Using a new avoidance measure in the clinic

A new measure of avoidance is a pretty good thing. Until now we’ve used self report questionnaires (such as the Tampa Scale for Kinesiophobia, or the Pain Catastrophising Scale), often combined with a measure of disability like the Oswestry Disability Index to determine who might be unnecessarily restricting daily activities out of fear of pain or injury. These are useful instruments, but don’t give us the full picture because many people with back pain don’t see that their avoidance might be because of pain-related fear – after all, it makes sense to not do movements that hurt or could be harmful, right?

Behavioural avoidance tests (BAT) are measures developed to assess observable avoidance behaviour. They’ve been used for many years for things like OCD and phobias for both assessments and treatments. The person is asked to approach a feared stimulus in a standardised environment to generate fear-related behaviours without the biases that arise from self-report (like not wanting to look bad, or being unaware of a fear).

This new measure involves asking a person to carry out 10 repetitions of certain movements designed to provoke avoidance. The link for the full instructions for this test is this: click

Essentially, the person is shown how to carry out the movements (demonstrated by the examiner/clinician), then they are asked to do the same set of movements ten times.  Each set of movements is rated 0 = performs exactly as the clinician does; 1 = movement is performed but the client uses safety behaviours such as holding the breath, taking medication before doing the task, asking for help, or motor behaviours such as keeping the back straight (rotation and bending movements are involved); 2 = the person avoids doing the movement, and if the person performs fewer than 10 repetitions, those that are not completed are also coded 2. The range of scores obtainable are 0 – 60.

How and when would you use this test?

It’s tempting to rush in and use a new test simply because it’s new and groovy, so some caution is required.

My questions are: (1) does it help me (or the person) obtain a deeper understanding of the contributing factors to their problem? (2) Is it more reliable or more valid than other tests? (3) Is it able to be used in a clinical setting? (4) Does it help me generate better hypotheses as to what’s going on for this person? (5) I also ask about cost, time required, scoring and whether special training is required.

This test is very useful for answering question (1). It provides me with a greater opportunity to review the thoughts, beliefs and behaviours of a person in the moment. This means I can very quickly identify even the subtle safety behaviours, and obtain the “what’s going through your mind” of the person. If I record the movements, I can show the person what’s going on. NB This is NOT intended to be a test of biomechanical efficiency, or to identify “flaws” in movement patterns. This is NOT a physical performance test, it’s a test of behaviour and belief. Don’t even try to use it as a traditional performance test, or I will find you and I will kill (oops, wrong story).

It is more valid than other tests – the authors indicate it is more strongly associated with measures of disability than measures of pain-related fear and avoidance behaviour. This is expected, because it’s possible to be afraid of something but actually do it (public speaking anyone?), and measures of disability don’t consider the cause of that disability (it could be wonky knees, or a dicky ticker!).

It’s easy to do in a clinical setting – A crate of water bottles (~8 kg) and a table (heights ~68 cm) are needed to conduct the BAT-Back. The crate weighed  7.8 kg including six one-litre plastic bottles. One could argue that people might find doing this test in a clinic is less threatening than doing it in real life, and this is quite correct. The setting is contained, there’s a health professional around, the load won’t break and there’s no time pressure, so it’s not ecologically valid for many real world settings – but it’s better than doing a ROM assessment, or just asking the person!

Does it help me generate better hypotheses? Yes it certainly does, provided I take my biomechanical hat off and don’t mix up a BAT with a physical performance assessment. We know that biomechanics are important in some instances, but when it comes to low back pain it doesn’t seem to have as much influence as a person’s thoughts and beliefs – and more importantly, their tendency to just not do certain movements. This test allows me to go through the thoughts that flash through a person’s mind as they do the movement, thus helping me and the person more accurately identify what it is about the movement that’s bothering them. Then we can go on to test their belief and establish whether the consequences are, in fact, worse than the effects of avoidance.

Finally, is it cost-effective? Overall I’d say yes – with a caveat. You need to be very good at spotting safety behaviours, and you need to have a very clear understanding about the purpose of this test, and you may need training to develop these skills and the underlying conceptual understanding of behavioural analysis.

When would I use it? Any time I suspect a person is profoundly disabled as a result of their back pain, but does not present with depression, other tissue changes (limb fracture, wonky knees or ankles etc) that would influence the level of disability. If a person has elevated scores on the TSK or PCS. If they have elevated scores on measures of disability. If I think they may respond to a behavioural approach.

Oh, the authors say very clearly that one of the confounds of this test is when a person has biological factors such as bony changes to the vertebrae, shortened muscles, arthritic knees and so on. So you can put your biomechanical hat on – but remember the overall purpose of this test is to understand what’s going on in the person’s mind when they perform these movements.

Scoring and normative data has not yet been compiled. Perhaps that’s a Masters research project for someone?

Holzapfel, S., Riecke, J., Rief, W., Schneider, J., & Glombiewski, J. A. (in press). Development and validation of the behavioral avoidance test – back pain (bat-back) for patients with chronic low back pain. Clinical Journal of Pain.




  1. Yes, it could be a useful test, but I would add an additional question to your list – Are there any risks associated the test, especially in those patients with significant biomedical pathology?

    1. As with any assessment, including quantitative sensory testing, range of movement, strength or cardiovascular testing, there can be physical risks. And yes, I’d include your question in my battery too! It’s part of my standard of practice to always have completed a normal history, examination and to use a heart rate monitor before asking the person to do anything “out of the ordinary”. My question to you is – what risks are there from bending to pick up a crate weighing 7.8kg from the floor and rotating to place it on a hip-height table? I can’t think of any – red flags should have been identified when taking a history or carrying out the examination. Most people being tested for avoidance will be in the sub-acute to chronic group, so the likelihood for doing harm is minimal. cheers Bronnie

  2. Hi Bronnie
    I rather like the use of this rather than a subjective questionnaire where i just see the answers but i am never really sure it is not just that they think it might hurt but they might actually do it and be fine and it was just a thought and not a real fear or it was a fear on paper that got dispelled

    1. I think they’re complementary – what’s nice about this is not just that it provides a practical assessment tool, but that it can be replicated, scored and norms can be developed. It’s a real challenge when thoughts and beliefs relevant to avoidance are so often idiosyncratic, and there is balance needed between reproducible and individualised assessment – but I think this approach certainly sheds more light on the factors relevant to avoidance than simply asking.

  3. I don’t like this…. I wonder how she explains to people why she’s using it. If she doesn’t tell them what she’s really testing, it’s not very ethical. But if she does, it wouldn’t work. Also if they push themselves to the point of causing a flare up, then it’s almost like they’re doing it for no reason, because she’s not testing what they think she is. Did I miss something?

    Sent from my iPhone


    1. I always explain why I use an assessment and what I’m looking for. I’m also interested in identifying whether a person is fearful of a flare-up – that’s just as important as understanding how much, how often and which positions are difficult to do. We collaborate to understand what’s going on. Given I have chronic pain (fibromyalgia) I think I can understand your concern, but as a clinician I also recognise how inaccurate so many of my own beliefs about my own capabilities can be. Drop in for a chat on here sometime, or on Facebook, and let’s get to know one another. I don’t really like being called “she” 😉

  4. My fave example demonstrating Catastrophising is to discuss PACE in the Chronic Fatigue community. Facts are irrelevant, data is skewed by bias, and it’s all a govt conspiracy!

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 )

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.