Shoe inserts for Sacroiliac Joint Pain?

It is always a pleasure to comment on some good research that really has the potential to change clinical practice; its just a shame that bad science keeps me distracted from commenting on the good science. Journals have a responsibility to not publish bad science, so hopefully no one does bad science. As much as I don’t like doing it, I am always going to call out journals and their peer review process when bad science gets published as they need to be held accountable. I was going to ignore this one as it was in a low impact factor (0.39) journal; the abstract was posted on Podiatry Arena, but then a friend tweeted about the study and sent me the full paper, so here it is:

The effect of gait training with shoe inserts on the improvement of pain and gait in sacroiliac joint patients.
Journal of Physical Therapy Science . Aug2015, Vol. 27 Issue 8, p2469-2471. 3p.
Purpose: The purpose of the current research was to identify how gait training with shoe inserts affects the pain and gait of sacroiliac joint dysfunction patients.
Subjects and Methods: Thirty subjects were randomly selected and assigned to be either the experimental group (gait training with shoe insert group) or control group. Each group consisted of 15 patients. Pain was measured by Visual Analogue Scale, and foot pressure in a standing position and during gait was measured with a Gateview AFA-50 system (Alpus, Seoul, Republic of Korea). A paired sample t-test was used to compare the pain and gait of the sacroiliac joint before and after the intervention. Correlation between pain and walking after gait training with shoe inserts was examined by Pearson test. The level of significance was set at α=0.05.
Results: It was found that application of the intervention to the experimental group resulted in a significant decrease in sacroiliac joint pain. It was also found that there was a significant correlation between Visual Analogue Scale score and dynamic asymmetric index (r= 0.796) and that there was a negative correlation between Visual Analogue Scale score and forefoot/rear foot peak pressure ratio (r=-0.728).
Conclusion: The results of our analysis lead us to conclude that the intervention with shoe inserts had a significant influence on the pain and gait of sacroiliac joint patients.

Pretty much what they claimed to show was that the group wearing the inserts got better and the group that did not wear the inserts did not get better. Seems pretty straight forward doesn’t it? However, like the study I reviewed last week, I question how papers like this get through the peer review and editorial process with flaws and inadequate information. There are just too many red flags to trust the results:

  • Firstly, the information provided does not come close to what is required by the CONSORT guidelines.
  • Secondly, like last weeks review, they used a within groups analysis rather that a between groups analysis. What is the point of having a control group if you do not statistically compare the outcomes of the intervention group with the control group?
  • The authors claim to have randomly selected the subjects – how do you do that?
  • The authors claim to have randomly assigned the subjects to the intervention and control groups, but ended up with exactly 15 in each group. While this can happen randomly, one thing poorly quality studies tend to have in common is the randomization process ends up with them having the exact same number in each group. The authors really need to explain how they allocated the participants to the groups to ensure they were properly randomized.
  • From the study, you have no idea what the actual intervention is! Was it a ‘foot orthotic’? If it was, what were the design features? Was it a flat insert to correct a leg length difference? They stated: “shoe inserts made of polypropylene material (Alpus, Seoul, Republic of Korea) “. I Google’d Alpus and found nothing, so that was not of much help. The whole point of the descriptions in the methods section is sufficient detail is given so that the study could be replicated. Not going to happen with this study, as we have no idea what they used, let alone use the intervention clinically if we want to try it.
  • Unlike last weeks review, they did describe the characteristics of each group (age, height and weight) so the baseline characteristics of each group can be compared, but they did not do a statistical tests to compare those characteristics. This is important as the weights of the two groups do appear to be different.

This simply means the results of the study can not be trusted, as much as it would be nice to know that this is a potential useful intervention for sacroiliac pain. The comments I made above should have been addressed in the pre-publication peer-review process and either the manuscript corrected to do the right analysis and provide the right information or it should have been rejected. Bad science should not be published. It is disappointing and distracting from good science to have to make these sorts of comments. Journals have responsibilities for improving the quality of the science.

As always: I go where the evidence takes me until convinced otherwise, and this is not evidence for using “inserts” (whatever they were) for sacroiliac problems.

Cho, B., & Yoon, J. (2015). The effect of gait training with shoe inserts on the improvement of pain and gait in sacroiliac joint patients Journal of Physical Therapy Science, 27 (8), 2469-2471 DOI: 10.1589/jpts.27.2469

Last updated by .

4 Responses to Shoe inserts for Sacroiliac Joint Pain?

  1. Rafael Pimenta October 2, 2015 at 2:05 pm #

    Perfect Craig!
    I know some Editors from good journals and they are very worried about it! Do bad science is a big problems because not everyone who read this kind of paper have the same knowledge of you and can´t distinguish between what is good and not! So, keep working hard! You are the best! Cheers from Brazil.

    • Craig Payne October 2, 2015 at 8:24 pm #

      …and I have some more bad review coming where the editors and peer review process really slipped up

  2. Ivan Rivera October 20, 2015 at 10:17 pm #

    “Sacroiliac joint pain” seems a little vague to me. I couldn’t get to the article, but do you know if they are they selecting a group that has sacroiliac joint pain from a particular cause (muscle imbalance X), or via a particular mechanism (impingement)?

    (I assume not, from the 5th bullet point).

    It seems like a bad study to cite as evidence for either side of the argument. Thanks for the heads-up.

    • Craig Payne October 21, 2015 at 12:31 am #

      I should have mentioned that as another bullet point! The points I made above focused on methodological issues; but their selection criteria for “sacroiliiac” pain was not that specific and vague, so is an issue.

      I would have preferred a more specific selection criteria in terms of diagnosis.

Leave a Reply