Here's my take on this. I'm an active researcher with a PhD in molecular biochemistry and a chronic tinnitus sufferer, so I think I offer a pretty balanced, critical view on this type of research.
The overall take is good, this appears to have a much greater effect than any intervention published previously, with almost 90% significant improvement among those who completed at least 5 treatments.
However, there is some issues with the data, many of which the authors recognise themselves in the discussion so I won't repeat that. My main gripe is the scale used has broad intervals, meaning 8 vs 6 is a marginal difference in perceptional but ensures the change will be statistically significant. Anyone with chronic tinnitus knows, how you would rate it day to day changes dependent on an array of factors so a superficial scale like this is not accurate enough for this purpose, imo.
Likewise, the way the data is displayed in the final graph is very striking, but also strange. Despite reporting distinct SD values in the table, all the error bars on the graph are the same size. The sample size being only 50 odd, this doesn't make sense. Not to mention no controls, etc. It would of been far more telling to display all the individual data points in a scatter.
I notice, although they report the parameters for how they selected patients as appearing a good unbiased approach, they also say ALL of the patients had previous nerve blocking procedures. I wonder how that effected results and what other treatments these people had already received.
I tried to find info about where this was done, it's in South Korea of course but it does not say where the procedures were performed. It appears to be the lead author who has performed the treatments, who is linked to a private pain clinic in their own name. I can't find any info about this clinic other than an address, which appears to be a one man back street chiropractor kind of affair. 3 of the 4 authors have the same (surname?). I'm not familiar with how South Korean names are organised but does this mean they are related, which could suggest a bias as the senior author is registered at a different institute through which this was published.
Lastly, this is published in Frontiers Neuroscience. Now, some Frontiers journals are better than others, but as someone in the trade, I can tell you Frontiers is the go to place when you've got weak data / want to avoid rigorous review. I have several papers in Frontiers journals with higher impact than their Neuroscience journal and can tell you it is incredibly easy to get published. On this paper, there is only 2 reviewers, both in the UK. Given the treatments were performed between 2017-2018, and only just now got published, and it's a very small data set which wouldn't have taken very long to process, I can all but guarantee this has been doing the rounds at better journals and getting rejected, eventually ending up in Frontiers. A paper like this does not take 3-4 years to submit and publish.
In summary, I'm excited by the results and it absolutely warrants a properly conducted clinical trial, and I hope the authors are trying to make that happen. However as it stands, I won't be flying to South Korea to get needles stuck in my head, just yet.
Member
