I mostly disagree. First of all, this column conflates "the public" and "scientists" without a clear division. As a science writer for many years, I've seen the danger of assuming that non-scientists in that public can get at the core ideas without some very skillful science communication --- that is creative, analogical writing with ample metaphors by a biomedical scientist who both understands the data deeply and has the creative skill necessary to accurately portray those data to non-scientists in a palatable form. Second, science does not ever rest on one set of studies. As the continually changing landscape of dietary fat recommendations illustrates, most of these highly complex topics require many years to accumulate research that clarifies and builds upon prior studies before any real clarity emerges. Quickly feeding one study to the public risks the entrenchment of one view in the collective psyche that may later be proven only partly valid, at which time, changing the mind of John Q. and Mary R. Public becomes a daunting task. Again, as a scientist/science writer, I've seen this dangerous scenario play out, first-hand.
2
Allowance for early stopping of a trial is embedded in the initial protocol, and the statistical criteria for stopping are carefully defined. Early stopping is indeed big news, but the researchers should be allowed time to reexamine the numbers and check through the roles of many covariates -- weight, age, gender, diabetes status, ethnicity, .......
It should also be remembered that the American public is not very good at sorting through medical information. The vaccination issue has been clouded by a misleading medical finding, and we walk through a storm of confusing nutritional advice. The local health food store is filled with strange and unusual chemicals, most of which are supported by anecdotal enthusiasm rather than serious study.
Please allow the blood pressure researchers a few months to get this all organized.
It should also be remembered that the American public is not very good at sorting through medical information. The vaccination issue has been clouded by a misleading medical finding, and we walk through a storm of confusing nutritional advice. The local health food store is filled with strange and unusual chemicals, most of which are supported by anecdotal enthusiasm rather than serious study.
Please allow the blood pressure researchers a few months to get this all organized.
4
The author seems to conclude that since printing random thoughts via the internet is fast, so should be medical research. Ironic he chose one of the few reviewed print journals left to voice his opinion, rather than a facebook post.
Good point. But that was not his point.
The Internet. Email, has taken several months off the review process.
I can see the vultures (Daiichi, Hoffman, Merck, Novartis, Pfizer, Sanofi, etc.,) circling .................
Today's use of the word "breakthrough" has lost it's luster, and we have the world of media to thank for that. Years ago, "breakthrough" in regard to news, meant something unimaginable had happened in the world. However, today it could mean something as minuscule as...BREAKING NEWS: SELENA WILLIAMS WINS WIMBLEDON!
In a world where everything is geared toward needing it NOW, we tend to forget that faster is not always better, especially in the world of medicine.
The trial mentioned consisted of 9,000 people, but did not provide the gender ratio of men to women, which is a vital piece of missing information. BREAKING NEWS: WOMEN ALSO SUFFER FROM HIGH BLOOD PRESSURE. Inequities in this area of research, still remain today in all areas of clinical trials. An accurate end result of a clinical trial is imperative to all it affects.
There's a distinct method in the madness of the waiting game. Just because one clinical trial looks promising early in the game, does not mean that this information should be released to the general population. There are risks to consider, which are only known IN TIME. Are we giving "false hope" to those in need, when there is no hope to give? Let's not forget about those who financially benefit from the streamlining of drugs: pharmaceutical companies, lobbyists, and those in Congress who vote for the fast-tracking of drugs and medical devices.
Tread lightly my friends..very lightly. As the article states, "lives depend on it."
In a world where everything is geared toward needing it NOW, we tend to forget that faster is not always better, especially in the world of medicine.
The trial mentioned consisted of 9,000 people, but did not provide the gender ratio of men to women, which is a vital piece of missing information. BREAKING NEWS: WOMEN ALSO SUFFER FROM HIGH BLOOD PRESSURE. Inequities in this area of research, still remain today in all areas of clinical trials. An accurate end result of a clinical trial is imperative to all it affects.
There's a distinct method in the madness of the waiting game. Just because one clinical trial looks promising early in the game, does not mean that this information should be released to the general population. There are risks to consider, which are only known IN TIME. Are we giving "false hope" to those in need, when there is no hope to give? Let's not forget about those who financially benefit from the streamlining of drugs: pharmaceutical companies, lobbyists, and those in Congress who vote for the fast-tracking of drugs and medical devices.
Tread lightly my friends..very lightly. As the article states, "lives depend on it."
1
As the authors note: "we do already know that aggressive treatment of high blood pressure with medication carries a risk of fainting and other important side effects". So perhaps we're not seeing a "breakthrough" but rather a shift in consequences for patients. Are the authors confident that the ultimate balance is positive?
For that matter, was not the NIH curious enough to complete the study and quantify the long-term effects of the treatment?
We are, after all, talking about people's lives.
For that matter, was not the NIH curious enough to complete the study and quantify the long-term effects of the treatment?
We are, after all, talking about people's lives.
1
"Don’t Delay News of Medical Breakthroughs"
"Antidepressant Paxil Is Unsafe for Teenagers, New Analysis Says"
Pick one.
"Antidepressant Paxil Is Unsafe for Teenagers, New Analysis Says"
Pick one.
5
Many children killed themselves, killed others or suffered terrible consequences when a medical journal article was published before the study, Paxil Study 329, was finished. The article stated that the antidepressant Paxil was safe and effective for children. It became the subject of a lawsuit by the attorney general of New York in 2011 and a judgement of a $3 billion penalty to be paid by GlaxoSmithKline for false marketing. One of the co-authors of that fraudulent article went on to become a celebrity and developed the current mental health programme for Ontario schools. This programme is subsidized by Lundmark, maker of 11 antipsychotic drugs and 5 antidepressants and that co-author pushes the use of antipsychotic drugs with children. Abominable. http://study329.org/what-can-i-do/
6
The authors have an excellent point. At the same time, we should not delay news of medical harm. It takes years, decades even, for news for harm from medical and dental devices to filter back to physicians and dentists. Thousands of patients and physicians have filed adverse event reports on devices such as dental amalgam, Essure, Mesh, morcellators, implants, stents and more. Yet until it shows up in major journal articles, or the slow wheels grind forward against all odds and device company opposition at FDA, physicians and dentists continue to put devices in people that harm them. This is a major gap in patient safety that needs to be addressed by the Senate in 21st Century Cures and FDA reform bills. Join the call for Medical and Dental Device Safety Urgent Reform: MEDDSURGE: http://www.petition2congress.com/18325/medical-dental-device-safety-urge...
1
Dr. Topol and Dr. Krumholz may have a point, but many of the comments are, in my view, more sensitive to the issues involved. As a physician and a scientist, I prefer to wait a few months to see the final report. And no, rarely if ever the results of any trial may be regarded as "chiseled in stone". Getting reliable knowledge takes time to conceive and design the study, to gather data, to analyze and report them, and, after pubication, to grasp the implications.
2
Has any medical association or consortium considered investing in a program to educate and hire increased numbers of research reviewers in order to handle the increased number of research papers with no loss in quality or turnaround time?
I think that's what the free market used to do in the old days, to ensure that jobs and wages kept pace with work and costs.
Seems to me it's a better idea than premature epublication.
I think that's what the free market used to do in the old days, to ensure that jobs and wages kept pace with work and costs.
Seems to me it's a better idea than premature epublication.
2
Yes, by all means - announce as fast as possible - but on a Etch-a-Sketch because next week, it will turn out that: omega-3 does not help prevent heart attacks, Vitamin E doesn't do whatever, we do/don't need more Vitamin D. We need to stretch before we exercise. No we don't. All calories are the same. No they are not. Eat less fat. East less carbs. Cholesterol matters. No, it doesn't. Statins are the answer. Statins are a pharmaceutical conspiracy.
Yup. Etch-a-sketch.
Yup. Etch-a-sketch.
3
I just read 4 comments by Jim, S. Naka, Schloss, and B. and find them more thoughtful than the article. Premature release of medical testing to the public can be a disaster, unless you are in the ad department of a drug company.
Releasing information to the public must be at a different pace to releasing the information to the medical profession. There is supposed top be a peer review period.
Releasing information to the public must be at a different pace to releasing the information to the medical profession. There is supposed top be a peer review period.
