There’s an extraordinary new article in The Atlantic, “Lies, Damned Lies, and Medical Science.” It echos the excellent article in our Journal of Participatory Medicine (JoPM) one year ago this week, by Richard W. Smith, 25 year editor of the British Medical Journal: In Search Of an Optimal Peer Review System.
JoPM, Oct 21, 2009: “….most of what appears in peer-reviewed journals is scientifically weak.”
Atlantic, Oct. 16, 2010: “Much of what medical researchers conclude in their studies is misleading, exaggerated, or flat-out wrong.”
JoPM 2009: “Yet peer review remains sacred, worshiped by scientists and central to the processes of science — awarding grants, publishing, and dishing out prizes.”
Atlantic 2010: “So why are doctors—to a striking extent—still drawing upon misinformation in their everyday practice?”
Dr. Marcia Angell said something just as damning in December 2008 in the New York Review of Books: “It is simply no longer possible to believe much of the clinical research that is published, or to rely on the judgment of trusted physicians or authoritative medical guidelines. I take no pleasure in this conclusion, which I reached slowly and reluctantly over my two decades as an editor of The New England Journal of Medicine.” (Our post on Angell is here.)
What’s an e-patient to do?? How are patients supposed to research if, as all three authorities say, much of what they read is scientifically weak?
More problematic, what’s an e-patient to do when doctors commonly insult them, saying “You don’t know how to research – stick to peer reviewed journals”? You know what reaction patients get when they question those journals? Commonly, doctors’ eyes roll. Both are reasons why we’ve covered this subject forever. See our category Understanding Statistics.
I haven’t read the full Atlantic article yet, but today it was the buzz of Twitter. I’ve asked for a post by Peter Frishauf, who authored a great commentary on Smith’s article last year: Reputation Systems: A New Vision for Publishing and Peer Review. Stay tuned.
The Atlantic article goes further than criticizing the publishing process in medical research. It criticizes the research process itself. On one level, bias can be introduced in study design for randomized controlled trials (RCTs)by overeager scientists who want their studies to confirm their hypotheses. But the article also alludes to deeper problems at the core of our current medical research system that holds up RCTs as the gold standard.
I think we’re at the threshold of new models of medical research that can account for more complex interactions and can be personalized for more sub-population groups. The Archimedes Model is one example of a new approach.
The availability of longitudinal outcomes data from large populations via patient and disease registries made possible through collection of data in EHRs is a big factor in making new types of research possible. I’m excited about the research possibilities with new rich data sources being made available. But, I don’t think the research methodologies have been worked out yet.
With respect to the medical publishing community, I don’t think they’re prepared for changes that are underway in research projects that are based on shared data sources and collaborative research, where the focus is on developing models/algorithms that can be applied directly to clinical decision support systems. Reading articles that summarize new research findings won’t be the primary means of information dissemination in the future.
Great comment, Janice. I get it about questioning the research process. What do you think about last year’s JoPM pieces?
since I provided one of last year articles I feel a need to respond.
George Lundberg says it best: “Journals are so Twentieth Century!”
Medicine is going to switch pretty fast into a group of rapid learning systems where the RCTs and attached publications will just be one of many forms of knowledge dissemination. The ACOR groups and PLM sites are both very early examples of what’s coming. As the data silos are crumbling faster than anybody expected even a year ago, we are closer to the fall of the Berlin wall in the word of medical science and knowledge.
I have a lot to say about the current peer review journal publishing model. Too much to summarize in a few words and within the timeframe I have today. But, since I always consider the business model when I look at a publishing operation, I can say that I see problems in the current funding model for medical research and publishing, because so much of it is funded directly or indirectly by Pharma. Ditto for continuing medical education and most patient education materials–especially online consumer health sites.
One other point for now (we can continue this discussion for a looong time): I think access to underlying data sets and the algorithms that were used to generate the research results will become more important than reading journal articles that summarize some aspect of the research. In time, the reward system will change, too, so that data analysis achievements will overtake written articles as the measure of quality in research. CVs may list algorithms created by a researcher that have been adopted in clinical practice instead of a list of articles in high-impact factor journals.
Wish I could lay out my thoughts with more clarity today, but I’ve got to get to my to-do list.
RE: I think access to underlying data sets and the algorithms that were used to generate the research results will become more important than reading journal articles that summarize some aspect of the research. In time, the reward system will change, too, so that data analysis achievements will overtake written articles as the measure of quality in research.”
