The wrong end of the stick

An interesting paper in Science Translational Medicine reports that responses to treatment of induced pain with an opioid analgesic are heavily influenced by expectation. Briefly, if volunteers were told that they were getting the opioid,  their pain relief scores were twice as high as when they were told it was placebo. This was corroborated by brain imaging data. Should we be surprised? I am not. It’s a quite elegant study, but we have known for decades that pain above all else is heavily influenced by expectation. The problem lies in the way it was reported. The full paper is sadly behind a paywall, but this is so common that abstracts have assumed more prominence than they deserve. At the end of their abstract, the authors say:

On the basis of subjective and objective evidence, we contend that an individual’s expectation of a drug’s effect critically influences its therapeutic efficacy and that regulatory brain mechanisms differ as a function of expectancy.

Their use of the generic term “a drug’s effect” is unfortunate. It implies that this applies to all drugs, when it only applies to analgesics on the basis of these data. This over-generalisation was of course carried forward by the BBC’s report, for which this link label was used:

Pessimism ‘stops drugs working’

In an age of dumbing down, this was one of the dumbest I have seen. To be fair, the actual article did state in the title that it was about analgesia, but it presented the whole thing as a totally new finding when it isn’t. Indeed the authors of the paper didn’t acknowledge in the abstract the substantial literature on clinical trials in this field. Maybe they did in the full text, but I can’t read that. Does anyone else have access? I would be interested.

But my major exasperation is reserved for the comment at the end of the BBC article by Professor George Lewith. To say that

“It completely blows cold randomised clinical trials, which don’t take into account expectation.”

is about as wrong as it is possible to be. Now I am quite prepared to consider the possibility that a BBC editor will have got Lewith’s statement completely wrong, but I doubt it will have been that much wrong. Is not the use of a placebo control a way of dealing with expectation? Is not this study a “cold randomised clinical trial” its very self? This from a professor who is employed to run “cold randomised clinical trials”. To claim that one randomised controlled trial (RCT)  in a single therapeutic area invalidates the entire discipline of clinical trials is bizarre and not remotely supported by the evidence. But this study simply doesn’t mitigate in the slightest way against RCTs, because it is an RCT. Sorry, but in case Lewith is reading this I have to make it crystal clear.

It looks a lot like bandwagon jumping to me.