3
Who exactly is supposed to make sense of a rapid dump of raw data as suggested? The mainstream media? The average person? (God forbid) internet news aggregators? Interested and knowledgeable but harried and overworked physicians? So many other commenters have properly said why this is not a good idea. Unlike so many other things in our world, the acquisition and dissemination of scientific knowledge is one thing that is mostly done correctly, and it's a system that is not broken and doesn't need fixing.
2
Didn't the NIH do what the authors wished by ending the trial two years ahead of schedule? They didn't delay the news of that action, as the title would suggest. And if the authors want the data faster than the established peer review process, shouldn't such a significant change be based on more than one example, especially one they describe as unusual, or outside the norm. (Interestingly, would the early sharing of data also allow for the sharing of liability?)
3
If the authors can identify a means to avoid all the quick-release quackery that abounds in the misguided interpretations of breaking news in science, medical or otherwise, then we have perhaps a proposition. But proper peer review requires vetting by independent researchers in the field, and there seems to be little means for coordinating that in time and space without putting everyone on the same payroll. I applaud the sentiment, but wonder if the authors wouldn't be opening Pandora's box with this idea.
3
Results of many studies are regarded as breakthroughs in medicine. Careful analysis by experts is essential before implementing or recommending a study's conclusion.
1
Employment Opportunities #1: The Medical Profession
September 18, 2015 - New York Times
Headline: "Don’t Delay News of Medical Breakthroughs"
Problem/Opportunity:
There is an ongoing backlog of medical research papers waiting to be reviewed, and their recommendations, if found to be medically beneficial, could help people all over the world
Solution Alternatives:
(_) Educate and hire more reviewers to keep pace with papers
(X) Publish papers without reviewing, fix medical "oops" later
September 18, 2015 - New York Times
Headline: "Don’t Delay News of Medical Breakthroughs"
Problem/Opportunity:
There is an ongoing backlog of medical research papers waiting to be reviewed, and their recommendations, if found to be medically beneficial, could help people all over the world
Solution Alternatives:
(_) Educate and hire more reviewers to keep pace with papers
(X) Publish papers without reviewing, fix medical "oops" later
3
Yes, lets rush to increase drug sales using any shred of evidence from any study, verified or not, since capitalism demands it. Funny how the preferred policy always coincides with what will generate the most revenue.
2
As a consumer of healthcare and health news, I am leery of an approach which is too speedy simply because we live in an instant information world. My first career was nursing, so I have more knowledge about the body and things medical than much of the general public. Still, I often feel tossed about by health news which is offered than retracted or by incomplete information or information about which I have questions no one seems able to answer.
Recently an online program used by my employer labeled my 122 systolic B/P as "pre-hypertension" and recommended all kinds of life-style changes. When I asked my doc about it, she dismissed it telling me that "they" are no longer using such categories and I was fine. Now I'm wondering if the new study has gone beyond saying that 122 is hypertension. When the goal was 140 with meds, it seemed that those below that were operating, at most, under a yellow caution flag. Now if 120 is the goal, does that mean that millions in the 120-139 range are now considered to have hypertension to be treated by meds? I have seen no answers. I doubt that a posting of study data would answer that question.
Whiles study data is interesting, all too often the basic, practical questions consumers have are not answered for months or even years after the headline news of the study. Yet the only relevant question for most of us is "What does this mean for my health and well-being and/or that of my loved ones?"
Recently an online program used by my employer labeled my 122 systolic B/P as "pre-hypertension" and recommended all kinds of life-style changes. When I asked my doc about it, she dismissed it telling me that "they" are no longer using such categories and I was fine. Now I'm wondering if the new study has gone beyond saying that 122 is hypertension. When the goal was 140 with meds, it seemed that those below that were operating, at most, under a yellow caution flag. Now if 120 is the goal, does that mean that millions in the 120-139 range are now considered to have hypertension to be treated by meds? I have seen no answers. I doubt that a posting of study data would answer that question.
Whiles study data is interesting, all too often the basic, practical questions consumers have are not answered for months or even years after the headline news of the study. Yet the only relevant question for most of us is "What does this mean for my health and well-being and/or that of my loved ones?"
14
Absolutely. Further, I wonder which of the BPs that I have during the day or at doctors visit should count. I've had systolic pressures of 110 or as high as 136 at different times during the same day. Treating the higher to below 120 will probably treat my 110 to a dangerous level especially when driving or otherwise involved. If these recommendations are only for a select group how will each physician determine to whom this applies?
1
The authors of this piece are eminent scientists recognized highly within their fields. What is disheartening is a de-emphasis on their part of the caveats of rapid dissemination of preliminary results. Science is irrevocably discredited when breakthrough research is publicized prematurely, found to be erroneous, and then retracted in the harsh light of the public--the STAP cells as a case in point. This is the nature of scientific investigation but each example weakens the public perception of the objectivity and therefore reliability of scientists. The changing perception fuels a justification to cut federal funding of research and divert the money to the cold, black-white logic of supporting military infrastructure.
17
Dr. Topol and Dr. Krumholz make a valid and fantastic point. However, there is one important distinction they seem to be missing in terms of releasing data: the technical knowledge gap between trained scientists (i.e., doctors) and the general public (i.e., patients). Transparency is great for those who can understand the context, but can be misleading for those who can't. Even among scientists the interpretation of raw data can vary drastically, which is a healthy thing because this is the way science is supposed to work. Therefore, when presenting medical breakthroughs to the general public, there should be stricter guidelines on how responsible institutions, researchers and the media convey this information. For example, all types of credible views should be included and there should be a very strong emphasis on disclaimers, especially in terms of what areas studies did not cover. Even the scientific community gets it wrong (think hormone replacement therapy or risk of suicide with SSRI antidepressants), which is not always a bad thing. The public needs to know that virtually all findings in science and in medicine in particular, are works in progress.
8
FWIW: I say stopping trial is ethical response for safety of participants. Broad application to population requires analysis and peer review. FIrst analysis should be done by study investigators & published after peer review. After that, release the data and have at it.
Really impressive comment stream. Nice to see such intelligence and civility in this forum.
Really impressive comment stream. Nice to see such intelligence and civility in this forum.
6
The only real news about findings of the SPRINT study is that it was stopped early because preliminary analysis by the study team found significantly lower cardiovascular events and deaths associated with aggressive treatment of blood pressure and a review panel verified their statistics. Anything else at this point is speculation, not science.
Most practicing physicians will await more thorough analysis, peer review and, perhaps confirming studies before adopting such aggressive treatment of high blood pressure.
It would be good to reduce avoidable delays in publication, such as agreements to delay publication until results have been presented at a conference or the lengthy process of getting an article peer reviewed an in print by a leading journal. But, short-changing the scientific process of a complex, randomized study is dangerous. Study participants must remain on their respective treatment regimens until final data have been collected. Data must be verified, standardized and entered to a database, then analyzed in much more detail than would be required to find a difference in overall rates. The information must be interpreted, alternative possible explanations considered and conclusions drawn. This will take months and will be time well spent.
Otherwise, encouraging premature release might result in parties with vested financial interests promoting bogus or even harmful treatment for millions of people with blood pressure above 120 mm/Hg
Most practicing physicians will await more thorough analysis, peer review and, perhaps confirming studies before adopting such aggressive treatment of high blood pressure.
It would be good to reduce avoidable delays in publication, such as agreements to delay publication until results have been presented at a conference or the lengthy process of getting an article peer reviewed an in print by a leading journal. But, short-changing the scientific process of a complex, randomized study is dangerous. Study participants must remain on their respective treatment regimens until final data have been collected. Data must be verified, standardized and entered to a database, then analyzed in much more detail than would be required to find a difference in overall rates. The information must be interpreted, alternative possible explanations considered and conclusions drawn. This will take months and will be time well spent.