From your lips to God’s ears, Janice. The skill set required to analyze data is very different from the one required to publish data. Currently, there is a premium on publication and very little attention to data analysis/re-analysis and sub-group stratification.
Like others have intimated. The publishing skill set is rewarded with headlines. The analysis skill set is rewarded with workable, interactive, individualized data.
Sign me up for any movement that gives us more of the latter.
Janice, I think you are absolutely correct with regard to the limitations on RCTs for providing individual answers and the possibility that larger data sets of (optimally) patient entered data could account for more complex interactions and provide individualized answers.
While I still see a place for RCTs I recognize that there is room for other types of data collection and research.
The challenge will be (as you’ve properly stated) making those in the research and publishing field recognize the value of other methodologies while implicitly and explicitly questioning the current foundations of medical and scientific orthodoxy.
Hopefully we’ll fare better than Galileo in this regard…
Janice, please tell us more about the Archimedes model.
Thanks for posting this, Dave. I’ve read and re-read the Atlantic article and find the discussion it has spawned online just as fascinating as the points that come up in the article.
As a physician; I had to struggle to overcome the initial “rolling of the eyes- reflex” at a story in the popular press which indicates that most peer reviewed research is useless.
However, the great majority of the issues pointed out in the article are (indeed) correct. I would hasten to add that many of the criticisms that Dr. John Ioannidi elucidates in the article vis-a-vis peer reviewed research are shortcomings that are recognized in the academic clinical world and which many of us struggle to rectify in practice, with students and (most importantly) in speaking to students.
The article rightly points out that conflict of interest is a serious problem in much research. It is correct in stating that there is limited hard research with which to guide the treatment of rare diseases.
I’m sure that most patients would be shocked to learn the number of times a major research finding is reversed by a subsequent study in medical journals.
I wouldn’t therefore be surprised if (after reading about the aforementioned problems)an aspiring e-patient threw up his or her hands in exasperation!
My feeling, however, is that this is a classic baby and bath water problem. It *is* possible to sift through the flaws and problems of the peer review bath water and keep the research baby to the benefit of patients.
Part of the fix to this problem is a matter of having appropriate filtering mechanisms for what is truly important in clinical research.
I suspect our information technology companions will find better ways for us to filter sound and noise in the medical research field and to make pertinent issues rise to the forefront for individualized care. Indeed, resources like this site and the SPM are integral in the process of making research in general important and relevant to the patient in particular. No research study will provide individualized patient recommendations. Such studies are aggregates. Your mileage will vary based on many other factors.
As I’ve said before, if your doctor is basing your management on the “most recent research” he read in a journal yesterday… get another doctor.
Another issue is the appropriate interpretation of data. I think it’s unfair for the author to state that “Much of what medical researchers conclude in their studies is misleading, exaggerated, or flat-out wrong.”
Talk about a cheap shot!
Context is everything. The language of research is (by necessity) limited in scope and specifically addresses the research question at hand.
Unfortunately, the way in which too much research is presented to the public loses any of the nuance found in the original research. Thus we have the evening news reports where something is “good” for us today and “bad” for us tomorrow.
As someone who has had my own research reported by others I can assure you that people in my position CRINGE when we hear the results of our diligent studies described in shorthand as being good or bad.
The scientists, doctors and media are all speaking different languages where research findings are involved and the patient loses in the end.
As always though, I remain hopeful that the emerging technologies we have at the present time will eventually move in the direction or remediating such problems and provide improved care to patients everywhere.
Hi Chuck – you spark so many thoughts!
> classic baby and bath water problem. It *is* possible to sift through the flaws and
> problems of the peer review bath water and keep the research baby to the benefit of patients.
Indeed, as Smith said in the summary of his JoPM piece, “It would be a bold funding body or journal that abandoned peer review, but could we at least do better?”
The teachable moment for patients (and providers IMO) (and mass media journalists!) is that we need to learn to filter *everything*, including the most highly respected journals – and indeed our whole concept of Phase 1, 2 and 3 trials as means of getting increasingly better answers.
Here’s an example that irritates me personally because it affects MOST Stage IV kidney cancer patients: Interleukin. I’m not personally an expert but here’s what I understand:
This is such a horrible failure of the “scientific method” (as applied in our culture) that it makes me sick. A community doctor who does this is blindly trusting what s/he was trained to trust. And an unwitting trusting patient can get seriously screwed as a result.
