Two tools for screening risk: STarT Back Tool and Örebro Musculoskeletal Pain Screening Questionnaire
I wrote yesterday about the difficulty there is in grouping patients so that the right treatment is given to the right person at the right time. Today’s post coincidentally follows a similar line – two screening tools that discriminate between ‘high risk’ and ‘low risk’ people with low back pain. The value of this isn’t lost on funders of various services for low back pain, and it shouldn’t be lost on clinicians either. Wouldn’t it be wonderful if the people who really need input get it early, before the problems bed down?

The two tools are the STarT Back Tool (SBT), developed in 2008 by Hill and colleagues to ‘identify patients with potentially treatment modifiable prognostic indicators using a brief, user-friendly tool and to validate cut-off scores for subgrouping patients into 1 of 3 a priori initial treatment options in primary care (low, medium and high risk groups).’ and the Örebro Musculoskeletal Pain Screening Questionnaire (ÖMPSQ), developed by Linton and Hallden in 1998, developed to ‘similarly differentiate primary care back pain patients, and it has a common conceptual purpose with the SBT of identifying high risk patients requiring targeted treatment.’

I’ve used the OMPSQ for some time now, and it is a common tool especially in New Zealand, as it has been available for us in ACC since the late 1990’s, although it has to be asked whether it has actually been used as part of case management. The SBT is unfamiliar to me, but the authors of this study suggest that it is shorter (nine items rather than 24), is scored in a ‘yes’ ‘no’ format rather than 0 – 10, and is thought to be much simpler to score and to use than the OMPSQ. The purpose of both is to identify people who are at risk of ongoing disability, so that services can be targeted. It could be used to identify those who will get better by themselves, or with minimal intervention – which is probably a good thing, as those people might be better left to just ‘get on with it’ rather than being hampered by over-inclusive input that could well hinder recovery.

This study asked 244 patients seen in eight GP practices in the UK to complete the two questionnaires, plus a number of other questionnaires – unfortunately these are not listed in this paper, although they are included in a previous paper (Hill, Dunn, Lewis et al, 2008). Correlations and discriminant analyses were carried out, and the results detailed in this paper.

Only 53% of the patients actually completed the questionnaires. That’s probably quite good in terms of usual response rate, but a little low for the purposes of being able to make effective generalisations. The correlations between the SBT and OMPSQ are reported as being ‘excellent’ or rs=0.80. The subgroups, however, differed a little. Although qualitatively they ‘looked’ the same – ie similar levels of disability, catastrophising, comorbid pain etc – the proportions allocated to each group (low, medium and high) were slightly different, with fewer in the ‘high risk’ group allocated using the SBT.

The implications of this are quite interesting. Putting the limitations of this study aside (cross-sectional data rather than prospective and the non-response bias), it’s worth thinking about why we might use one tool over another when one identifies fewer ‘high risk’ people.

In identifying ‘high risk’ patients, we can assume, on the basis of hundreds of studies around the world, that these people are the ones who will cost the most in terms of long-standing disability. Costs that are not just economic or treatment-associated, but personal and social. I think it’s worth being somewhat over-inclusive rather than under-inclusive in this high risk group (ie identifying some who are at less risk but giving them the benefit of more intense treatment). While this will cost more in terms of up-front treatment costs, it may just prevent some of those people from progressing along that trail toward long-term disability.

Of course, if a ‘stepped care’ approach is used, this might be less of a problem. In a stepped care approach, the same interventions are applied to everyone from the beginning, but if the person fails to progress with one level of intervention, he or she progresses to the next, more intensive, level of care – and so on. People who recover don’t need the greater level of intervention.

This approach might work well if the interventions are simply different levels of intensity, but it doesn’t work so well if the interventions are qualitatively different – and it’s my suspicion they could be.