Otherwise, encouraging premature release might result in parties with vested financial interests promoting bogus or even harmful treatment for millions of people with blood pressure above 120 mm/Hg
6
Lots of interesting comments here. It is important to know that the NIH has concluded that the results are final enough to announce publicly (they did) and to stop the multimillion dollar study. The data were thoroughly reviewed by independent experts (the data safety and monitoring committee). What we are saying is that given that conclusion, it would seem useful to release the information that was provided to the committee and used as a basis to stop the study. Right now we have news that this remarkable study was so successful that it was stopped years before its planned completion - and the NIH said the implication was that the knowledge could be lifesaving. What we are saying is that rather than wait months for the publication, we need to know more know to inform our patients. This is not so early that the results will change - in fact, studies are stopped early because the results are so definitive that continuing the study is not necessary. The final results should still be published, but transparency now can be helpful. And there are lots of examples of delays. Another recent study took 9 months from the time the study ended until it was published. What is an acceptable delay? This is different from the situation when the study has ended but the results of the study are not yet known; here the fact the study is positive has been announced, we just lack details that would allow some idea about what it might mean for patients.
2
I remember reading the announcement of this blood pressure study in the NY Times, and had a number of questions about how this study was conducted. But we couldn't read the study because the results had not been carefully peer reviewed and published yet. Good heavens, how many errors, questions, and problems do peer reviewers identify with almost any study published?? I also read a number of comments by readers which brought up a host of concerns about the reliability, validity, and reasons why such results could be suspect and/or cause harm.
Call this the New Risky Commercial Model of Research, but it flies in the face of the long-standing Scientific Model of Research, which is based on caution, skepticism, critical peer review, and verification of results. This argument is especially troubling, since these cardiologists are not talking about plant research but medical research, where the consequences of rushing to use and then discovering harmful consequences can be disastrous.
Can these eager physicians not imagine the ethical problems associated with a rush to market? Conflicts-of-interest plague medical research these days, which is all the more reason to wait for caution and verification.
Good thing we did not follow this Commercial Model of Research with thalidomide in the U.S. when other countries in Europe concluded too early that the drug could be given to pregnant mothers.
Better safe than sorry still holds
Call this the New Risky Commercial Model of Research, but it flies in the face of the long-standing Scientific Model of Research, which is based on caution, skepticism, critical peer review, and verification of results. This argument is especially troubling, since these cardiologists are not talking about plant research but medical research, where the consequences of rushing to use and then discovering harmful consequences can be disastrous.
Can these eager physicians not imagine the ethical problems associated with a rush to market? Conflicts-of-interest plague medical research these days, which is all the more reason to wait for caution and verification.
Good thing we did not follow this Commercial Model of Research with thalidomide in the U.S. when other countries in Europe concluded too early that the drug could be given to pregnant mothers.
Better safe than sorry still holds
19
In one of the hundred or so scientific papers that I have peer-reviewed, one was so faulty that it was never published, even decades later. Every paper required at least one correction.
It is unbelievable that two academically based cardiologists are advocating the early release of data without some level of peer review. Yes, this study was prematurely terminated, but only after a data and safety monitoring board went through the data and recommended early termination. This situation occurs only very rarely and when the results are clear cut.
When a scientist wants to publish data, they always make some conclusions based on the data. Sometimes those conclusions are only partially supported by the data. The purpose of peer review is to insure that unsupported claims are vetted. What kind of process are Topol and Krumholz advocating, an unsubstantiated data free for all?
When a scientist wants to publish data, they always make some conclusions based on the data. Sometimes those conclusions are only partially supported by the data. The purpose of peer review is to insure that unsupported claims are vetted. What kind of process are Topol and Krumholz advocating, an unsubstantiated data free for all?
14
What is the purpose of peer review? Interpretation. But when you run a randomized trial, there is less need for interpretation because the design was laid out in advance. Releasing the protocol and the raw data would be helpful to the medical/ scientific community and enhance transparency.
Never have I seen such an irresponsible op-ed concerning medical matters written by two esteemed cardiologists. The comments are unanimous. Perhaps the editors want to publish a retraction? Many will be reading the print edition without the benefit of perusing the comment section. This was a classical case of " . . . on the other hand . . . but . . . "
What do Drs. Topol and Krumholz have to say to readers that may decide to up their dosage of already prescribed BP meds? Some will hasten the development of known side effects. Some will undoubtedly have syncopal episodes with possible trauma (patients who faint may have resultant nasty head injuries, hip fractures, or worse- imagine fainting when behind the wheel of a car). The rebuttal to this piece should be published on the front page. And despite their warnings of increased side effects and fainting episodes, pharmaceutical executives are positively giddy. How many more billions of pills will be sold in the quest to lower systolic BP from 140 to 120? And how many tens of millions previously told their BP was fine will now be labeled hypertensive and placed on pharmaceutical regimens? (The majority outside that 49 million current hypertensives do not have systolic BPs under 120.)
This is why, frustrating as it may be, medicine advances slowly. Old standbys taught for decades are now snake oil. Mouth to mouth in resuscitation. The ubiquitous use of oxygen. Proceed with caution. It's not a black and white field.
What do Drs. Topol and Krumholz have to say to readers that may decide to up their dosage of already prescribed BP meds? Some will hasten the development of known side effects. Some will undoubtedly have syncopal episodes with possible trauma (patients who faint may have resultant nasty head injuries, hip fractures, or worse- imagine fainting when behind the wheel of a car). The rebuttal to this piece should be published on the front page. And despite their warnings of increased side effects and fainting episodes, pharmaceutical executives are positively giddy. How many more billions of pills will be sold in the quest to lower systolic BP from 140 to 120? And how many tens of millions previously told their BP was fine will now be labeled hypertensive and placed on pharmaceutical regimens? (The majority outside that 49 million current hypertensives do not have systolic BPs under 120.)
This is why, frustrating as it may be, medicine advances slowly. Old standbys taught for decades are now snake oil. Mouth to mouth in resuscitation. The ubiquitous use of oxygen. Proceed with caution. It's not a black and white field.
4
I agree that all information about this study should be released immediately.
But what seems to have been released is a generalized conclusion that lowering blood pressure more than previously thought is generally of great benefit.
But wasn't this study of a sub-set of people, rather than the general population? From my faulty memory, I recall: over 50 years old, not diabetic, kidney disease or high risk of heart disease.
So what, for example, defines "high risk of heart disease"? Does it have anything to do with the wildly varying LDL levels set by the so-called expert panel on statin guidelines? Those numbers come from a panel of ACC/AHA, most of the members having financial ties to the pharmaceutical industry.
Are we talking about having lower blood pressure numbers naturally, or is it worth taking perhaps three medications?
Is there no financial motivation because the medications are generic? The numbers are huge, so why wouldn't there be huge profit in generics?
And since the Supreme Court decision of 2013, generics aren't responsible for design flaws in the medications. Not even responsible for putting newly-found dangers on the warning labels.
Doctors, please protect us! Please show great care in giving medications to healthy people because they're "at risk". There's so much money sloshing around!
Statinvictims.weebly.com
But what seems to have been released is a generalized conclusion that lowering blood pressure more than previously thought is generally of great benefit.
But wasn't this study of a sub-set of people, rather than the general population? From my faulty memory, I recall: over 50 years old, not diabetic, kidney disease or high risk of heart disease.
So what, for example, defines "high risk of heart disease"? Does it have anything to do with the wildly varying LDL levels set by the so-called expert panel on statin guidelines? Those numbers come from a panel of ACC/AHA, most of the members having financial ties to the pharmaceutical industry.
Are we talking about having lower blood pressure numbers naturally, or is it worth taking perhaps three medications?
Is there no financial motivation because the medications are generic? The numbers are huge, so why wouldn't there be huge profit in generics?
And since the Supreme Court decision of 2013, generics aren't responsible for design flaws in the medications. Not even responsible for putting newly-found dangers on the warning labels.
Doctors, please protect us! Please show great care in giving medications to healthy people because they're "at risk". There's so much money sloshing around!
Statinvictims.weebly.com
6
The whole point behind the scientific process, of which peer-reviewing results before publication is a big part, is to prevent bad results from being accepted as truth. Rushing to push results out that may be proven wrong serves only the business who stand to profit at the expense of the public's trust in science. We're already experienced that *exact* problem with the false vaccine/autism fiasco, and indeed we're still suffering from it (if you saw the GOP debate you know what I am referring to). Things are bad enough; we don't need that to become the new normal.
