Double blind seems to be one of those jazzy terms we get excited about when reading a paper because it boosts our faith in its credibility. This is when you perform an experiment where both the tester and the participant are unaware of who is receiving treatment and who is a control – ultimately it aims to reduce bias. Intuitively double blind studies seem completely necessary when testing placebos – but are they? Many don’t think so and are doing just the opposite – that is, an experiment where both the tester and the participant know who is receiving a placebo. This is called an open-label design and it really had me stumped. This summation explained my perplexion well …
1. Find people in pain.
2. Enrol them in a study.
3. Admit you can’t do much to help.
4. Give them a fake pill.
5. Tell them that’s exactly what you are doing.
Perhaps, like me, you are new to this term ‘open-label’. Perhaps also, the idea of a open-label placebo study seems slightly oxymoronic. My understanding of placebo was that it involves some sort of trickery such that the patient thought they were receiving the experimental treatment, when in fact, they weren’t. I thought the aim of placebo testing was to assess the power of the perception of receiving a treatment, compared to actually receiving it. How then can you test this perceptual illusion if the patient is aware?
It turns out my understanding of placebo was somewhat misplaced (I console myself in the belief that I don’t think I’m alone here). It turns out deception is not necessary. To provide a placebo is to provide a treatment known to be inert or ineffective – like a sugar pill. The fact that most studies deceive, does not necessitate the need to do so.
Why run an open-label placebo trial?
These open-label trials often come after it has been shown that placebo is just as beneficial (if not more) than a certain treatment. Then comes into question, why does the placebo work? Some believe that it is due to the psychosocial effects of the therapeutic encounter – its interactions, rituals, and symbols. If this is the case, deception shouldn’t be required.
Open label studies also address the ethical conundrum that practitioners face when using treatments that are thought to work because of placebo effects. You see, concealing the knowledge that you are giving a placebo from a patient violates the ethical principles of respect for patient autonomy and informed consent. But if the placebo treatment works without having to lie about it – problem solved!
A few studies have shown positive results even when the placebo’s inert content was divulged to the participant – irritable bowel syndrome, major depressive disorder and ADHD. To add to this line-up, a randomised controlled trial published in Pain in October 2016 (open access) has investigated this phenomenon for those with chronic low back pain.
The test: Take 83 people with chronic low back pain who responded to a flier for “a novel mind–body clinical study”. Tell them that half will be given a placebo – an inactive substance like a flour pill, that contains no active medication. Explain that placebo can be powerful, that the body can automatically respond to it (like Pavlov’s dog), that a positive attitude is helpful but not necessary and that they must take the placebo pill faithfully for 21 days. Then randomly give half the pill (remember to write ‘placebo’ on the bottle) while the other half continue treatment as usual. Three weeks later check for changes to pain and perceived disability.
Are the results clinically relevant?
Turns out the placebo group recorded significantly lower pain scores (Numerical Rating Scale) and lower self-percieved disability (Roland-Morris Quesionnaire). Keep in mind that the pain score reduction was just over 1-point (0-10) and the disability quesionnaire dropped just under 3 points (0-24). While these may seem low for clinical relevance, they were both significantly better than the control group, and importantly had no side-effects or cost.
What, why, how …
There must be some mechanism at play that can explain why honest placebos work. Some offered by the present study include:
This all occurs in the presence of a postive practioner-patient relationship and rationale which may be convincing enough to allow participants to suspend their disbelief.
Participation implies a belief or hope that the treatment might work – and engendering hope when participants feel hopeless about their condition can be therapeutic.
Non-conscious processes actively contribute to placebo responses (i.e. twisting the bottle and swallowing) – but bear in mind, 90% of the participants in this trial were taking pain medication (primarily NSAIDs) before and during the trial.
Spontaneous fluctuations in pain might be interpreted as evidence that placebo is working, thereby strengthening expectations of relief.
So what do you take from all this? Obviously an open-label study introduces a lot of bias compounded by the fact that the outcome measures were subjective. But can placebos really work without the deception? Or have they still been deceptive by describing to the participant the powerful, beneficial effects of placebos prior to testing (after recruiting those interested in a ‘novel mind-body’ study)?
– Hayley Leake
THE 2017 NOI CALENDAR IS SHAPING UP, HERE ARE THE CONFIRMED DATES
Melbourne 31 March – 2 April EP and GMI
Adelaide 26-28 May EP and GMI
Darwin 4 – 6 August EP and GMI
Brisbane 25 – 27 August EP and GMI
Newcastle 8-10 September EP and GMI
Details and dates coming soon for Wollongong and Sydney
Check out our courses page for details and to enquire
HAVE YOU DOWNLOADED OUR NEW PROTECTOMETER APP YET?
Just search the App Store from your iPad for ‘Protectometer’