Thought-provoking posts on science, health and beliefs

The blogosophere is pretty hot right now – some really interesting topics being discussed, enough to make me think again about my own biases.

The first one is the popular Science-Based Medicine blog that rarely fails to challenge those that prefer ‘belief’ over ‘evidence’ – do not proceed to this blog if you think homeopathy or reiki is ‘good for your health’, this blog does not mince words!  Three posts have particularly caught my eye recently – this one on some of the reasons people (even scientists!) find it hard to change tack when new evidence is found; this one on the misinterpretation of acupuncture trials; and this one on the role of adenosine in pain relief.

The first post really struck me as I look at my own bias towards nonbiomedical interventions – yes, even I can be tempted to ‘overlook’ studies that show equivocal findings for CBT-based approaches, or to focus on studies that show positive outcomes… I remember my initial concern about mirror therapy and graded motor imagery for CRPS – and while I still hold concerns that this treatment doesn’t seem to reduce pain and increase function in many people (notably the complex presentations we see at Pain Management Centre), I do think it holds promise for people with CRPS now that the evidence is gathering.  That’s science you see – revising what we accept in the face of accumulating evidence.

Movin’ Meat is the rather curious name of a blog written by an ER physician (and administrator, bless him).  This blog is often about    but this post is a repost and hilarious.  Do not read if you’re easily offended, but given that people with chronic pain DO have a sense of humour (how would you survive without it?!), this post on a Brand New Pain Scale hit my funny bone – read and enjoy!  After my experiences last year with tonsillectomy, I totally appreciate the need for a ‘beyond what I want to cope with’ number on a pain scale.   I’ve always thought that putting a number to my pain is a bit silly, wouldn’t it make more sense to start with ‘no pain’, move to ‘a bit of pain but it’s not bothering me’, go from there to ‘yup, this is pain and I’d like it to go no’ and finally ‘I’ve had enough, MAKE IT GO AWAY’.

And lucky last, but certainly not least, is BodyinMind blog and the rip-roaring debate going on there about exercise and back pain, and the most recent discussion about acupuncture and a critical review of some of the most recent papers on it in pain management.

If you’re a Facebook participant, head to the very informative and rational page Neuroscience and Pain Science for Manual Physical Therapists.  If you click on this link and you are not a member of Facebook, beware, you won’t be able to get in.  If you’re clicking on this at the Canterbury District Health Board, forget it, Facebook isn’t available.  Otherwise, click away and enjoy the discussions on a whole range of neuroscience and pain management topics.  Some fabulous people and over 2500 fans have joined this page and comment regularly – makes Facebook worth visiting just for this alone!

I hope you enjoy these links – let me know what your thoughts are!


  1. I am a scientist, and I think evidence is very important. So I typically research evidence for treatments my doctors offer me. Over time, I have come to a conclusion that trials matter, but personal experience matters as well even if numbers don’t support it.

    An interesting case in point is the use of TENS machines. A recent Cochraine review says “In summary, the review authors found conflicting evidence regarding the benefits of TENS for chronic LBP, which does not support the use of TENS in the routine management of chronic LBP. ”

    Now, a TENS machine is something I use, very effectively, to reduce the need for painkillers. So I know it works for me, and since the review said that there isn’t even consistent evidence of pain relief, I dug deeper. Turns out that the data reviewed included trials with treatments ranging “20minutes to 3 hours per day”. This resolves the surprise. If I were only to use TENS for 20 minutes a day, it would be ineffective. NHS instructions I received from my physio say to wear the machine and turn it on and off as needed, and expect about 30-45 minutes of pain relief after TENS is turned off. Manufacturer’s instructions say the same. So the trials which used 20 minute per day treatment, or even an hour per day treatment, do not provide a valid evaluation for practical use. But you would not be able to tell this from the summary, which is all many lay people, and presumably busy medical professionals, will ever read.

    This is just one example – I found a couple others as well in my reading. My conclusion was, numbers can lie. For anything other than straightforward drug regimens (CBT, physical therapy, acupuncture, exercise…) one has to be very careful about the way studies were run, and whether they will apply to your particular situation. And personal experience matters, even if evidence doesn’t match. So if you have doubts about mirror therapy, there may be a good reason – perhaps it indeed does not work for presentations that you see. Knowing when a therapy is applicable is important, too, especially given that statistically significant effects may mean something like “39% people improve” (which is the last one I read on CBT for pain – things may have moved on).

    This is not to say that evidence is unimportant. I still do my research 😉 But I find places like Science-Based Medicine blog go too far the other way – they seem to think that trial outcomes are all that matters, and anyone who dares believe something not supported by science is deluded at best.

