THE CONCEPT IN CARDIOLOGY OF “LIVES SAVED” IS TOTALLY MISLEADING
It Inflates The Benefits Of Drug Treatment and Is Intellectually Dishonest
By Red Flags Columnist Dr. Malcolm Kendrick
(email - firstname.lastname@example.org)
How is it that cardiologists save lives, when oncologists only manage to increase median survival? Okay, not the snappiest question ever. But you try engaging in a discussion on median survival rates without creating instant narcolepsy.
Here is another question for you. If I were to use a defibrillator on a man having a heart attack, and I get his heart beating normally again, have I saved his life?
Is this a trick question? In a way. Perhaps you would say that it all depends how much longer he lives. If he only lives an extra five minutes, then dies, did I save his life? No, yes, maybe? Did I increase his lifespan? Yes, of course, if only by five minutes.
What if he lives for another year? Did I save his life?
The reality, of course, is that you cannot actually stop anyone from dying. You can only give them more time. How much more? Well, you’re never going to know unless you keep track of them for the rest of their life.
‘So what,’ I can hear you cry. So quite a lot, actually. Because cardiologists talk about saving lives all the time. Superficially this may seem reasonable. However, in reality, this represents a statistical sleight of hand that results in a massive distortion of the results of clinical trials, stretching benefits to breaking point – and beyond. It makes the bogus use of relative risk figures fade into insignificance. You probably think this is a bit of an exaggeration, but bear with the argument.
Moving back to oncologists for a moment. They recognised early on, I am not quite sure how or why, that their drug treatments rarely cure cancer (a moment of unexpected humility from the medical profession). Anti-cancer agents zap tumors, but they most often do not get rid of the cancer altogether, as it often returns.
Therefore, in most cases, all that a new drug treatment in cancer can achieve is an increased length of survival. Thus, when they do a clinical trial on a new drug treatment, oncologists normally report their studies according to the increase in median survival. Median survival means the time at which fifty per cent of patients are still alive – or fifty per cent have died.
Why don’t they use the average? Because, gentle reader, if you want to know the average lifespan, you have to wait until almost everyone has died before you can do the calculation. As some patients may actually be cured by the treatment, you might have to wait for fifty years before they die. Which is a bit long for any clinical trial.
Anyway, in cancer treatment, if they did a trial in which one hundred more of the ‘treated’ patients were alive at the end of, say, two years, no-one would dream of claiming that one hundred lives had been saved.
Yet cardiologists have no hesitation in claming that lives can be saved. The home page of the Heart Protection Study (HPS), starts with the quote, ‘Tens of thousands of lives could be saved each year by changing prescribing guidelines for statins,’ says Dr Rory Collins, lead investigator of the HPS study.
The implication here is that every extra person who has not died, has been cured, and will live out a full and healthy life. But this is complete nonsense. The reality is that drug treatment, at best, slows disease progression, and may give you some extra time.
Therefore, when Rory Collins says ‘Tens of thousands of lives could be saved, what he actually should say is that, tens of thousands of people may have their lives extended by about six months to a year by using statins (and only if you treat more than one million people).
Why doesn’t he say this? I suspect you may already know the answer to this question. He doesn’t say this because – if you claim that you are saving lives – the benefits sound fantastic. However, if you claim that you are merely delaying disease progression – it seems rather less wonderful.
I will just try to feed some real figures into this theoretical discussion, to give you a better idea of the level of distortion achieved.
A study published in May 2003, in the Journal Of The American Medical Association,
looked at all clinical trials on blood pressure lowering. Nineteen of these trials looked at drug treatments (various) versus placebo. There were 42,972 patients in all these trials and they lasted, on average, three and a half years or so. In total, therefore, these trials represented 133,741 years of drug treatment.
Adding the results from the trials together, there were 1453 deaths in the placebo group, and 1303 in the treated patients. Which means that there was a grand total of one hundred and fifty more people alive in the treated groups.
Now, if each of the one hundred and fifty people who survived was ‘cured’ and lived an extra thirty years this wouldn’t be too bad a result. I don’t think.
To work out the exact benefit, you could do a calculation as follows:
150 lives saved x 30 extra years = 4,500 extra years of life.
As it took nearly one hundred and forty thousand years of drug treatment to achieve four and a half thousand extra years of life, this means that each extra year took about thirty years of drug treatment. Not brilliant, but not too bad. Although I don’t think it would convince me to take a tablet each day, for the rest of my life.
If, however, you know that no-one was actually cured, and all you have done is delay death by about a year (max), the equation looks rather different.
150 lives increased by one year = 150 extra years of life
At 140,000 years of drug treatment, this means you need almost one thousand years of drug treatment to gain one year of extra life. In my opinion this changes the result from ‘not bad’ to ‘a complete waste of time and money.’ Especially when you consider that all drugs have side effects. Some worse than others.
You could add, if you were a troublemaker like me, that in reality it takes about one thousand years worth of side-effects to create one extra year of life. And, at about $1,000/year for the drugs, it also costs almost one million dollars per year.
Perhaps you could put this another way. According to the figures, the average person taking a blood pressure lowering medication for thirty years will gain 30/1000th of a year of extra life – or about nine days. Thirty years of treatment for nine days extra life! Whoopee! Raise the flags and fire a cannon in the air. There have got to be better ways of spending money to improve health than this.
To be frank, I am not sure if the cardiology community is even aware that the concept of ‘lives saved’ is utterly misleading. It is so ingrained into the thinking that no-one even questions it. Most cardiologists would (I know I’ve tried it) just look at you in bemusement if you suggested that talking about lives saved hugely inflates the benefits of drug treatment and is therefore intellectually dishonest.
They cannot, or will not, see that there is a problem here. And I am sure that, until you read this, you perhaps didn’t realise there was even an issue, let alone a problem. ‘Saving lives, sure, that seems reasonable, what’s wrong with that.’
Maybe I should try to arrange a meeting where oncologist and cardiologists discuss how they set-up, then present, the results of drug trials. Although, I suspect that the oncologists may be seduced by the cardiologists, rather than the other way round. ‘You mean you’ve found a way to make the benefits of treatment seem three thousand per cent better….wow… way to go.’
* * *