There is a way of testing new medicines that allows inferior drugs to come on to the market. I don’t mean inferior in the sense that the drug doesn’t help as many people as an existing medicine; I mean that it can be linked with more deaths than an existing medication, and still be approved.

An excellent paper in The Lancet recently explained how non-inferiority trials work, why they are unethical and why they should be banned. I relay its contents because it is quite possible that you are taking a pill that has travelled this dubious route from laboratory to medicine cabinet.

As the name implies, a non-inferiority trial is designed to prove that the medicine being tested is no worse than an existing medication. The added value is thought to lie not in its effectiveness but in some other aspect: for example, perhaps it needs to be taken once, rather than thrice, daily.

The problem, according to Professor Silvio Garattini and Dr Vittorio Bertele, from the Mario Negri Institute for Pharmacological Research in Milan, arises in the definition of “non-inferiority”. The drug under investigation – let’s call it M – must produce results that are similar to a standard treatment within a prescribed limit.

If this limit is set at 7.5 per cent, it means that a death rate of 7 per cent among the patients taking M is acceptable, even if the death rate among those on the standard treatment is 5 per cent. In a trial of 1,000 patients, that amounts to an extra 20 deaths with M – yet it could still receive a regulator’s blessing. “To expose patients to such risks in the trial, and in real life, with no benefit in exchange, is clearly unethical,” Garattini and Bertele fume.

The only beneficiaries of non-inferiority trials are drug companies, because of lower research and development costs; proving that M is noninferior to the standard can be done more quickly and with fewer patients than proving its superiority. The authors ask: “Why should patients accept a treatment that, at best, is not worse, but could actually be less effective or less safe than available treatments?”

Why, indeed?

–– Last week I received a press release about “the science of wrapping”, including a formula created by a Leicester University student that supposedly calculates how much paper is needed according to the dimensions of the gift. My thoughts on this “revelation” are unprintable.

So, thank you to the Laboratorytalk blog, which offers a more measured appraisal: “(Warwick) Dumas’s discovery . . . is so shallow as to be laughable. . . The whole thing heaps embarrassment and shame upon him, the university and science in general.” The formula was commissioned by a well-known shopping centre. The only thing it proves, Mr Dumas, is that some gift horses are worth looking in the mouth.

Times Recommends

• It’s not just about Iraq, it’s about Blair’s style
• You can’t rank hospitals like football teams
• Minarets are not an essential part of Islam