4
While I am not competent to evaluate this argument either way, as a multiple drug compliant controlled high blood pressure patient for 25 years, I have more than passing interest. I rely on the guidance of my medical doctor professionals. Medicine is part art and part science. This is all magic and mystery to me.
2
This is a complex issue because, well, life is complex! It is rather ironic to see this post, on the same day that BMJ published the re-analyzed data on "study 329" indicating that, lo-and-behold, paroxetine and imipramine may NOT help depressed adolescents, but may in fact hurt them- this, too, was life-saving information, that has been under some degree of "wraps" for too long. The instantaneous and social nature of the web is certainly tantalizing (yep, I've "tweeted" today about 329), but the purpose of the multiple levels of review, analysis, etc that data undergo to reach publication is to be SURE of the validity of the information.
The Study 329 story and the oseltamivir story are two that demonstrate that even the best efforts of "science" don't always produce well-founded and reliable patient-oriented evidence. Nevertheless, to try to advance sharing of information TOO rapidly, risks over-enthusiasm for interventions that may NOT in the end help, or may be harmful.
Any time I see news of a "life-saving breakthrough," my first reaction is "oh, really???" - most supposedly miraculous interventions, aren't.
If we can find ways to make genuine scientific inquiry, peer review, and vetting of results and evidence move more expeditiously towards sharing of solid, patient-oriented evidence, then great!
I fear that a too-quick rush to share supposedly "breakthrough" information, may degrade the overall quality of information available to patients and professionals.
The Study 329 story and the oseltamivir story are two that demonstrate that even the best efforts of "science" don't always produce well-founded and reliable patient-oriented evidence. Nevertheless, to try to advance sharing of information TOO rapidly, risks over-enthusiasm for interventions that may NOT in the end help, or may be harmful.
Any time I see news of a "life-saving breakthrough," my first reaction is "oh, really???" - most supposedly miraculous interventions, aren't.
If we can find ways to make genuine scientific inquiry, peer review, and vetting of results and evidence move more expeditiously towards sharing of solid, patient-oriented evidence, then great!
I fear that a too-quick rush to share supposedly "breakthrough" information, may degrade the overall quality of information available to patients and professionals.
3
Didn't we learn when the "Significant" news that studies showed that vaccinations caused autism were released and picked up by the media, before they were thoroughly reviewed?
18
Unfortunately, that study WAS published in a peer-reviewed journal, was later (too late) determined to be research fraud, and the doctor who made up the data had his medical license revoked. Peer review, while necessary, is not sufficient to ensure validity, especially in cases of deliberate research misconduct.
While this op-ed is a hot mess of scientific ignorance, the comment section is inspiring. How wonderful to see so many people who understand and applaud the scientific process and peer-review system!
24
The one sole example given in this article sounds like a very unusual case, from which little can be inferred.
There have been so many sharp U-turns in our understanding of how to prevent and treat disease thanks to experiments that cannot be duplicated, and so many valid studies whose results have been either suppressed or exaggerated for commercial reasons.
We should be extremely wary of jumping to accept research results that have not been duplicated by disinterested parties.
There have been so many sharp U-turns in our understanding of how to prevent and treat disease thanks to experiments that cannot be duplicated, and so many valid studies whose results have been either suppressed or exaggerated for commercial reasons.
We should be extremely wary of jumping to accept research results that have not been duplicated by disinterested parties.
11
There's a lot of misunderstanding in this article about how and why science publishing works the way it does.
Sometimes researchers release early results before they have finished going through their own data (i.e. understanding it themselves) or getting it reviewed by others (checked for errors by other people in their field). As the author notes, this data is incomplete and imsufficient for making medical care decisions. What he fails to acknowledge (understand/accept?) is that at this stage in the research process he shouldn't be trying to apply anything.
It is dangerous for doctors to make medical decisions based on incomplete preliminary data.
Large scale studies have mountains of data that needs to be analyzed before recommendations can be made. He might want a ton of specific information from a preliminary press release, but that's impossible.
It's preliminary: detailed analysis don't yet exist.
It's a _press_ release: a general overview written for a broad audience.
What he actually wants is for research to be finished faster. Understandable, but also wrong. Fast=sloppy and sloppy medical research are retracted at best and kill people at worst.
Sometimes researchers release early results before they have finished going through their own data (i.e. understanding it themselves) or getting it reviewed by others (checked for errors by other people in their field). As the author notes, this data is incomplete and imsufficient for making medical care decisions. What he fails to acknowledge (understand/accept?) is that at this stage in the research process he shouldn't be trying to apply anything.
It is dangerous for doctors to make medical decisions based on incomplete preliminary data.
Large scale studies have mountains of data that needs to be analyzed before recommendations can be made. He might want a ton of specific information from a preliminary press release, but that's impossible.
It's preliminary: detailed analysis don't yet exist.
It's a _press_ release: a general overview written for a broad audience.
What he actually wants is for research to be finished faster. Understandable, but also wrong. Fast=sloppy and sloppy medical research are retracted at best and kill people at worst.
9
I have a better idea. Rather than releasing raw data, maybe the media should quit trumpeting the study-du-jour from the myriad of junk science published every week?
18
Medical achievements are often only put into their true light when they are applied to current practice and nasty and inconvenient issues such as side effects and drug interactions dampen the intial enthusiasm for use. Let's be patient. Medical practice is not the same as a clinical trial.
7
The devil is in the details. A clinical study might come to a definitive conclusion like aiming for a systolic pressure of 120 mmHg instead of 140 mmHg, reduces heart attacks, strokes and deaths. But it is not the systolic blood pressure or diastolic pressure by themselves. Instead it is the Mean Arterial Pressure which is of consequence. And the MAP has a normal range from 50 mmHg to 150 mmHg. As you become hypertensive there is a rightward shift of the curve meaning a higher MAP is required to keep blood pumping in the sclerosed blood vessels. Risk of fainting is real for patients who have a rightward shift of the curve leading to hypoxia ( decreased oxygen in the blood ) to the organs. Setting all this jargon aside, the bottom line is without knowing and confirming the side-effects and it's consequences, it is dangerous to just put the results i n the newspapers. Eric Topol should know better with his experience from Vioxx.
10
I admire the past contributions of Eric Topol and Harlan Krumholz to medical research but they have not struck the right balance here. First, releasing unstructured data prematurely nowadays often leads to misguided sensationalist handling in the media that leads to adverse outcomes. Recent examples include: the wave of terror engendered by fears of radiation over the last several years which has, led many patients to avoid or refuse much needed perfectly safe testing; and the vaccine-autism fiasco which has resulted in reemergence of preventable diseases such as measles and whooping cough, with significant suffering and a few deaths. Secondly, there is a far bigger problem in hypertension treatment than benefit of lower target BP: namely patients with untreated hypertension due to refusal to take prescribed medications. This "noncompliance" contributes much more to death and disability in the U.S., particularly for minorities, than any further reduction of risk in treated patients taking their medications can possibly deliver and is largely neglected. Finally, a major obstacle to rapid dissemination of important information has been embargos on release of such information by leading meetings and journals prior to their sessions or publication dates. What could work is an NIH sponsored program of rapid ad hoc meetings devoted to such breakthroughs as soon as they emerge but after completion of detailed analysis by the investigators.
4
Vaccine-phobia is not responsible for all the cases of whooping-cough, or pertussis.Pertussis is now treated with a different vaccine, that is short-lived.
This vaccine should be given several times, throughout life.I was surprised to find this out, as never heard of a vaccinated adult, needing to be re-vaccinated.
Believe me, you don't want to have whooping-cough.It's horrible to live through, and can be deadly to infants.
This vaccine should be given several times, throughout life.I was surprised to find this out, as never heard of a vaccinated adult, needing to be re-vaccinated.
Believe me, you don't want to have whooping-cough.It's horrible to live through, and can be deadly to infants.
3
You are expecting the general public to understand statistics and make rational decisions based on them? You expect physicians to act based on data when their patients are clamoring about what they read in the headlines or who knows where on the internet?