Let me explain: for many people with low back pain, the problematic attitudes, behaviours and situations don’t exist, or don’t exert the same level of influence over the individual. These are the ‘low risk’ people who will generally manage their recovery well. Do these people need to have input to coach them through ‘how to cope’ with catastrophic thoughts and avoidance? Probably not. They might need simply some effective analgesia and the green light to go ahead and return to normal activities.

For people with high risk factors influencing their recovery, it’s not simply more of the same (more analgesia, more urging to go back to work, more fitness), it’s actually about thinking differently about their pain and doing things differently. Depending on which of the factors that are more influential, it could be about helping family members stand back a bit, or about a workplace that needs help to accommodate the person. It might be about seeing a psychologist, but it might just as easily be about seeing an occupational therapist to help problem-solve through a daily activity plan while coping with the negative thinking that often goes with having to limit activities.

I’m not sure that simply being able to identify low, medium and high risk is enough to ensure the right treatment is provided to the right person at the right time – sure it’s part of it, but until treatment providers and funders appreciate that there need to be tailored approaches to address the specific risk factors, we might be better to avoid the screening process for now.

Hill, J., Dunn, K., Main, C., & Hay, E. (2010). Subgrouping low back pain: A comparison of the STarT Back Tool with the Örebro Musculoskeletal Pain Screening Questionnaire European Journal of Pain, 14 (1), 83-89 DOI: 10.1016/j.ejpain.2009.01.003

Hill JC, Dunn KM, Lewis M, Mullis R, Main CJ, Foster NE, et al. A primary care back
pain screening tool: identifying patient subgroups for initial treatment. Arthritis
Rheum 2008;59(5):632–41.

Linton SJ, Hallden K. Can we screen for problematic back pain? A screening
questionnaire for predicting outcome in acute and subacute back pain. Clin J
Pain 1998;14:209–15.



  1. Welcome back!
    Establishing high and low risk is a hot topic at the moment in NZ. new contracts for ‘private funders’ ie not ministry of health, will apparently reflect the intention of defining these two groups.

    I think it has the potential to be a fantastic move, but as ever my optimism is tempered by the administrators motivations. Rarely are the health outcomes at the forefront of their minds, and far more often the bottom line. Now this should suit us quite well (clinicians also want patients to ‘get better’ and go back to work) but we are probably happier to see that over a longer time frame.

    Does identifying risk levels and then changing intervention lead to better (and faster) outcomes?

    i have recently been involved in a number of pain presentations of less that 6months duration, which is not the norm for the pain management service I work within. The patients have presented quite differently, and required different intervention. Both the patient, the GP and their case manager have all struggled with the concept of chronic pain, and appropriate intervention – at least with someone with 3+ years chronic pain everyone agrees that what has been tried hasn’t fixed them, and it’s likely to stay around.

    So ‘yay’ to earlier recognition of risk factors for chronicity – but watch out if you think you can group them with people with longer term experiences.


    1. Risk for chronicity is not the same as pain duration – which is one of the reasons I almost would prefer ‘chronic pain’ to be left out of the diagnosis, and instead refer to ‘risk for disability’. At least that way it would be the disability that is addressed rather than the pain intensity!

      You’re quite right, it’s difficult for funders, case managers, GP’s and other health professionals to distinguish between ‘chronic pain’ of 6 or more months and ‘chronic pain’ of 3 – 6 months – even though one of the definitions for chronic pain is pain that has continued for 3 months or more. I can understand why tertiary pain management services were established for this group of people with more long-standing chronic pain, because their problems are often more complex than the so-called ‘sub-acute’ group. And the needs are quite different although the end result is often sadly the same. I guess the hope was that tertiary services wouldn’t need to see the ‘sub-acute’ group because community-based secondary services do exist and, with some help, could address many of the problems of these patients.

      If we could only get the attention OFF the pain and ONTO the disability and the factors that prolong disability, maybe we’d have a better chance of helping these people. But IMHO that will only happen when the medics get outcome measures imposed on them to the same extent other health providers have.

Leave a Reply

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

You are commenting using your 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.