In a recent interview, Tim Harford, the Underground Economist, said,
That’s what makes medicine such an effective academic discipline.
By “that” he meant certain methodologies, especially randomized experiments. I disagree with this assessment. My opinion is that health care is in a state of profound stagnation, unable to make much progress on major problems.
Here are six signs of the stagnation in health care (by which I mean everything related to health):
1. The irrelevance of Nobel Prizes. Year after year, the Nobel Prize in medicine is usually given for research that is so far useless (e.g., teleomere research) or irrelevant to major health problems.
2. The obesity epidemic. Starting in 1980, obesity rates climbed fast. Thirty years later, doctors seem to know no more about how to cure obesity than in 1980. Low-fat diets, popular in the 1980s, are still popular! Low-carb diets are ancient — the Banting diet became popular in the 1860s.
3. Ancient treatments for depression still popular. SSRIs were introduced in 1988. Cognitive-behavioral therapy began in the 1980s, combining earlier ideas. Neither works terribly well — and notice how different they are.
4. The high cost of ineffective care. Americans pay much more for health care than people in other rich countries, yet American health is no better. All that new technology that Americans are paying for isn’t helping. In an article complaining about our education system, Joel Klein, the former head of New York City schools, wrote, “unlike in health care . . . in education, despite massive increases in expenditure, we don’t see improved results.” Actually, that’s exactly what we see in health care when we compare America to other countries. Tyler Cowan makes this point in The Great Stagnation.
5. Statins. A defender of modern medicine would claim that statins were an important innovation. They are heavily prescribed, yes. Yet in recent tests they have been stunningly ineffective — so much so that the earlier favorable evidence has been questioned.
6. The stagnation has become invisible — the normal state of affairs. Allowing Harford to make that comment. Harford, like Dr. Ben “Bad Science” Goldacre (whom Harford praises), believes you judge science by whether it follows certain rules. By making various rules (e.g., the need for placebo controls) and then following them, medical researchers have drawn attention — at least Harford’s and Goldacre’s — away from lack of progress. They’re making progress, they say, because they’re following self-imposed rules. Well, what if the rules make things worse? (For example, placing high value on placebo controls may draw attention away from non-pill treatments.) Better to judge by results.
What do you think are the clearest signs of health-care stagnation — if you agree with me about this?
How about the idea that the obesity epidemic is a problem, considering that “overweight” people live a little longer than “healthy weight” people? Chart.. Article about chart.
I’m not convinced that medicine is stagnating in all areas– people do seem more apt to recover from cancer than they used to be.
That 75% of all medical expenses go to maintaining someone’s life for an extra six months….
First, many health care revolves around ideas that are questionable in the first place (like statins), and many doctors dismiss nutrition as being silly. Americans in general have terrible diets, and it’s a long way from improving. It seems like preventive medicine just means going to the doctor more often, so he can diagnose you.
I think the technology point you made ties in well with nutrition/prevention. Everything is developed for diagnosis and treatment, but treatment is often ineffective because of unhealthy lifestyles. Of course, nobody considers that–we just keep pushing “better” “treatments.”
A clear sign of health-care stagnation: it’s not helping us live longer.
From David H. Freedman’s recent book “Wrong”:
On distinguished professor put it to me this way in an e-mail note: “Our life expectancy has almost doubled in the past seventy-five years, and that’s because of experts.” Actually, the vast majority of that gain came earlier in the twentieth century from a very few sharp improvements, and especially from the antismoking movement. As for all of the drugs, diagnostic tools, surgical techniques, medical devices, list of foods to eat and avoid, and impressive breakthrough procedures and technologies that fill medical journals and trickle down into media reports, consider this: between 1978 and 2001, according to one highly regarded study, U.S. life spans increased fewer than three years on average- when the drop in smoking rates slowed around 1990, so did life expectancy gains.
The problem is with imposing too many rules and procedures, while researchers have no reeal understanding of why these are needed.
Designing a formal study nowadays, involves you in a myriad of useless rules, most of which are perfectly unrelated to getting valid results.
Randomising groups + double blind, is not the main barrier to having results. Most of the effort get spent on endless more details that naive students are told in school that has nothing to do with getting scientifically reliable results (think approval committies).
It is absurd. Theyball say you need many studies to prove something, which is empirically true, but laughable logically. If you do one really good experiment, you got to get vaid results.
The reason so many experiments get refuted is because people do experiments in invalid ways, or have no idea how to handle data. They only knnow that you need much work and endless procedures.
I wrote elsewhere on the problem of focusing on the measurable and on what experiments are easily done,
The academic bias. https://yzilber.blogspot.com/2008/08/academic-bias.html
Thanks, Nancy. It seems that a fatso like me is somewhat safer than someone in the lightest subset of the “normal” range.
By the by: my cardiologist told me recently that the results of my angiogram were “normal”. I asked him whether he meant “normal” as in commonplace, or “normal” as in desirable. He goggled at me; I inferred that not only had no-one ever asked him before, but that he’d never himself reflected on it before. Doctors, eh?
“How about the idea that the obesity epidemic is a problem, considering that ‘overweight’ people live a little longer than ‘healthy weight’ people? Chart.. Article about chart.”
I wonder how and when they calculate those weights? At death? After 12 months of radiation/chemotherapy, for example?
Until we know more about things like that, the chart really isn’t meaningful. At least not to me.