Last night one of the presidential candidates in the debate said that vaccines cause autism. A story on the e-front page of the NYT says that doctors often or always change the immunization schedule when requested by parents.
We should wait until the data are sufficiently analyzed to make a decision so that medical advice to further reduce blood pressure can be balanced against other outcomes, e.g., fainting, people feeling tired so they do not exercise, populations in which benefits are or are not seen. Although heart attacks and strokes are acute events, they are associated with chronic high blood pressure. The events are neither caused nor prevented overnight.
Last night one of the presidential candidates in the debate said that vaccines cause autism. A story on the e-front page of the NYT says that doctors often or always change the immunization schedule when requested by parents.
We should wait until the data are sufficiently analyzed to make a decision so that medical advice to further reduce blood pressure can be balanced against other outcomes, e.g., fainting, people feeling tired so they do not exercise, populations in which benefits are or are not seen. Although heart attacks and strokes are acute events, they are associated with chronic high blood pressure. The events are neither caused nor prevented overnight.
6
As a scientist with 45 years of experience with complex biological systems, though not as complex as humans, I was dismayed by the rush to utilize findings of single complex studies before full analysis.
Regardless of how well one tries to select subjects to be the same, humans of the age to have heart issues have huge carry-over of decade of differing life conditions. Such complex human health research is inherently susceptible to unknown variation and varying results from identical treatments. As such we are far better off not responding immediately to each study, but to a body of evidence that comes to a strong consensus.
Regardless of how well one tries to select subjects to be the same, humans of the age to have heart issues have huge carry-over of decade of differing life conditions. Such complex human health research is inherently susceptible to unknown variation and varying results from identical treatments. As such we are far better off not responding immediately to each study, but to a body of evidence that comes to a strong consensus.
4
Peer review carried out by reputable scientific and medical journals provides a necessary and important step in trying to ensure that what is announced is solid and readers, other scientists and potential patients can depend, within certain limitations, on the outcomes. Another questionable set of reasons, as laid out in this column, to accelerate the process is misguided at best and dangerous at worst. Medicine does not need to change its approach to releasing new important information. Practitioners need to maintain their guard against the many perverse pressures that exist to hurry to judgement.
9
The Lancet, a British medical journal, published Wakefield's article in 1998.That's one that contains autism-vaccine link.The Lancet was founded in 1823 and is one of the premier journals, or shall I say, was. It is peer-reviewed.
It wasn't until 2010, when The Lancet retracted that article.The 12 years can never be returned to a child who died of measles.
Peer-review? I don't think so.
It wasn't until 2010, when The Lancet retracted that article.The 12 years can never be returned to a child who died of measles.
Peer-review? I don't think so.
3
It does not follow that rushing to release data from clinical trials automatically results in better care. The opposite is more likely, as clinicians are questioned relentlessly about the new findings, but have not yet had the chance to see the final statistical analysis and determine if the results have credibility. Yes, a review panel halted the SPRINT study in hypertension - but other arms of it go on, including looking at kidney function and cognitive decline. We don't know how high blood pressure was at the start of the study (baseline) in either the standard treatment group (treated to a systolic of 140 mmHg or below) or the intensive treatment group (treated to a systolic of 120 mmHg or below) - and that drop is important, particularly in older patients. We also don't know how quickly the BP was dropped in these patients - was that linked to adverse events? Was the level of fitness of patients at baseline also a factor in how much their cardiovascular risks dropped after their BP fell? So is a sedentary person still at higher risk, even at a lower BP, than a physically active one? We expect answers to questions like these in a final report, which takes time to develop - so a rush to publish numbers, even among a select few experts, might spread generalizations that are false or even dangerous to some patient groups. With respect, we disagree with the authors, who mean well but have not thought through the potentially life-threatening implications of a rushed data dump.
7
It's interesting to me that Eric Topol is the same researcher who (rightfully) fought the battle against Vioxx, which was withdrawn because neither Merck nor the FDA had done an adequate analysis regarding Vioxx's harmful side effects. Rushing medicine through peer review and then approval allows unsafe drugs to come to market, which is why the argument he and Harlan Krumholz put forward is dangerous.
Just last week Harlan Krumholz tweeted about this same study that the entire study needs to be ready for release before publishing findings; to do otherwise is "not good enough."
Both authors, then, seem to be inconsistent in their opinions about such matters, which should lead at the very least to more explanations, but which could also raise concerns about their motives for this article.
Just last week Harlan Krumholz tweeted about this same study that the entire study needs to be ready for release before publishing findings; to do otherwise is "not good enough."
Both authors, then, seem to be inconsistent in their opinions about such matters, which should lead at the very least to more explanations, but which could also raise concerns about their motives for this article.
12
Topol and Krumholz are cardiologists, doctors, they understand the significance of this study, they really care about the hypertensive men and women who would live longer if the message of the study is accepted. The delay means lives lost.
If all people who were going to read the "accelerated data presentation", as the Topol and Krumholz suggest, were like Topol and Krumholz themselves, good willing, intellectually honest men/women and scientists, with no other interest other than well-being of mankind, the proposal would be acceptable. But as we all know it is not so.
If all people who were going to read the "accelerated data presentation", as the Topol and Krumholz suggest, were like Topol and Krumholz themselves, good willing, intellectually honest men/women and scientists, with no other interest other than well-being of mankind, the proposal would be acceptable. But as we all know it is not so.
So the authors are advocating a data dump absent context or statistics. The public can then, what, do a keyword search and see how many of the thousands of patients died of cancer and create an erroneous conclusion? Run T tests (if they've taken college stats)? And apparently to heck with long term results.
12
I am wondering if Drs. Topol and Krumholz don't have an intellectual conflict of interest in this matter. Have they previously published a contrary opinion? Are they upset that they weren't included in this trial? I can't say that they do, but (as reflected by the many other commenters) they should know that their recommendations here are irresponsible. One can't help wonder if there are other reasons why these two are eager to get their hands on the data.
9
I agree with the strategy, just not the tactics. Now is a crossroad in how we use information technology to accelerate data analysis and the peer review process. The ability to disseminate raw data quickly to a multitude of independent groups for collective analysis would be the ideal situation. I believe, and the authors seem to contend, that the internet could be a useful tool for this peer review revolution.
2
Medical studies have to be carefully reviewed before publication, otherwise they risk becoming marketing collateral for people with a stake in the financial outcome.
A more novel approach to getting needed medicine faster out of the pipeline and to people who need it was recently proposed: Tailor the approval process to the severity of the disease and the lack of other effective forms of treatment. A person with highly treatable Type II diabetes, for instance, doesn't need a new drug on the market as fast as someone with lethal forms of pancreatic cancer, which don't have effective treatments. Check out this article: http://fivethirtyeight.com/features/how-the-fda-could-change-the-way-it-...
A more novel approach to getting needed medicine faster out of the pipeline and to people who need it was recently proposed: Tailor the approval process to the severity of the disease and the lack of other effective forms of treatment. A person with highly treatable Type II diabetes, for instance, doesn't need a new drug on the market as fast as someone with lethal forms of pancreatic cancer, which don't have effective treatments. Check out this article: http://fivethirtyeight.com/features/how-the-fda-could-change-the-way-it-...
4
There are two sources of delay in the publication of scientific studies. One is the time required for the authors to digest, carefully consider, and present their results, and to have them reviewed by independent peers. The other is the time required for the publication process in professional journals.
Scientists have become increasingly impatient with the second of these for quite some time. There are growing efforts within the scientific establishment to take the commercial publishing companies out of the process, and the internet presents an alternative means for releasing scientific reports. After all, the crucial work, such as peer reviewing, is done for no recompense by scientists themselves.
If this is what the authors of this article are referring to, then they can only be applauded.
But if they are suggesting that the process of scientific deliberation be hastened or compressed, then this must be treated with extreme caution.
Once the results are out there, it is difficult to get them back in the bottle. As the case of the spurious linkage of vaccines to autism illustrates.
And let us not forget thalidomide. It was only the warning of an influential researcher in Australia which saved that country from experiencing the plague of birth defects which occurred in the US as the result of its approval.
