Cancer drugs: when NICE says no

What happens when NICE says no? If NICE refuses to fund an expensive intervention to treat cancer, but the patient wishes it anyway, the patient must forgo all ‘free’ NHS care and pay for the intervention, plus all the rest of their care – ie be subsequently treated entirely as a private patient. Care then becomes very expensive.

There has been a lot of debate about this recently, with many arguing that this is wrong, and that patients should be allowed to pay for ‘top-up’ care for whatever additional treatments they may wish to have.

Freedom of choice is obviously desirable, but, as I’ve argued before, it has to be a meaningful choice. It is incredibly hard to look at newly generated evidence and to try and decide what it means for one as an individual. This becomes even harder when the data is 1) not free to access in its entirety 2) not yet peer reviewed (eg selected data being presented at a conference) 3) presented in ways where the best possible interpretation of the treatment is used (e.g. outcomes described as a reduction in relative risk rather than in absolute risk)  4) when the trial has been small scale and/or short term, which, among other things, may not be long enough or large enough for significant outcomes or adverse effects to be made apparent.

It gets even more complicated. As many before me have pointed out,  if clinical trials do not get published, the publicly available information about that intervention becomes skewed. Usually – in what is known as publication bias – we end up with evidence that suggests a treatment is more effective than it would appear were all the clinical research on it available.

There is a recent article in a US publication The Oncologist covering this subject. The authors examined what happened after clinical trials in cancer had been registered.  (The registration of trials is a relatively recent requirement for publication in most high profile medical journals. Registration of a trial means that there is a record of what trials are going on where. It is not a legal requirement to register a trial, but it is a marker of good practice.) The authors found that less than one in five studies in cancer that were registered went on to be published. They found that studies sponsored by the pharmaceutical industry were least likely to be published. True, it takes time to analyse research results and then to have them published – but out of trials registered before September 2004, less than one in four of these trials had ended up published in widely accessible journals. The authors had particular concerns that negative trials – where an intervention is not useful and has disappointing results – are not being published. This means that important information about treatments which are not effective is going unseen.

This means that it is not just very hard to make good decisions about treatment interventions, but approaching impossible. 

Organisations like NICE need to grow teeth. They need to be able to insist on all clinical data about an intervention to be made publicly available before they – or anyone else – can make a fair judgment on effectiveness.

It is my belief that all effective treatments should be free – but the question of what the effective treatments are has to be known first.

Margaret McCartney’s Blog

This blog is no longer updated but it remains open as an archive.

A forum on healthcare policy and professional issues, by Glasgow-based GP and FT Weekend columnist Margaret McCartney.

FT Blogs