All I want is for people to open their fricking minds to the reality that knowledge is not nearly as certain as they think – including knowledge from the sources we were taught to trust.
Glad you want to read Dr. Zuckerberg’s blog, because she has a post on a phenomenon you know about with Interleukin, briefly stated in a quote from it below:
“The vast majority of drugs – more than 90 per cent – only work in 30 or 50 per cent of the people,” Dr Roses said.
Here is the entire post of hers:
I think, as her post indicates, we are just realizing we need to develop clinical trial designs which take into account this phenomenon.
[Editor: I moved this correction into the previous comment, where it belonged.]
Well, Dave, I need another correction, because I inexcusably got Dr. Zilberberg’s name wrong this time around..my sincere apologies.
btw, Chuck, what do I / we do with docs who sneer at patients who want to discuss the validity of research??
I know of a couple whose superb surgeon responded to subsequent questions by saying, angrily (they say), “I need patients who don’t question my judgment” and stalked out. (He eventually returned, but still…)
Honestly: what will it take to open the minds of these trained professionals to the idea that WHAT THEY WERE TRAINED TO TRUST is still subject to critical scrutiny?
Is there an effective approach to this conversation?
That is actually an excellent question, Dave. I’ve found myself being the recipient of such patients. We speak about the concept of social media “conversations” that I think its easy to forget that in the eyes of many physicians a doctor’s visit is NOT a conversation but rather a soliloquy.
Unfortunately many physicians feel that a patient asking questions is a challenge to their authority.
I’m still struggling to come up with solutions at this point. One thing I would recommend is for the patient to make it clear they they are not asking questions to be confrontational. (It shouldn’t be necessary but it may soothe the doctor’s ego and facilitate conversation). I also think that when patients are frank regarding the fact that they may seek a second opinion they can preemptively get the doctor’s attention and more individualized care.
Regardless of the technique, it’s important to keep in mind that dealing with an e-patient is going to be a learning experience for the physician as well. Not all educational experiences are comfortable; but if done properly, they can be beneficial for both parties.
I’d love (and it would be perfect as a Society activity) to develop a 2-3 minute YouTube modeling how it can work.
We already have a counter-example – the doc at the start of the Seinfeld episode in our playful-but-serious HIPAA post. :–)
I love everything you write but I have to disagree with you on this one.
As long as patients have to provide an excuse for being what they should have been all along, ie autonomous, we can’t expect a real reform of the system.
It is now the responsibility of health professionals to engage educated and activated patients as full partners. Anything less is just not acceptable and if we keep on being polite we will just help to retain the status quo ante.
Doctors must understand that educated patients are better patients. How can we convince them? Via their wallet. I think Martin has asked on Twitter the most interesting question, one that we should research ASAP: are e-patients more or less litigious than non e-patients.
This is really a reply to Gilles’ reply to Chuk (this is confusing!), but I would say that although Gilles is correct about patients being autonomous, there is no reason to be deliberately confrontational about it. As someone who was known to be sometimes tactless myself at work, I learned the hard way that there is a good way and a bad way to get what you want. Yes, many docs have over-active egos, but many also complain of lack of respect from patients who tell them “I just wasted my co-pay” or other such remarks. I think respect needs to go both ways. Dave’s idea of a how-to video is a great one; maybe docs need one too!!
Bev, what on earth made you think I was talking about a video to only teach patients??:) Sorry for giving that impression.
I often say this is like learning a new dance. Hard to demonstrate with just one party.
Check out the CDC training video on hand washing. This one’s directed at newly admitted patients, but does demonstrate scenarios.
To me it’s no more complicated than an architect working with a client on blueprints for a new deck. With my primary Danny Sands, I like the give & take of “Have you heard about x?” “Yes, don’t go there.” “Have you heard about y?” “No, let’s look.”
Re the CDC video, some people don’t like the scene where someone says, “Doctor, I’m embarrassed to ask you this…” but if we want to create change we need to reach people in their current state of mind. That’s messaging 101: you get no points for speaking at a level where your audience isn’t.
The same goes for patients introducing a question to a physician. Zero points for being a jerk and/or presuming the doctor has read this discussion. It’s about partnering.
And if the offer of partnering on something important is declined, I’d start looking for a doc who’s a better fit for my style.
