, and the GP’s asked to complete the questionnaire again for a third time. Not all GP’s were asked to complete all three questionnaires to minimise any learning effects.
The PABS test-retest reliability was good for the biomedical dimension, while the psychosocial was also good but less strongly so than the biomedical. The authors suggest that because they conducted the retest measure about three months after the original measure, their variance was slightly greater than normal – this is because just giving a test can influence an individual to think about it. Despite this, the test-retest reliability was good.
As a measure of change, however, the questionnaire didn’t perform quite as well, nevertheless, it did change in the direction expected – GP’s who scored more highly on the biomedical dimension moved closer to the biopsychosocial dimension.
What can we learn from this?
One of the points made by the authors of this study is very, very important: a change on a questionnaire doesn’t necessarily represent a change in actual practice.
Hopefully, the responses on the PABS did reflect in some way the thoughts and beliefs of the GP’s, but without an additional measure of what occurs in a consultation, and more importantly, what referrals are made and the factors that are addressed, we can’t tell whether education or information make any difference.
An ongoing complaint about ‘Guidelines’ for evidence-informed practice is that many of them appear to have minimal impact on behaviour. The pragmatist in me suggests that the main influence over provider behaviour is payment – and we can certainly see this in New Zealand with ACC influencing who works in pain management and who does not. 20 years ago it was common for most physiotherapy practices in the private sector to give hands-on treatment in an almost unlimited way – now there are an increasing number of multidisciplinary teams with a clinical psychologist involved and a lot less ‘hands-on’ treatment.
I’d love to see the PABS, however, used alongside a couple of other measures – like the Orebro Musculoskeletal Screening Questionnaire, and referral rates to providers like occupational therapists and pain management centres – to establish whether clinicians have a biomedical orientation, or the more evidence-based biopsychosocial orientation. At least that way there could be some monitoring of those who fail to identify glaring psychosocial risk factors – with ‘education’ or some sort of corrective action for those who do not, or will not attend to psychosocial risk factors.
Here’s another thought: what if you and I took this PABS ourselves, and asked ourselves some hard questions about our own orientation. Probably readers of this blog will have a fairly biopsychosocial approach to pain management, but maybe a score on PABS would give us each a bit of a shake-up to review our own beliefs.
Bowey-Morris, J., Purcell-Jones, G., & Watson, P. (2010). Test-Retest Reliability of the Pain Attitudes and Beliefs Scale and Sensitivity to Change in a General Practitioner Population The Clinical Journal of Pain, 26 (2), 144-152 DOI: 10.1097/AJP.0b013e3181bada3d