    1. Hi Mary
      It’s true that the Devil is in the details! And even systematic reviews are grouping data together so there is some loss of individual variation that would normally be present. The EBHC movement acknowledges that while meta-analysis and systematic reviews collect information so as to find large effects at the same time, individuals are individual and it’s always up to the clinician and the person to work out what works for this person at this time.
      Despite this, I’m more in favour of starting from the treatments that do have robust support from RCT’s and then progressing to less strongly supported approaches, than the other way around.
      It’s also really important, as you say, to read the details of the methods used. In the case of TENS, I wonder how the NHS and the manufacturers came up with the instructions you got – and why they differ from what was researched. I’d hope that NHS and the manufacturers did some research before coming up with their instructions!
      Regarding mirror therapy – I know that the people I see are probably on the more severe and chronic end of the spectrum, and there could well be other mediating variables influencing how they respond – so I keep my eyes open and look at the studies as they come through. It’s something I’m quite relaxed about, because data is independent from me and my opinion! Over time the evidence gradually accumulates. It’s just when strong evidence is ignored that I start to get titchy.
      Thanks for taking the time to comment – it’s great to know there are people who question their treatments and want to know how and why things work!

  2. Oh, I agree completely. When I was looking for options for my own treatments, I started with well supported ones and progressed to those which have less evidence when the “gold standard” didn’t work. I believe that at a minimum every treatment has to be evaluated for negative side effects – I would not try anything that hasn’t been in at least some clinical trials and established reasonably safe. I have, sadly, seen too many people do outright dangerous things claiming that science is irrelevant.

    It’s an interesting question about the TENS instructions. As it happened, when I was doing my research on TENS, I haven’t came across the Cochrane review, which may be just as well 😉 I was going by various reputable medical articles explaining how TENS operates, plus experience of several friends who used TENS in labor and for osteoarthritis, and found that it worked for them. The explanations for TENS that I read all said more or less the same thing: the current interferes with pain signal, and therefore provides temporary pain relief. The instructions I read were consistent with that theory – the idea is, you wear it all day, and turn it on when pain gets bad enough so that the tingling TENS sensation is actually preferable to the pain. I found it worked exactly as advertised. I did find some sites making more general claims about acupuncture points, or about long-term relief after consistent use. They didn’t seem believable given the general TENS theory, so I just ignored them. In a sense, the trials prove that the those wider claims are untrue, but I still find it surprising that the review did not look at the disparity between standard medical information and trial conditions.

    Part of the problem may be the difficulty with setting up proper trials. The review wanted “high quality evidence”, which required blinded trials. I think this again is nearly impossible with TENS. The way to make blinded trials is to issue some people with the machines that don’t produce any current, and tell everyone to set machine dials at the same setting. But this again goes against standard usage instructions, which say “turn current up to the highest level of sensation you can tolerate”. This again is based on the theory of electrical signal interfering with pain, but of course is not “blindable”. So it is possible here that the need for accepted and verifiable scientific methodology (i.e., blind trials) here interfered with proper evaluation. I think many other interventions are equally affected, including exercise and acupuncture discussed in the posts you mentioned.

    I work in another field (educational technology) where statistics are used to evaluate effectiveness. So I am well aware of both the benefits of collecting evidence and the traps that go with it. I think science still has some ways to go to come up with objective evaluation methods for inherently individualized interventions. But, as you said, ignoring mounting evidence is not an answer as well – we just have to live balancing the collective evidence and individual experience, and searching for what works.

    1. Implementing a completely blinded study with sham controls is really difficult, although recently with acupuncture I believe a truly sham acupuncture intervention has been developed (a patch that looks exactly the same as the ‘active’ acupunture patch but doesn’t pierce the skin). De-tuned TENS has been used, I know, but I’m not very au fait with TENS as an intervention. In terms of CBT vs a ‘control’ condition, most of the studies use ‘attention’ or just talking to people about a ‘general’ topic unrelated to pain or CBT as the control – but because one of the ‘active’ ingredients in CBT could be the therapist’s own persuasiveness (even if they don’t mean to persuade!), this again may not be terribly effective as a control.
      So, you’re right, there are loads of confounds in RCT’s, one of the most difficult is to identify an appropriate control condition. In terms of the ‘standard useage instructions’, I do wonder whether there is much standardisation of instruction – at least in NZ there are so many variations as to how to teach someone to use TENS, it could be a pretty hit and miss affair if you got a ‘standard’ instruction!

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.