Recently my girlfriend has started doing hypnosis therapy before going to sleep. For the sake of privacy, let's assume it's to cure her of her shameful enjoyment of farting in crowded spaces, and I was curious to know whether it's actually helping or not. In fact this is one of many dubious treatments she is trialling in the attempt to rid herself of this embarassing perversity, and I was unsure of which to tackle first - next week CBD!
Before I continue, I should be quite clear that I wasn't very hopeful from the start. I expected either little to no effect, for the simple reason that if hypnosis really did work in the same way that e.g. penicillin does (did?) then we would be prescribing it en masse to change people's behaviours, health related or otherwise. That last one is a bit scary actually, so now I'm hoping that hypnosis is ineffective. Facebook et al are doing a sinister enough job of changing our habits thank you very much.
First thing I wanted to know was whether hypnosis, as in the kind that is subjected to Top Gear presenters, actually worked. Turns out that googling this is quite unhelpful - I don't trust many of the results (sorry WebMD). So I went for the more specific search of "Cochrane meta analysis hypnosis".
This proved much more fruitful. A meta analysis of hypnosis for "smoking cessation" (i.e. quitting) with 1926 subjects in total found limited to no effect with low certainty on the results. Poor trial quality and low number of participants was a common theme among these meta analyses in fact. Another meta analysis on hypnosis for treating depressive symptoms supposedly concluded a positive effect, but this seems fishy given the low number of total participants (258) and trial quality. Finally, a meta analysis on hypnosis for insomnia similarly reported positive effects but the authors were honest enough to state they were uncertain of the results given the low number of participants and poor trial quality.
It was at this point where I chanced upon this Vox article which reported on hypnosis and weight loss. First of all, it is hard (impossible perhaps?) to double blind hypnosis trials, since double blinding requires that both the hypnotizer and hypnotee are ignorant of whether they're giving or recieving hypnosis treatment. This makes it difficult to discern whether the placebo effect is at play or not. Many of the meta analyses cited in the Vox article come from one Irvin Kirsch, who has also published a paper in 1994 (in the American Journal of Clinical Hypnosis no less!) about hypnosis as a "non-deceptive placebo". I have no idea what that means, but it sounds interesting.
In case you're wondering, when it comes to hypnosis and weight loss I'm not sure. As the Vox article points out, an initial meta analysis from Kirsch suggested it did slightly, then a rebuttal claimed otherwise. To this Kirsch published a third analysis (lovingly sub-titled: "another meta-reanalysis") which again claimed increased weightloss when hypnosis therapy was given. I don't know if there was a follow up, but I like to think that everyone got bored and started researching the relationship between petanques and drub abuse.
So that's that. I know I should have some sort of cool summary here, but I've already spent 3 hours doing this and I honestly have better things to do. Like watching Borat again.
 To be precise, these meta analyses were investigating "Hypnotic enhancement of cognitive-behavioral weight loss treatments". I should also add that it seems that losing weight is easy - it's maintaining that weight for 3, 6 months, several years after which is the hard part.
First written 23rd of January 2021