Scientists have become increasingly impatient with the second of these for quite some time. There are growing efforts within the scientific establishment to take the commercial publishing companies out of the process, and the internet presents an alternative means for releasing scientific reports. After all, the crucial work, such as peer reviewing, is done for no recompense by scientists themselves.
If this is what the authors of this article are referring to, then they can only be applauded.
But if they are suggesting that the process of scientific deliberation be hastened or compressed, then this must be treated with extreme caution.
Once the results are out there, it is difficult to get them back in the bottle. As the case of the spurious linkage of vaccines to autism illustrates.
And let us not forget thalidomide. It was only the warning of an influential researcher in Australia which saved that country from experiencing the plague of birth defects which occurred in the US as the result of its approval.
5
Correction. It was indeed an Australian researcher who blew the whistle on Thalidomide, but it was already being used there and in many other countries. It was the USA that largely escaped the plague due to the efforts of an FDA inspector.
2
Good point.
McBride was not an influential researcher in Australia who reported the problems with thalidomide.
He was a mainstream OB GYN who reported an observation; much to his credit
and his patients benefits!
McBride was not an influential researcher in Australia who reported the problems with thalidomide.
He was a mainstream OB GYN who reported an observation; much to his credit
and his patients benefits!
2
sorry to be pedantic.
But the FDA inspector, Dr Kelsey, largely based her concerns
on the letter McBride wrote to the Lancet journal.
McBride ( now dead) was not too smart and pompous, who based the letter on an observation of his resident.
He later lost his license for fudging lab results on rabbits and debendox.
BUT he did report thalidomide and SAVED many thousands of children from this dreadful drug.
But the FDA inspector, Dr Kelsey, largely based her concerns
on the letter McBride wrote to the Lancet journal.
McBride ( now dead) was not too smart and pompous, who based the letter on an observation of his resident.
He later lost his license for fudging lab results on rabbits and debendox.
BUT he did report thalidomide and SAVED many thousands of children from this dreadful drug.
1
If snake oil salesmen are going to rake in the dough with both fists under a national single payer system, where all medicine and medical treatment will be thoroughly politicized, that whole deliberate peer review thingie will have to go.
1
So established scientific method should yield to instant gratification?? Why not just tweet the data as it is obtained ........
9
In the mid 1960s, the US Veterans Study followed patients treated and untreated for hypertension to find out if treatment improved health outcomes. After one year, 27 of 70 untreated patients sustained strokes, versus 2 of 70 in the treated group. The medications were primitive in the 1960s, but the benefits were unambiguous.
1
I used to give lectures on cancer chemotherapy. Once for my final lecture I talked about some results that looked especially promising and exciting that had just been published in the New England Journal of Medicine. A colleague who knew more about this than I did got up and explained to the students that the results were still very preliminary and they should not get too excited about them. After the lecture my colleague explained to me in private that the study had been hyped up and was not as promising a cure as had been presented by the authors of the study. He did not want the students to return home and start telling people with cancer about this marvelous new cure and having them trying to get into the NIH program.
Just today there were news in Australia about fabrication of results and retraction of papers by a cardiologist.
The internet has not changed science. There will always be those who try to make more of their results than are justified and those who will fabricate data which is why it is essential especially for clinical studies that time be given to insure the validity of the results and that they are applied appropriately to patients.
Just today there were news in Australia about fabrication of results and retraction of papers by a cardiologist.
The internet has not changed science. There will always be those who try to make more of their results than are justified and those who will fabricate data which is why it is essential especially for clinical studies that time be given to insure the validity of the results and that they are applied appropriately to patients.
3
Who will train journalists not to do sensationalism with medical breakthroughs?
6
I favor early publication that is free and open source, with data in digital form that can be used with a free and open source statistical package. For important papers, some proprietary journals will publish early and free and open. These papers are produced with taxpayer money and should be free to all, especially if they concern public health, life and death for all.
However, early termination of a study does not necessarily imply that it should be published early. Termination is based on sufficient statistical power achieved, not necessarily positive results. An early end also means that we won't know from this study what happens to subjects for more than a few years, as they won't be followed; another study will have to be done to follow more subjects longer, or if not, a retrospective study after some years. Extrapolating from some 9,000 subjects to ALL people with high blood pressure and requiring them to take three drugs might require more statistical power, but might be too expensive. Perhaps the answer would be to have all of us collect data about our own BP and decide what to do with a personal physician, making the "lifesaving" guidelines and the haste to publish secondary.
However, early termination of a study does not necessarily imply that it should be published early. Termination is based on sufficient statistical power achieved, not necessarily positive results. An early end also means that we won't know from this study what happens to subjects for more than a few years, as they won't be followed; another study will have to be done to follow more subjects longer, or if not, a retrospective study after some years. Extrapolating from some 9,000 subjects to ALL people with high blood pressure and requiring them to take three drugs might require more statistical power, but might be too expensive. Perhaps the answer would be to have all of us collect data about our own BP and decide what to do with a personal physician, making the "lifesaving" guidelines and the haste to publish secondary.
3
"Medical scientists shouldn’t wait until their data are chiseled in stone to publish results."
If this was a thing -- useful knowledge, cures, and treatments being held up , it would be terrible. But as it is, most published medical research can't be reproduced. That goes from everything from cancer to periodontics. The last thing we need is a proliferation of nonsense providing new "market" incentives for bogus treatments and medical rent-seeking in the currently malignant U.S. health care system.
If this was a thing -- useful knowledge, cures, and treatments being held up , it would be terrible. But as it is, most published medical research can't be reproduced. That goes from everything from cancer to periodontics. The last thing we need is a proliferation of nonsense providing new "market" incentives for bogus treatments and medical rent-seeking in the currently malignant U.S. health care system.
5
The real problem with the blood pressure results from the Sprint trial isn't how quickly the rest of its facts can get released, it's whether the science behind the theory is sound. Is one number, one marker, systolic blood pressure, enough to decide that aggressive treatment is needed?According to the doctors involved in this study, yes. These sorts of one-size-fits-all medical solutions rarely pan out in the long run.
The easiest way to lower systolic blood pressure is with medicine. Exercise and weight loss also help, but maintaining these approaches is always a stretch for many people. So, the pharmaceutical industry will be delighted with this new, aggressive treatment of systolic blood pressure. How doing so effects overall mortality rates is going to be a whole other study, one which will have to wait. It wouldn't surprise me if overall mortality rates hardly inch down at all, even after mass medication treatments of the affected populations.
2
Appreciate all the comments. The point is that if data are hardened enough to make the decision to stop a multi-million dollar study and the results are considered groundbreaking, then why not release the data upon which that decision is made. Surely we want a careful analysis of the final data... but there is no reason not to be transparent about what is known. And the info about such a trial is likely to evolve over years with ancillary publications. And the recently published BMJ paper on Paxil shows that re-analysis by other investigators can even provide different interpretations. Science is progressive; not static. The main trial results should be easy to report; understanding the nuances will take a lot of time. And the reason for the issue here is because the result is considered lifesaving by those who have seen the results.
Yes! In this era of instant communication- we could now make mistakes even faster- and bigger- than ever before! Full speed ahead! Or- whichever way it is we're heading.
6
This is embarrassing for these two cardiologists and one wonders about their actual research vs clinical backgrounds. Skipping the rigorous peer review process should only be done in very rare circumstances. Otherwise, we return to a medieval non-evidenced based approach to medicine, where we simply accept and disseminate unsubstantiated findings.
59
They aren't suggesting skipping completion and peer review. They're suggesting releasing a little more detail and color at the study-stop announcement, which today is most often restricted to the length and depth of a single press release of 1-2 pages.
One wonders if this is the same Eric Topol who, according to Dollars for Docs, received close to half a million dollars from between August 2013 and December 2014. At the least, it could make him more sympathetic to the point of view of drug companies. But then, according to the same source, Harlan Krumholz received only $26 during that time.
1
I, for one, am sick and tired of bogus news about amazing new breakthroughs regarding Alzheimer's Disease, Lou Gehrig's Disease and other severe maladies about which doctors are essentially clueless. Let them do their work. Let it be funded. And let them shut up about it, until they really have something to say.
