4 of 4 people found the following review helpful
Punchy and eminently readable,
This review is from: Testing Treatments: Better Research for Better Healthcare (Paperback)This punchy and eminently readable book, by two doctors and a healthcare-research campaigner, takes a critical look at how treatments are assessed - and how so often the way they are evaluated falls short of the ideal.
The essence of good treatment evaluation boils down to the `fair test', the clinical trial that compares different treatments in a way that avoids bias, takes account of the play of chance and assesses all relevant, reliable evidence.
It might be supposed that clinical trials are based on well-constructed research protocols and done with due safeguards for unbiased reporting, but that doesn't always happen. All too often it is a question of profit-driven rather than evidence-based evaluation - pharmaceutical companies want to put their results in the best possible light, because their profit depends on it. One ploy is to withhold negative results from publication, but even independent researchers may neglect to write up their reports if the results are disappointing. Astonishingly, at least half of all clinical trials are never fully reported, which makes the collection of all relevant material for an assessment a daunting task.
The authors want to draw patients into helping to make decisions about their own treatment, and to this end the book tells them how to judge whether claims for treatments are trustworthy. "Getting the right research done", the authors conclude, "is everybody's business".
The book is filled with examples of good, bad and unnecessary research; injudicious use of screening ("early diagnosis is not always better"); the proliferation of "me too" drugs (new, expensive medicines that offer no advantage over older, similar ones); and the misuse of statistics. It is also liberally sprinkled with apt and sometimes pithy quotations - from thoughts on certainty and opinion (from Charlie `Peanuts' Brown) to observations on the overdiagnosis of prostate cancer. And it is beautifully written.
Tracked by 1 customer
Sort: Oldest first | Newest first
Showing 1-1 of 1 posts in this discussion
Initial post: 8 Oct 2012 17:21:45 BDT
I can truthfully say that it was one of my main frustrations, throughout my research career, that if a study asked the question "if we do this, does that happen?" and the answer was "apparently not", then, generally speaking, nobody was interested in its publication. Not necessarily some profit-driven ploy, as you acknowledge; the research didn't have to involve some putative wonderdrug BadPharma was wanting to flog. There seemed to be a basic human failing at work, whereby the answer "no" was seen as inherently less interesting than the answer "yes" .. frighteningly shortsighted, and mindbogglingly, heartbreakingly disappointing. After all, nobody should go into science with the idea that every experiment will give you the answer yes to a prejudged outcome. If you come away with a better question to ask, this was also a good day's work.
‹ Previous 1 Next ›