I have to disagree with Chukwuma (and a little with Bev as well). While there is no need to be confrontational, there also should be no need to worry about a MD’s ego. We are all adults, and our Mommies should have made sure that we can engage in a civil discussion at a certain level. For too long now we have accepted the sanctity of a MD’s knowledge, and while for some types of patients it is still a comfort, I should think that a doctor should be eager to engage with a patient who wants to know more.
Bev, Marya, Dave and Giles.
I think we are all advocating the same strategy with different tactics. The real question in my mind is what should be the response of an empowered patient to a physician who is (initially) dismissive of their desire to play a more participatory role or who wants to scrutinize accepted practice.
Yes. They could walk away. In some cases that would be the best thing to do; but we should all recall that participatory medicine is a learning process for both parties. So. In that vein. The patient has to (metaphorically) put on a white coat and “treat” the physician.
I would think that the first obligation is to determine if the physician is amenable to therapy or if he is an incurable case.
As I said, the initial reaction to the physician may be the “eye-rolling reflex”. Okay. Handle this case like a trained professional. Remember, this is *simply* a reflex. Perhaps this doctor has never been encoutered with someone who questions their recommendations. (You’d be surprised at how many patients who tell me that they want *me* to make decisions for them after they are presented with options).
First: Determine if the reflex is transient, alterable or fixed and ingrained. If the eye-rolling reflex is alterable then you can still save this case. If it is fixed then you would likely be better served with a new physician.
Although these statements are made in a seemingly flippant way; they belie a very important issue.
I have had cases where patients will come to me as a third, fourth or fifth opinion. In many cases; we have a cordial and effective working relationship because we communicate effectively. However, I will not say that this is all due to my personal skills. Indeed. In *most* cases, I am not doing anything radically different from the other consultants with which the patient has seen. The difference, of course, is that we are actually communicating and making progress rather than being stuck in the arbitrarliy and historically mandated role of “doctor” and “patient”. Part of the transformation in the relationship is that I am willing to engage patients. But (as many patients will tell me) part of the issue is that the patient has improved their ability to tell their story and to determine what the doctor needs to do their own job.
Dave, this is actually one of the things I’ve learned from your personal story. The fact that you were learning more about yourself and the health care system as the process unfolded toward your eventual treatment. I truly believe you improved your care and made your doctors better physicians.
This process does not have to be confrontational. And while there may be doctors who are beyond hope; I suspect that there are many more who simply need to be educated as to the reality of the new type of patient out there.
I’ve actually gotten some feedback via Twitter and Email from young doctors who have told me that this discussion that we are having here is opening their mind to aspects of patient care they were unaware of.
Here is one Tweet: http://bit.ly/aOzDaA
This process must continue. In the end it will benefit everyone involved. (Okay, perhaps not the people who don’t want to be helped….)
Let’s try that link again…
If you look at this issue rationally it is perfectly normal.
But unfortunately we live in a society with a majority of people who demand absolutes in every realm of their life. Since a very large number of Americans are proud religious fundamentalists how could they ever accept the greatest reality of all: we know very little about anything and whatever we know is subject to change at any time. As I always say; “Knowledge is in constant beta!”. The fantastic story is that although we know very little we already know enough to move robots to Mars. But obviously not enough to understand the genetic networks and their associated epigenetic effects responsible for the million variations of cancer.
Dave, I have to agree with my esteemed colleague Chukwuma above — the issue is not that all science is bad, the issue is how we interpret it. I have been blogging a lot lately about the imprecise nature of clinical research. For this reason touting “science-based” medicine as providing all the answers is disingenuous and paternalistic. The reality is that there will always be more questions than answers and the complexity will continue to demand integration of knowledge. And humbleness.
Thrilled to meet you, Dr. Z! In my talk last May at the Institute of Clinical Systems Improvement colloquium, one of my requests was “Give up the illusion of certainty.” See my other reply to Chuk tonight.
I wonder how much registration there would be for a breakout session at a conference titled “The Illusion of Certainty.” :–) Not an inspiring thought – but true.
I hope to chase your blog too – anything that catches Bev MD’s eye is good enough for me!
Thanks, Dave, really nice to meet kindred spirits.
For anyone on this thread who hasn’t yet chased BevMD’s link to Dr. Z’s blog – read this! “HTE” = heterogeneous treatment effect – “the effectiveness and safety of a treatment varies across the patient population.”