56
Rush the results? That's how the drug companies would like to do it, so they can market their products as quickly and widely as possible. Of course, you end up with results like you got with Thalidomide, or Risperadal (see today's Nick Kristof column on the subject).
55
It's worth noting that while many states have legalized marijuana for medical uses, there is very little research supporting its use for any of these conditions. It seems the medical marijuana proponents have no trouble recommending its use despite this major deficiency. It's not only drug companies that would like to rush products to the market before there is evidence to support their use.
Papers, even after peer review, have not been living up to the standards they once were. While high profile data like this one are not going to be reviewed as casually, premature release without the thorough review is a mistake. The NIH expert panel's review should be the first step in expediting the peer review and publication, not short cut it. Preliminary conclusions are out there in popular press. Patients can aim to reduce their blood pressure by changing their dietary habits and life style before the physicians are ready for medical intervention to implement the guidelines that are thoroughly vetted. Jumping the gun with hard to reverse practices is the last thing we need.
20
"The percent reduction means little without knowing, in absolute numbers, how many people benefited or how many suffered a stroke, heart attack or died. We also don’t know about the risks of aggressive treatment or the types of patients who derived particular benefit."
It is difficult to believe that such a statement came from a professor of such high regard. The NIH would not stop a study based on only a few patients since the results would likely be statistically invalid. Further the NIH does not analyze all the secondary outcomes.
When a study is stopped early, much is sacrificed, including some secondary outcomes such as the ability to identify subsets of patients who may benefit, or the types of benefits derived from an intervention. Analysis of these outcomes is severely hampered by the inherent decrease in the number of patients analyzed, and the decreased time over which data were collected.
Also, these types of analysis take time and need to be done carefully. Data needs to be circulated among key investigators to insure proper integrity and interpretation.
It is difficult to believe that such a statement came from a professor of such high regard. The NIH would not stop a study based on only a few patients since the results would likely be statistically invalid. Further the NIH does not analyze all the secondary outcomes.
When a study is stopped early, much is sacrificed, including some secondary outcomes such as the ability to identify subsets of patients who may benefit, or the types of benefits derived from an intervention. Analysis of these outcomes is severely hampered by the inherent decrease in the number of patients analyzed, and the decreased time over which data were collected.
Also, these types of analysis take time and need to be done carefully. Data needs to be circulated among key investigators to insure proper integrity and interpretation.
20
For two cardiologists at premier institutes to pen such a naive column is rather surprising. Yes, of course, let's rush to publication. No need to carefully review the data, ensure it is accurately captured and interpreted. And later, when a more careful review is done and problems with the data or interpretation arise, no big deal. Come on--I assume both physicians have participated in clinical trials. Otherwise, their opinions are not exactly based on knowledge. With the biomedical literature awash with irreproducible results, we do not need to add to the lower quality of the literature by rushing to publication. As anyone who has participated in clinical trials knows, the data need to be carefully QC'd, interpretations need to be considered. Just dumping data on a website for anyone to analyze has appeal, I understand. But it is likely that this approach will result in all sorts of inappropriate analyses and mis-interpretations. Better to be certain rather than rush.
73
While I wasn't surprised to see an article like this in the first place, I was surprised to see it written by two people who should know better. Rushing to proclaim results, especially results that haven't even been through peer-review, is absurd.
27
As a former statistician in the pharmaceutical industry, I would like to offer a big AMEN to your comments, Mr. Kuhstoss. Well said. It is usually especially complex and time-consuming to tease out the full picture of the risks; and heaven help us if data base updates or re-interpretations cause that assessment to have to be modified post-publication.
16
I don't think a wholesale early data dump is a good idea either. But it seems to me that the gist of this recommendation: the rationale for an early study halt should be included in, or posted to accompany, the press release about the early study halt, in order that its significance can be properly understood. That's a smaller-scale proposition and it seems eminently sensible.
1
It seems to me that the problem here is not that the details haven't been released, but that the broad results (i.e. what was reported by the media) were released too early. I agree that they should come out roughly simultaneously, but definitely disagree that the details should be rushed out.
Of course a delay could cause the deaths of certain individuals. But a premature release could as well. Caution is warranted here no matter how impatient you are to see the details of the study. We'll all have to accept that whatever some day kills us could be curable the next. Even so, releasing the details/recommendations after cautious and deliberate thought is much better for society in the long run. Just because technology has made life appear to move fast, does not mean we have to make life and death decisions in a split second. Relax and wait for the results.
Of course a delay could cause the deaths of certain individuals. But a premature release could as well. Caution is warranted here no matter how impatient you are to see the details of the study. We'll all have to accept that whatever some day kills us could be curable the next. Even so, releasing the details/recommendations after cautious and deliberate thought is much better for society in the long run. Just because technology has made life appear to move fast, does not mean we have to make life and death decisions in a split second. Relax and wait for the results.
15
I disagree with the authors' recommendation. The medical community already suffers from public skepticism from perceived flip-flops (most or which are preliminary correlations reported as established cause and effect). Waiting till the researchers have fully analyzed the data and peer review is complete will help prevent some major retractions that are certain to occur otherwise.
36
Though it would be great to shorten the timeline from when studies are concluded to when they are published, part of the reason it takes as much time as it does is that these studies generate huge troves of complex data. This information needs to be thoughtfully processed. This takes time.
As it stands, medical knowledge published in the peer reviewed literature is of varying quality. Adding to that with an unprocessed, non-peer-reviewed data dump will only add to the confusion and perhaps allow people to seek out results that support their own beliefs (e.g. a link between vaccines and autism).
If it takes an extra year (but still less than the 8 years originally planned) to allow physicians and scientists to carefully analyze this data and better understand who will benefit from more aggressive BP reduction, that is ok. Releasing the top line summary statement early (fewer heart attacks and strokes) helps build the story for physicians and patients alike that getting BP under strict control is important. Then, when more data is analyzed, the treatment guidelines can be fine tuned.
As it stands, medical knowledge published in the peer reviewed literature is of varying quality. Adding to that with an unprocessed, non-peer-reviewed data dump will only add to the confusion and perhaps allow people to seek out results that support their own beliefs (e.g. a link between vaccines and autism).
If it takes an extra year (but still less than the 8 years originally planned) to allow physicians and scientists to carefully analyze this data and better understand who will benefit from more aggressive BP reduction, that is ok. Releasing the top line summary statement early (fewer heart attacks and strokes) helps build the story for physicians and patients alike that getting BP under strict control is important. Then, when more data is analyzed, the treatment guidelines can be fine tuned.
47
I do not agree with the reason you suggest for studies supposedly taking so much time.
For one thing, a study of a new treatment must not only ensure that the treatment works, but what are the side effects, some of which can take a long time to show up.
For another thing, just as with any contractors, the more a study goes over budget — and an easy way for this to happen is for it to go over the projected time constraints — the more money the researchers, or at least their sponsoring institution, will be paid.
Computers, on the other hand, can implement the latest statistical techniques on tons of data in the blink of an eye.
It is extremely important that all financial considerations be removed from studies of treatment effectiveness, whether through clinical trials, longitudinal studies, or observational data.
For one thing, a study of a new treatment must not only ensure that the treatment works, but what are the side effects, some of which can take a long time to show up.
For another thing, just as with any contractors, the more a study goes over budget — and an easy way for this to happen is for it to go over the projected time constraints — the more money the researchers, or at least their sponsoring institution, will be paid.
Computers, on the other hand, can implement the latest statistical techniques on tons of data in the blink of an eye.
It is extremely important that all financial considerations be removed from studies of treatment effectiveness, whether through clinical trials, longitudinal studies, or observational data.
The need to "thoroughly process" data does not generally delay release of the data. Competent experiments are well planned with analytical routines determined and tested before data is even acquired. If there is enough information to release a tentative conclusion, then there is tabulated data that can be published to a website. Publication of data is delayed by the desire to be the first to publish an article in a journal. Indeed, in many cases, the data underlying an article is never published. The authors are correct to criticism this practice and advocate that key information be released to peer review quickly.
