Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

On being little (Editorial)

On drawing a line
On the net

Roman augurs never had it like this. They could tear the entrails out of a chicken and make of it what they liked, or perhaps what they were paid to make out of it. After all, taking the auspices was one thing, what happened was another, and if the two did not match someone else was to blame for offending the gods.

One of the classic problems with evidence is when there's not much of it, and we have to tease out what it means. The trouble is we are not allowed to guess. We have to get it right. So this month we concentrate on looking at little amounts of evidence.

Three meta-analyses, of treatments for impetigo, methylxanthines for COPD exacerbations, and trazodone for erectile dysfunction make the training ground. One confirms us in our beliefs, though it appals us that there is so little evidence for treatments used so often. One shows that guidelines can be wrong, and that a treatment probably does more harm than good. One teases us with suggestions of efficacy, but only in some, and then perhaps.

And that is what it is often like, taking the auspices from too little information. We have, in the end, to make much from little.

On drawing a line

Another venture into homeopathy, with a systematic review of systematic reviews. Homeopathy doesn't work. It is time to draw the line. Some will call for more trials, but Bandolier cannot see why its taxes should pay. Surveys show that huge amounts of money are taken from the public for alternative therapies with little or no evidence. Make them pay, as happened in Holland where a "loser pays" deal was struck.

And they will. Hail the manufacturer of magnetic insoles for foot pain that funded a randomised trial showing that they were no better than non-magnetised insoles. Don't weep. The US public spend $500 million a year on magnetic products for pain.

On the net

next story