Whilst there are times when we have straightforward answers to this question there are also times when we do not know the best treatment. However in most cases the best way of answering this question is with a clinical trial – particularly a trial in which a patient is allocated to a particular treatment completely by chance, if possible when neither the doctor nor the patient knows which treatment they have received.
The study will continue for a time which is decided at the beginning- usually a number of years, and then an outcome – which again is decided at the beginning will be looked for at the end. If there is no known effective treatment then the study is often placebo controlled ie an inactive substance is used as a treatment for some patients, if there is a known and established treatment then the studies are performed comparing that established treatment.
The best example of a large randomised placebo controlled study that made and continues to make a large difference in medical treatment was the ISIS-2 study. This study looked at treatment of patients with an acute heart attack. Each patient had strict criteria for the diagnosis of heart attack and if they had a heart attack as was defined by the study and correct age etc. were enrolled into the study ( obviously with their consent).
At the time of the study there was no agreement as to the best treatment for a heart attack and so it was perfectly possible that any treatment might have made things worse rather than better – hence it was proper and ethical to include a placebo ( or inactive drug). At the time of giving the medication neither the doctor or the patient in the study knew which treatment they were getting making the study a double blind study. So each patient could have had the active agent aspirin tablet, or a placebo tablet, and they also could have had the first lytic agent streptokinase (so called clot-busting drug) by infusion or a placebo infusion .
So any patient could have both placebos, both drugs or just one drug.
The result was that the patients receiving aspirin and streptokinase did better than those receiving just one of the drugs and better than any patient that received no active drugs just placebo. This study was done so well and was so large and included a wide cross-sections of people and places that it had an immediate effect and changed the treatment of heart attacks from sitting and waiting for complications to clearing the blocked coronary arteries with either drugs or much later instruments.
What is of interest to me after all these years is that it is often forgotten that the benefit of aspirin alone was the same as that of the lytic agent alone, but that both agents together gave an additive benefit. The benefit of either of these treatments was increased by giving the treatment as soon as possible. Still I think doctors and patients underestimate the beneficial effect of aspirin in these circumstances.
However the only way of “knowing” these right treatments was by doing this study. It was prospective so the quality of the information collected was uniform. Nobody knew what treatment was given. The outcome was straightforward as well – whether the patient was alive or not at a particular time after their heart attack.
This kind of study in a large number of patients is the gold standard study for evaluating treatments and when properly done yield results that are uniformly accepted and acted upon. It is clinical trials like this that might take 5 or more years from asking the question to getting the answer that drive our medical knowledge forwards. Once a study has been done it must be published in a place that doctors interested in the study can read and ask questions about how the study was done the types of patients that were used etc., and expect useful answers. That prolonged process is the reason why advances in treatment are often seen as slow.
When studies are done in small numbers of patients or are not set up prospectively or not randomised properly or either the patient or the doctor can guess which treatment was given, or the study is not actually published or published in a journal that few doctors read – then the answers cannot be relied upon as much or maybe at all.
Usually when we are fairly certain of the best treatment one of these prospective randomised double-blind studies has been done, and when we are not so sure one of these studies has not been done or has been done but not so well or that there are different interpretations of the results.
When you hear a piece of new information be aware that the quality of the research is more important than how new it is, and the difficulties involved in sometimes giving a clear and honest answer to the question of “what is the best treatment”.
Lancet. 1988 Aug 13;2(8607):349-6
![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_c.png?x-id=78febe56-3522-4daa-847c-d324d3701785)