That’s a pretty alarming thought if you’ve come to believe that a well designed drug approval process will lead to certainty … and, more importantly, if you think nothing is valid if it HASN’T passed that standard.
May I recommend a book by Ben Goldacre, called “Bad Science?”
Ben is a NHS psychiatrist, founder of http://www.badscience.com, and is on Twitter @bengoldacre.
This book should be essential reading for everyone in or even closely allied to the medical profession.
It details all the bad science out there, permeating big pharma, alternative medicine, allopathic medicine, nutritionist practices – no scared cows here!
Brilliant, brilliant, brilliant!
Go on – buy it!!
And I have no personal gain whatsoever in saying so!!
Last thing is that junior doctors and residents are pushed to publish research – “publish or perish” – even if much of it is !@#$.
Martin, does “Bad Science” also cite examples from established medicine, or only from the alternatives you list? If the latter, I’d assert that it’s guilty of its own crime…
Dave – examples from all over medicine – it really is a remarkable work, and one that proactive medical ‘consumers’ like you would relish!
One example is the underhand measures some pharmaceutical companies use to try to prove their new drug is better than any other. Another relates to the incredible damage done to people by the ‘anti-vaccination/autism” lobby.
Goldacre is really very critical against those who use ‘bad science’ to further their own ambitions.
I agree completely with Dr.’s Onyeije and Zilberberg, particularly with regard to the comments about nuances getting lost in reporting, and how we interpret the science. I recall, but cannot locate, a similar comment by (I think) the founder of the term “evidence based medicine” – to the effect that collecting the evidence is the easy part; it’s interpreting it that introduces unavoidable subjectivity.\
And BTW, this led me to Dr. Zilberberg’s excellent blog, which I recommend!
Thanks, Bev, really appreciate your thoughts!
> As long as patients have to provide an excuse for being
> what they should have been all along, ie autonomous,
> we can’t expect a real reform of the system.
Bingo. Thanks for saying what was stuck in my craw, unidentified.
Archimedes was founded by Dr. David Eddy, who is sometimes called the “father of evidence-based medicine”. Kent Bottles has a helpful write-up of a lecture Dr. Eddy gave at ICSI last year, which will provide some insight into his thinking and research: http://bit.ly/7miYQy.
In their own words, Archimedes describes themselves as follows:
Archimedes is a Healthcare Modeling company that has built a large scale simulation model of physiology, disease, interventions, and healthcare systems.
The Archimedes Model is written at a high level of detail using object-oriented programming and run on a distributed computing network. The core of the Model is a set of ordinary and differential equations that represent the physiological pathways pertinent to diseases and their complications. The Model includes aspects of diseases and healthcare systems needed to analyze downstream clinical events, utilization, and costs including: signs and symptoms; patient behaviors in seeking care; patient encounters with the health care system (e.g. emergency room visits, office visits, and admissions); protocols and guidelines; tests and treatments; provider behaviors and performance; patient adherence to treatment recommendations; and clinical events that affect logistics, utilization, and financial costs. The Model is flexible and helps decision makers understand the likely outcomes of interventions.
To learn more about the Archimedes Model visit http://www.archimedesmodel.com/modeling
The company has been offering free webinars by Dr. Eddy that describe their model in great detail. The next webinar is scheduled for Oct 28. See: http://www.xtalks.com/xto421ArchimedesDemo.ashx
Note, Archimedes was funded by Kaiser Permanente Innovations and is still fully-owned by KP. Definitely an interesting company.
My response to The Atlantic article is here:
Indeed, the emperor has been without clothes for some time. These insights are necessarily new, but the mathematical prediction is astounding. That 41 percent of these often-cited studies turned out to be false — and yet are still cited even when later debunked — should be a wake up call for anyone who believes evidence-based medicine is some sort of cure-all for our current research/practice ills.
There are no easy answers here. The journals don’t want to play gatekeeper, the government doesn’t want to play gatekeeper, and the universities can’t be trusted to play gatekeeper. That leaves us with the researchers themselves — again relying on individual integrity and morals to keep the on the right path.
Sadly, as this article and oodles of research demonstrate, researcher integrity can be sadly malleable.
It’s great to know researchers like Ioannidis are out there doing this work.
that is exactly why, now that we have the tools at our disposal, engaged patients must take their right place in the health care system and move rapidly from the most underutilized resource of the system to a position of arbiter, with the help of patient networks, databases and maybe even AI in the not-so-distant future.