Joe
Seth, perhaps a failure to understand science lies at the heart of our healthcare follies. Here is a great video by Tom Naughton. Funny too.
https://www.youtube.com/watch?v=y1RXvBveht0&feature=player_embedded#at=221
I disagree with you on the stagnation of mental health treatments. In the past five years, medication advances for mental illness have been tremendous. There is a much greater understanding of the nuances of mental illness and many, many more medication options available. As someone who has had a lifelong struggle with dysthymia and major depressive disorder, I can attest to that personally. After trying treatments and giving up so many times earlier in my life, I was amazed and relieved by just how much the field has changed.
“SSRIs were introduced in 1988. Cognitive-behavioral therapy began in the 1980s, combining earlier ideas. Neither works terribly well — and notice how different they are.”
But combined they’ve been shown to be quite effective. I believe, very strongly, that the main problem in the treatment of mental illness in America is the disconnect between therapy/close observation and medicating. (I can’t speak for the rest of the world as I’m not familiar with other mental healthcare systems.)
The standard in the USA has been for a psychiatrist to see a patient once a month for 15 minutes. Unless they’re getting reports from the patient’s therapist, there’s no way the psychiatrist would be able to accurately assess for themselves how well the patient is doing. The doctor would be relying 100% on the patient’s self-assessment and their observation of the patient.
That day’s mood or recent circumstances can easily taint a patient’s perception so that the report they give their physician would be inaccurate (and even more likely, incomplete). Even the doctor’s observations could easily be misinformed. For instance, the patient might have just had a hard time finding a parking spot after running later than they’d have liked to the appointment. That’s not a major enough life event that they would necessarily mention it, but it would affect their body language and behavior.
Even more than that, the standard protocol has been to wait 2-8 weeks between making medication adjustments (changing doses or prescriptions), which, to someone suffering from severe mental illness, is an eternity-especially when they’re fully aware that if that next switch doesn’t work, they’ll have to wait several more weeks to try something else that’s new. It’s no wonder so many patients give up on meds, especially when those adjustments are often made on faulty information.
A health care provider meeting with a patient on a weekly basis or so and providing both therapy and medication fixes all those problems. But it’s a rare patient who even knows that option even exists. Psychiatry is the black sheep of the medical field, and it doesn’t help that some of the medical students who end up in the field are there because they weren’t good enough to get into their first-choice specialty and they see it as a cushy job. It has the lowest malpractice rate of any medical specialty, dermatologist-like hours (without the uber-high competition to get into the field), and so on.
I’m actually starting grad school next January to become a psychiatric mental health nurse practitioner so that I’ll have the prescribing rights but will still be able to do therapy.
/soap-box
Autism.
OK, I’ll bite on Point 1. Barry Marshall recently got a Nobel for demolishing entire industries (with a bit of personal science) and I note the ‘so far useless’ qualifier in the example of telomeres. The ‘health’ of the ends of chromosomes and ‘health/redox state’ of cells are correlated and life can be debilitated and short in diseases where telomere repair is compromised. E.g., ataxia telangiectasia, which is kind of a disease of premature aging, in part is caused by loss of DNA repair and subsequent chromosomal instability. It is in fact a genetic, genomic/cancer, neurodegenerative and metabolic doozy that includes immune, insulin signaling and cellular respiration irregularities etc. It may turn out that telomeres are the useless bit of research into genomic and metabolic damage repair mechanisms but that will only be obvious in hindsight.
For a not uncontested metabolic and nutritional slant on cancer management – including the possible link between telomerases, glucose metabolism and cellular respiration – see Seyfried 2010 ‘Cancer as a metabolic disease’.
I wanted to point out that you appeared on the Discover Magazine blogs today.
https://blogs.discovermagazine.com/discoblog/2011/05/19/ncbi-rofl-holy-correlation-batman/?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+DiscoverDiscoblog+%28Discoblog%29
Barry Marshall, “personal science”? No he did his self-experimentation as part of his job, not to help himself. It isn’t much of a counterexample because ulcers are not a major health problem and because the bacterium that supposedly causes ulcers lives harmlessly in a billion people. The big and important question that Marshall failed to answer is what distinguishes infected people who get ulcers (1%?) from infected people who don’t get ulcers (99%?).
It isn’t progress to come up with 20 more medications. Number of treatments doesn’t matter, what matters is their effectiveness. It’s progress to come up with something that works much better than existing treatments.
The weights are often measured prospectively. For example, in 1970, and the death rate after 1970 is measured.
yes, that’s a good example: little improvement in life expectancy.
thanks, hadn’t seen it.
The fact that cancers are still treated with one drug at a time, rather than with a cocktail as the AIDS activists helped make standard for AIDS sufferers. In both cases the disease is a moving, evolving target, and the best outcomes come from hitting it on all fronts. (Google results about “treatment-naive” patients; they do better on new drug treatments than ones who have already had other, _unrelated_, drug treatments.)
Of course, giving someone several drugs is bad science. (And more expensive, and with more chance of weird side-effects, yadda yadda yadda.)
I wonder how much stagnation in medical care is the result of lack of knowledge, and how much is the result of inadequate practice.
For example, I’ve got a lot of anecdotal evidence that it’s hard to find a doctor who will listen and think. (The estimated percentage is pretty low– would people care to post their own estimates? I’m concerned that I’ve been anchoring the responses I’ve been getting because I’ve been giving a number.)
There’ve been the work showing that doctors aren’t reliably washing their hands, and that checklists make a huge difference to hospital outcomes.