Margaret McCartney: Only data can say if new is betterBMJ 2017; 357 doi: https://doi.org/10.1136/bmj.j2191 (Published 08 May 2017) Cite this as: BMJ 2017;357:j2191
All rapid responses
The problem with data is provenance. These days evidence costs money and journals charge to publish (or at least to allow free access). Inevitably those with money such as big pharmaceutical companies can produce more data and get it published.
The problem with new is anonymity. Those with large marketing departments such as big pharmaceutical companies can get the message out more widely.
My current experience as a GP innovator is an example. I and a colleague have discovered that emulating the neonatal gaze involving 3rd and 4th cranial nerves thereby stimulating the most rostral parasympathetic nucleus while focussing on interoception (visceral awareness) can calm anxiety symptoms in under 5 minutes. I have used the "Baby-Gaze" successfully in 10 minute appointments for over a year and have published an initial audit (1).
Patients love the idea of learning a technique to use themselves without relying on a prescription for example those with flight anxiety prefer to Baby-Gaze than take diazepam. Initiating SSRIs for anxiety has become ancient history for me (Citalopram prescriptions for anxiety have been estimated as costing £19million per annum in UK). Today a woman who had been suffering for the two years since the death of her sister returned to tell me that by Baby-Gazing every other day since I taught her a fortnight ago, her "life has been transformed".
Colleagues make me feel like a snake oil salesman. For example, on offering one colleague the opportunity simply to see the technique in action for that GP's patients the responses (copied and pasted from emails) were:
1. hello, please send me your work that shows it's evidence based
2. I am afraid that your publication doesn't help me - no control group, no time scale, etc, so thanks, but no thanks.
Big data takes big money. Marketing ideas takes money. Money saving innovation is the enemy of the drug companies who peddle their wares with Marks & Spencer sandwiches, fancy brochures and "free" education.
If taking up new ideas is to be based on data, only the ideas of the best funded (drug companies in the main) stand a chance.
I hope that one day the colleague I made the offer to demonstrate the Baby-Gaze to will accept.
If "we" are to gather data we should remember that someone had first to create it.
At least I know how Lister must have felt suggesting that his
(1) Ashworth AJ, Dutton PV Baby-Gaze: A neurobiological method of anxiety relief in trauma Trauma Emerg Care, 2016 Volume 1(3): 36-39 doi: 10.15761/TEC.1000115
Competing interests: No competing interests
. “.....the obvious question is why it was approved without knowing whether the valsartan (costing around half the price) or the sacubitril was responsible for any benefits”.
In my days of advising and writing expert reports to get veterinary drugs authorized, combination drugs required proof from actual studies that each of the drugs in the combination actually contributed to the result. This was actually a specified requirement of the EMA in their guidance. Unfortunately, that was many years ago; I cannot remember the specific location.
However, It certainly caused a lot of problems for older combination therapies and resulted in many useful products being removed from the market because of the cost of proving the value of each constituent in the combination.
The above quotation would seem to breach that regulation for Big Pharma companies.
“The trial was stopped prematurely because it found an “overwhelming benefit” from the new combination drug.”
I am afraid that when I see that a study “has stopped early” I suspect that all is not well. The JUPITER study is a case in point particularly with its hidden/obscured data.
“The difference in mortality alone was smaller: 17% in the test group and 19.8% in the enalapril group. Is this reason enough for NICE approval?”
19.8% minus 17% is a mere 2.8% difference or an improved efficacy for 1 in 35 patients. In short 34 patients are unlikely to benefit, a probability of NO BENEFIT being p = 0.97. in short approaching certainty of NO BENEFIT. Cost wise this represent a massive increase for all 35 patients with only one patient benefiting as a consequence.
Then we have Prof. S Post’s study, Lancet. 2000 Jan 15;355(9199):175-80.Flawed Systolic blood pressure and mortality based on Framingham data.. which demonstrated that the model used in BP studies was in fact wrong and significantly so. This important paper has simply been ignored. Its conclusion The Framingham data contradict the concept that lower pressures imply lower risk and the idea that 140 mm Hg is a useful cut-off value for hypertension for all adults. There is an age-dependent and sex-dependent threshold for hypertension. A substantial proportion of the population who would currently be thought to be at increased risk are, therefore, at no increased risk.
Then we have the US CDC and NHBLI claim of “an epidemic of Congestive Heart Failure” which was rapidly removed from the internet, presumably because the rise was in lock step with the increasing sales of cholesterol lowering drugs and the known “revolving door” between the CDC, NHBLI, FDA and Big Pharma. the NHBLI fact sheet now resides at:
It seems to me that data of which Big Pharma does not approve is rapidly dumped.
Competing interests: No competing interests