I hope an algorithm can be developed that does exactly that — it certainly seems within our reach.
There are just tons — and especially in psychological research, the majority — of ‘bad’ research whose results should be barely mentioned. Instead, the results are blasted all over the media, with few outlets bothering to take the time to put anything into context.
Making the job all the harder is that “context,” as Ioannidis’ research shows, is virtually impossible in today’s fractured universe of journals and the current publishing paradigm.
All of this must change if we are to move forward, faster, with our medical (and psychological) knowledge.
In a related thread and thanks to Cameron Neylon who asked
I found this pearl:
The entire thread is worth reading
Wow. There’s much here that’s over my head, yet also much to comment on, and it’s really difficult to figure out how to organize it, even after trying to sort it out on paper! But, I’ll try as I think e-patient involvement in such discussions are important.
[I will declare here that I have NOT followed the numerous links in the comments, and I am a self-educated e-patient, having learned only a bit more than what’s generally available because I accept nothing at first glance.]
As to Patients
Patients need to be (or become) informed about the scientific model currently in use, and its flaws. The ones I had already found before this article included:
• Clinical trials cannot always be taken as gospel because the size of the trial groups can be incredibly small. I have been flabbergasted to read the info (sometimes) included with an Rx or sample gotten from a doctor. I can’t recall which medication it was for, but I read one where the trial size was only 10 each in the group taking the medication and the control group. It seems downright irresponsible for a medication to be approved by the FDA and put on the market with such an infinitesimal cross-section of the population at large.
• Clinical trials cannot always be taken as gospel because they are often funded by pharmaceutical companies with a vested interest in the outcome. This is also true of “articles” on the web, such as those at WebMD. Experimentation can be (and I believe is) slanted from the beginning in the language of the the funding. Scope can be limited, the pharmaceutical company can “own” the results and have a non-disclosure clause prohibiting the scientist’s releasing the results. How many trials, I wonder, are collecting dust at pharmaceutical companies because the outcome was negative?
• Just because a medication has been tested in trials and approved by the FDA doesn’t mean it’s safe. I learned this lesson the hard way, though not as hard as some. My doctor suggested that I try the new Cox-2 inhibitorsfor my OA when they came out on the market. Neither Vioxx or Celebrex worked for me, but I was absolutely. thrilled with the results of Bextra. Right up until it was taken off the market due to a high percentage of cardiovascular events. It’s a general rule for me to now wait a year (Ok, maybe 6 months if it looks real promising and I’m desperate) after a new medication is on the market before I’ll try it.
Other things that need to be understood by patients include:
• Doctors are NOT gods. This is a myth that’s permeated all of society. We all need to be retrained. Medical schools need to say it out loud to their students (“Patient Relations 101”?) The medical news sections of the internet and TV, and in papers and magazines need to keep this in mind when reporting medical news and make their reporting geared towards and understandable by patients, with the caveat to ask their doctors (rather than marching in and demanding). Asking a question should not be taken as a challenge or insult, and if that’s how your doctor takes it, find a new one.
• It’s not that difficult to become an e-patient. Is it time consuming? Yes. Doing your own investigations to become educated about medical subjects takes time.
But, fortunately, there are are now plenty of sites that have learned to forego writing in “medicalese” and use English instead. (But investing in a good medical dictionary is something I recommend.) there are also support groups available for a lot more conditions than thetre usef to be. Join one, or if you can*kt find one, start one. Bouncing ideas off others is always beneficial.
• If you are already an e-patient, spread the word! Tell your family (teach your kids!), friends, co-workers and (if you have a blog) as an off-topic post to your readers. They’ll forgive you if you make it clear you’re expressing your experience and opinion on a subject you find of utmost importance. Be understanding if others don’t feel as you do. My experience has been that, the older and less ill the patient, the more resistance there will be.
That’s going to have to be it for now. I’ve run dowN the battery on my phone. And I apologize for any typos. ;)
The link for “understanding statistics” comes up with a nothing found… just thought you might want to fix it.
Yikes, sorry, Stitch – it’s fixed now. The correct link is http://pmedicine.org/epatients/archives/category/understanding-statistics. Thanks for letting us know.
I wrote an article, “Doubt is Healthy When Interpreting Medical Research Studies” for HealthTap that summarizes some of this discussion for a more general consumer audience — with my perspective, of course: http://bit.ly/gLts3J. Comments welcome!