We already suffer from science by press release, especially in the increasingly bitter competition for grant money. I would be very concerned if the deliberate approaches medical science usually takes was short circuited to run in "Internet Time". While some studies will produce profound results that warrant immediate release because they can save lives I suspect that those are fairly rare. Without retesting and replication of results, we really don't have a way to confirm (or refute) a study. I'm not sure that without a clear demonstration of a benefit and a review by peers that a shortcut would do anything except put people at risk.
28
It might be useful to offer up the myriad of prior studies that contradict this finding, offer caveats in wholesale medication for 130-140 BP which had until last week been perfectly acceptable, and a disclaimer as rigorous as that offered for many of the meds that the cure is not without risk. What is the number to cure here (how many folks without issues will benefit from a 120 BP given the risks of the meds?). Frankly, at 70 years of age, with acceptable blood pressure, the idea of saving myself for cancer and/or dementia is not appealing. BP is just, after all, a number it is not the key to immortality.
7
Just to clarify, a blood pressure of 130-140 has not been perfectly acceptable. for the last many decades the standard for systolic blood pressure has been 120 mmHg. Up until about 10 years ago, a systolic blood pressure of 130-140 mmHg termed "high-normal". However, patients weren't making the changes in lifestyle to bring blood pressure down because they viewed themselves as still being in the normal range. So the powers that be (American Heart Association, American Medical Association, etc) changed the terminology to refer to this level of blood pressure as "pre-hypertensive", in an attempt to try and motivate people to reduce their systolic pressure to at least 120 mmHg. So 130-140 hasn't been deemed by clinicians to be perfectly acceptable for decades.
1
I disagree for another reason. My concern is that the trend toward stopping trials early (which seems to have begun before the social media era) will only accelerate. This is a problem because, despite the apparent benefit of an intervention, significant risks can emerge over time that will be obscured by early stoppage. Trendlines can reverse. The decision to stop a trial early is an ethical/moral one, but if the science ultimate turns out to be flawed, nobody wins. This a dangerous trend.
16
There are also issues with the statistical analysis of trials that are stopped early. The number of adverse events in the two groups of patients varies randomly over time in addition to whatever the effect of the treatment is. Halting the study at the point where the difference between the groups is maximal instead of at the pre-determined endpoint will necessarily exaggerate the benefit or harm of the treatment and stopping greatly complicates the statistics.
1
No, the last thing we need are more premature, unrepeated results released to the lay public, creating false hope and cementing bogus information into people's minds that is impossible to remove. In fact, many of the current scientific studies out there turn out not to be true upon later attempts to replicate the results with similar studies. We already have too many breathless articles in the popular press telling us stuff that is either just not true, or is only half true, with many downsides. Early releases of studies will mostly just create a din of noise as speculative findings are made public, followed by results of later studies, often years later, totally debunking the first study. This does not serve the public, it would only make us more cynical, waste time and enegy, and damage our health.
35
By coincidence, today's NY Times provides a perfect example of results of medical studies that were embraced, spawned whole industries, changed diets, and now appear false. It turns out that the idea that Omega-3 fatty acids are probably not the panacea for heart health that had been thought. http://www.nytimes.com/2015/09/22/science/inuit-study-adds-twist-to-omeg...®ion=Marginalia&src=me&pgtype=article
This is not based on mere experimental evidence, the researchers actually found that genetic differences are more likely why the Inuit people have such good blood levels of healthy fats. (Of course, it may turn out to be more complicated, wait for more studies) It is not due to their Omega-3 rich diet after all. So, if you are European, forget about those Omega-3 supplements, you have been misled by those who had assumed the correlation between Inuit's remarkably low heart attack rate was due to Omega-3. Think how much time, effort, money, and hope has likely been wasted eating excess Omega-3. Michael Polan was right when he said not to trust nutritional studies in his book, Omnivores Dilemma.
This is not based on mere experimental evidence, the researchers actually found that genetic differences are more likely why the Inuit people have such good blood levels of healthy fats. (Of course, it may turn out to be more complicated, wait for more studies) It is not due to their Omega-3 rich diet after all. So, if you are European, forget about those Omega-3 supplements, you have been misled by those who had assumed the correlation between Inuit's remarkably low heart attack rate was due to Omega-3. Think how much time, effort, money, and hope has likely been wasted eating excess Omega-3. Michael Polan was right when he said not to trust nutritional studies in his book, Omnivores Dilemma.
2
Dr.Topol I hold you in great respect, and I am certain that you would bring a sophisticated level of personal analysis to the raw data but I can't agree with your plea to make the data and the results of the NIH hypertension trial available before the researchers say they are ready to present their work. Surely you can recall times when findings have been misinterpreted or miscommunicated. You can also recall when medical misinformation had proved itself to be extremely difficult to retract, modify and correct once it has emerged into the internet. There is no problem with the NIH publicizing that a study has been concluded early and that formal results will follow. Long ago, the ACAS study concluded early, but the data and conclusions needed to be thought through carefully
before recommendations would be made to change the direction of medical and surgical intervention for carotid stenosis. A prudent practitioner wouldn't act
until the recommendations are clear and the individual patient situation is reassessed. I would support saving the urgent dissemination of NIH study data only for the most unusual, emergent situation--as in, again long ago, the announcement of the efficacy of AZT for HIV.
before recommendations would be made to change the direction of medical and surgical intervention for carotid stenosis. A prudent practitioner wouldn't act
until the recommendations are clear and the individual patient situation is reassessed. I would support saving the urgent dissemination of NIH study data only for the most unusual, emergent situation--as in, again long ago, the announcement of the efficacy of AZT for HIV.
41
I assume this BP study is correct, as it is NIH funded and the investigators are field leaders, yet the principal this commentary is spot on. If the prelim data is strong enough to stop a trial and ready for a press release, then the results should be shared in a meaningful way with the public. Not doing so puts out a medical recommendation without the data and results to evaluate it.
Of note, in the 1990's the NIH prepublicized a large trial about giving high dose steroids for traumatic spinal cord injury. The treatment was rapidly adopted. When the data were released many physicians were not convinced, yet it took 20 years to reverse the practice that was begun by the early press release.
more info here: //bit.ly/1iD92AE
Of note, in the 1990's the NIH prepublicized a large trial about giving high dose steroids for traumatic spinal cord injury. The treatment was rapidly adopted. When the data were released many physicians were not convinced, yet it took 20 years to reverse the practice that was begun by the early press release.
more info here: //bit.ly/1iD92AE
7
1. It is well known that keeping ones blood pressure under adequate control is essential to ward off congestive heart failure, myocardial infarction, renal failure to say the least.
2. This study seems to suggest that one needs to bring his/her blood pressure near normal" of 120/80 as we all know. Unfortunately, many seniors do not tolerate this. The problem is as we age our blood vessels become a whole lot more rigid compared to an age of 20. This causes one to have syncopal episode if they get up suddenly from a lying down or seated position.
3. This study adds more drugs to the current therapy. Does it address other modalities such as weight reduction, diet modification (which most of MY patients did and had great results) etc?
4. There is a great discussion about this topic at one of the "doctor" sites.
Majority of the doctors discussing this topic have similar opinion such as mine.
5. lastly, was this study funded by any of the drug companies? if yes, there is a significant "conflict of interest" . I will do my" research" and will let you all know what comes out of this.
2. This study seems to suggest that one needs to bring his/her blood pressure near normal" of 120/80 as we all know. Unfortunately, many seniors do not tolerate this. The problem is as we age our blood vessels become a whole lot more rigid compared to an age of 20. This causes one to have syncopal episode if they get up suddenly from a lying down or seated position.
3. This study adds more drugs to the current therapy. Does it address other modalities such as weight reduction, diet modification (which most of MY patients did and had great results) etc?
4. There is a great discussion about this topic at one of the "doctor" sites.
Majority of the doctors discussing this topic have similar opinion such as mine.
5. lastly, was this study funded by any of the drug companies? if yes, there is a significant "conflict of interest" . I will do my" research" and will let you all know what comes out of this.
1
These circumstances are not similar simply because a spinal injury is so devastating and there are no treatments available. Anything that might help was tried because the other options were merely stabilizing the fracture. Physicians grasped at straws. It turned out the straws didn't pan out. So what are we left with? Nothing.
Plenty of evidence abounds, should an investigative reporter inquire.