Sir Iain Chalmers - Interview 30

Age at interview: 62

Brief Outline: Iain is a keen advocate for well-designed randomised controlled trials to test treatments.

Background: Sir Iain Chalmers practiced as a medical doctor until 1973. Until recently, he was a director of the UK Cochrane Centre. He is now Editor of The James Lind Library (a website to help people understand why and how treatments are tested). He is married with 2 adult children. Ethnic background/nationality' White British

Iain explains the worry that screening tests may raise anxiety, especially if it is unclear what...

Well I’d be worried first of all about the side effects of a positive test, and the biggest side effect is probably anxiety. And it’s, I think, probably not sufficiently taken into account that some people, and I’m one of them, can be quite anxious faced with the result of a test which isn’t certain, and then go on to have other tests which are more invasive, if you like, than a simple blood test. So people need to know about that, but in particular they need to know well if it does turn out that this test identifies a cancer that’s there, a real cancer that’s there, what are the treatment options and what difference do they make to, for example my life expectancy, as one example, and what are the side effects of the treatments as well. And when not enough is known about those one can find oneself being treated for a disease and being labelled as being a cancer sufferer if you like, a good deal earlier than might otherwise have been the case and you want to be very clear that that’s got advantages as compared with, for example, going to your doctor when you’ve got some troubling symptoms, and then the doctor examines you and does things in response to the fact that you’ve gone to the doctor rather than the doctor has offered a screening test in spite of the fact that you appear to be well.

 

I mean there are some conditions, blood pressure, raised blood pressure is a good example, where you may not have any symptoms and because there are important and effective treatments for them it is a good idea for a doctor to check your blood pressure and see whether it’s raised and if it is, may be to do some further investigations and to invite you to consider going on treatment. Because we know that treating high blood pressure can reduce your risk of stroke, which is a very serious condition. But the situation with PSA [prostate specific antigen] screening is not like that yet, may be it will become that, but the evidence that’s so far available doesn’t put it in that class of a screening test at all.

Treatment for prostate cancer can have unpleasant side effects. Because the best way of treating...

Would you like to go back to prostate cancer, what is the evidence at the moment about treatments?

 

Well one of the reasons that there is uncertainty is that prostate cancer is quite common, far more common than people realise, but it doesn’t kill people. In other words, you can die with prostate cancer but not of prostate cancer. So it’s quite important, given that fact, that one is very clear about the quality of the evidence upon which you intervene early with some of the quite radical things which people are intervening with. For example radical prostatectomy, quite a major operation, to remove a prostate with some cancer in it, has side effects. It can cause incontinence, men can become impotent as a consequence of it and then of course it has the risks associated with any large operation, those associated with the anaesthesia or bleeding and so on.

 

And then there is treatment given with radiotherapy as an alternative to that. There again there are problems which can be associated with the radiotherapy, so some people say, even when they know there is cancer there, well actually let’s just watch the situation and if it does seem to be deteriorating fast only then intervene with one of these more radical treatments. And that uncertainty is currently reflected in the fact that men are participating in this country in a very big controlled trial comparing those three different options, the radical surgery, the radical radiotherapy and active monitoring of the situation. And as a result of that study and other similar studies, although I think the British one is probably the biggest and the best, people like me are going to be in a far better position to take an informed decision if we find ourselves in the circumstances where we do have an early prostate cancer diagnosed. And indeed, if I went to my doctor with symptoms and it did lead to a diagnosis, I think I would want to be invited to participate in that controlled trial. It’s a way of hedging my bets because we don’t know which is best, but also helping to produce the information that will make decision making in future by people in my position more informed than it would otherwise have been. 

Iain explains why he thinks randomised trials are the only ethical way to make progress when we...

Someone who's properly informed about PSA screening [for prostate cancer] should know what the treatment options are before they go in for it. I think that's part of the information they need.

 

And the side effects of the treatment presumably?

 

Absolutely, yes. Doctors who are paid on a piece work basis are paid for every radical prostatectomy that they do. You can imagine that some doctors are very keen on maintaining their income by doing lots of radical prostatectomies. Now if there was really good evidence that these operations actually did have an impact both on the duration of life and on the quality of life - to the extent that individual men would consider important - then we would be in a better situation than we are. As it is, we don't know. We do know that the operation can sometimes cause incontinence and impotence. I imagine that incontinence is going to be a distressing complication for most men; impotence some will be bothered about it, others won't. But those things have to be set against whatever advantages the operation has to offer and I think they're far less clear than some people are led to assume by those promoting these operations.

 

In the current state of knowledge I think it's unethical to offer treatment outside the context of randomised trial. Why should it be ethical to give a treatment the benefit of the doubt because doctors think it's useful, but also because they're actually going to get paid on a piecework basis for the treatment that they're offering? Why should that be ethical when there's no strong evidence to support the treatment and yet treating people within the context of a controlled trial not being ethical? The whole thing is completely upside down.

 

So if you were a urologist and a man came to you with an enlarged prostate at the moment, what would you recommend that he did?

 

If systematic reviews of controlled trials had shown that, in the circumstances in which that particular patient found themselves, there were useful treatments, useful in the sense that at least they had been shown to have some beneficial effects, even if there might be side effects too, the first thing to do would be to share that evidence with the patient. If on the other hand their condition was such that there were no clear answers, then the patient should know that. And if there was a randomised trial as an option for offering treatment, then I would want to suggest to that person that they might like to participate. Faced with uncertainty, the most appropriate way forward is to deal with the uncertainty by contributing to efforts to reduce the uncertainty, and that's certainly what I'd want for myself. I carry around a card with me saying 'invite me to participate in any randomised controlled trial for which I am likely to be eligible'. I don’t do that out of any altruism; it's purely selfish.  

Iain thinks a random sample of the public should be asked for their view of the evidence before...

If you want to introduce a screening test on a population basis - as a policy - then two thoughtful authors, Paul Glasziou and Les Irwig, made an interesting suggestion in an article published last year [2000]. You go to a random sample of the population and you invite that random sample to look at the evidence and to say whether that they think that this screening test should be applied to the population of which they are the representatives. If they come out with a clear ‘Yes’, then that gives you the legitimacy to go ahead and offer the screening test on a population basis. If they come out with a clear ‘No’, then that also gives you a clear direction about what you ought to be doing on behalf of the whole population. If there's a real mixture of views and no clear cut off at all that allows you to say that a majority of people are in favour, that obviously means that the information is sufficiently uncertain or equivocal that more information is needed before you have a license, a mandate, in other words, to approach a whole population with a screening test. I think that's a very good suggestion and I don't know whether it's ever been tried.