Quick reminder from last post: The aim of our study[1] was to evaluate the assumptions that were made when translating the individual study criteria[2-6] (eg, all the criteria from the original subgrouping studies) into the classification algorithm.
To evaluate the impact of these changes made to the individual study criteria, we recruited 250 patients with acute/subacute low back pain and gave them a standardized assessment. Then we classified them into treatment subgroups using both the individual study criteria AND the classification algorithm. To be fancy we also looked at the number of people who were classified using the top part of the algorithm (aka ‘clear’ classifications) and the bottom table of the algorithm (aka ‘unclear’ classifications) and looked at the reliability of the classification algorithm decision.
We found that the algorithm tends to prioritize manipulation. When using the individual study criteria approximately 35% of patients met the criteria for the manipulation subgroup. However, when using the algorithm, 42% of patients were placed in the manip subgroup. Direction specific exercise, on the other hand, went from 45% in the individual study criteria to 31% in the algorithm.
We also found that 25% of patients meet more than one subgroup using the individual study criteria. This confirms one of the assumptions – the need for the hierarchical ordering in the algorithm, because otherwise we wouldn’t know what treatment to give these patients. Unfortunately, this finding just creates more work for researchers, because it doesn’t tell us which treatment a patient should get should they meet 2 or more subgroups. The algorithm by nature forces people into one subgroup, but we don’t know yet if this subgroup is the right one.
Last, we found that 25% of patients do not meet any of the subgroups based on the individual study criteria. This confirms another of the assumptions – the need for the creation of a bottom table in the algorithm – because otherwise we wouldn’t know what treatment to give these patients. Now these 25% of people not meeting any subgroup (using the individual study criteria) should correspond to the number of ‘unclear’ classifications (when using the algorithm) as this was why the bottom table was created. Interestingly, more people had unclear classifications using the algorithm than we would expect (eg, 34% of people). This extra 9% is really important because we also found that the reliability of the classification decision for unclear classifications is really bad. This opens up a whole can of worms such as – if we can’t reliably decide on a treatment for these people with unclear classifications with the algorithm, do they actually benefit from receiving ‘matched’ treatment? Are people with unclear classification actually unsuitable for application of the algorithm?
Clinically our data suggest that the most important role of the clinician, if using the algorithm to guide treatment, is to closely monitor the patient’s response to the treatment. Many of you may be thinking, ‘well…duh’, but I think it warrants blatantly stating the importance of clinical thought and reason. While an RCT suggests that people do better when they received ‘matched’ treatment based on the algorithm (compared to unmatched treatment),[7] I’d argue our findings suggest that we shouldn’t just blindly follow the algorithm’s suggestion.
(Disclaimer – it was never Julie Fritz’s intention that the algorithm dictate treatment choice, rather it should GUIDE treatment choice, but sometimes I just like to stir the pot). Stir, stir, stir….
Tasha Stanton
Tasha Stanton is a postdoctoral research fellow working with the Body in Mind Research Group both in Adelaide (at University of South Australia) and in Sydney (at Neuroscience Research Australia). Tash has done a bit of hopping around in her career, from studying physio in her undergrad, to spinal biomechanics in her Master’s, to clinical epidemiology in her PhD, and now to clinical neuroscience in her postdoc. Amazingly, there has been a common thread through all this hopping and that common thread is pain. What is pain? Why do we have it? And why doesn’t it go away?
Tash’s research interests lie in understanding the neuroscience behind pain and its clinical implications. She also really likes nifty experiments that may have no clinical value yet, but whose coolness factor tops the charts. Last, Tash is a bit mad about running, enjoying a good red with friends and organizing theme parties.
References:
[1] Stanton TR, Fritz JM, Hancock MJ, Latimer J, Maher CG, Wand BM, & Parent EC (2011). Evaluation of a treatment-based classification algorithm for low back pain: a cross-sectional study. Physical therapy, 91 (4), 496-509 PMID: 21330450
[2] Flynn T, Fritz J, Whitman J, Wainner R, Magel J, Rendeiro D, Butler B, Garber M, & Allison S (2002). A clinical prediction rule for classifying patients with low back pain who demonstrate short-term improvement with spinal manipulation. Spine, 27 (24), 2835-43 PMID: 12486357
[3] Hicks GE, Fritz JM, Delitto A, & McGill SM (2005). Preliminary development of a clinical prediction rule for determining which patients with low back pain will respond to a stabilization exercise program. Archives of physical medicine and rehabilitation, 86 (9), 1753-62 PMID: 16181938
[4] Browder DA, Childs JD, Cleland JA, & Fritz JM (2007). Effectiveness of an extension-oriented treatment approach in a subgroup of subjects with low back pain: a randomized clinical trial. Physical therapy, 87 (12) PMID: 17895350
[5] Long A, Donelson R, & Fung T (2004). Does it matter which exercise? A randomized control trial of exercise for low back pain. Spine, 29 (23), 2593-602 PMID: 15564907
[6] Fritz, J., Lindsay, W., Matheson, J., Brennan, G., Hunter, S., Moffit, S., Swalberg, A., & Rodriquez, B. (2007). Is There a Subgroup of Patients With Low Back Pain Likely to Benefit From Mechanical Traction? Spine, 32 (26) DOI: 10.1097/BRS.0b013e31815d001a
[7] Brennan GP, Fritz JM, Hunter SJ, Thackeray A, Delitto A, & Erhard RE (2006). Identifying subgroups of patients with acute/subacute “nonspecific” low back pain: results of a randomized clinical trial. Spine, 31 (6), 623-31 PMID: 16540864