Which Medical Practice Will Be Discredited Next?

An editorial in the current British Medical Journal makes a very sharp point that many of us have probably been thinking about in the last few weeks while reading the latest medical news in the papers:

It’s easy to feel contempt for deluded practitioners of the past who advocated bloodletting and tonsillectomies for all. Easy, that is, until one considers emerging evidence that coronary stenting and postmenopausal hormone replacement therapy may well be the contemporary equivalents of those now discredited practices.

This recent New York Times article tells of the compelling study which found that coronary stenting is typically no more effective than heart drugs, even though it is far more invasive — and, to be sure, profitable for the medical personnel involved.

Indeed, another article in the same issue of BMJ suggests that stenting is less common in Europe than in the U.S. mainly because “there have not been the same financial incentives to carry out some stenting.”

FWIW, we mentioned this subject briefly in Freakonomics:

If you were to assume that many experts use their information to your detriment, you’d be right. Experts depend on the fact that you don’t have the information they do. Or that you are so befuddled by the complexity of their operation that you wouldn’t know what to do with the information if you had it. Or that you are so in awe of their expertise that you wouldn’t dare challenge them. If your doctor suggests that you have angioplasty-even though some current research suggests that angioplasty does little to prevent heart attacks-you aren’t likely to think that the doctor is using his informational advantage to make a few thousand dollars for himself or his buddy. But as David Hillis, an interventional cardiologist at the University of Texas Southwestern Medical Center in Dallas, explained to the New York Times, a doctor may have the same economic incentives as a car salesman or a funeral director or a mutual-fund manager: “If you’re an invasive cardiologist and Joe Smith, the local internist, is sending you patients, and if you tell them they don’t need the procedure, pretty soon Joe Smith doesn’t send patients anymore.”

The BMJ editorial also asks a wise question. Perhaps some of you can offer an answer: “What other medical orthodoxies might join hormone replacement therapy and stenting on the ever-growing list of discredited interventions?”

I have one prediction: late-stage chemotherapy for many types of cancer.


I have one prediction: late-stage chemotherapy for many types of cancer.

This is typically done at the behest of the families and patients, not because doctors think there's any hope. Most docs would rather prescribe hospice care for late-stage cancer patients, but people like to cling to false hope, so unpleasant treatments continue up until the very end.

It'll stop when insurance companies decide to stop paying for it.


As more and more "targeted" therapies are developed, that is, therapies that attack specific proteins or pathways in tumor cells, you will see fewer cases of non-specific chemo used for late-stage.

As these drugs get better and our techniques for identifying which proteins and pathways are mutated in a person's indvidual tumor improve, physicians will be able to make better selections for chemo regimens.

This is happening in isolated places today and the clinical trials to determine its effectiveness are just getting started. Regardless of the short term, I don't think you'd find too many people in the field who would argue that this isn't the way it will be in 20 years.


I actually would pose a complementary question, what simple things are doctors not doing that could help patients significantly?

It seems to me that most doctors know very little about nutrition, diet, and exercise. If they do know something, they're certainly not sharing it. Why not? No incentives. Offering comprehensive diet and exercise plans to patients brings in no revenue and significantly decreases the chances that patients will need more care.

Interventions that I think may end up being discredited:

HPV vaccine for all young women (very expensive, unknown long-term risks, insignficant improvement of health outcomes)

Use of statins to lower cholesterol (the link between these drugs and reduced risk of heart attack sounds pretty shaky)


I think the trend in cardiology will increasingly be towards discrediting practices for specific groups (i.e., women and minorities).

I'm not sure cancer will follow the same path. Cancer research has been increasingly specialized since the 1970s, but most of the apparent improvement is still ultimately due to a handful of statistical artifacts. Because so many researchers have incentives to prevent this artifact from becoming public knowledge, I suspect that it will continue to fly under the media's radar.


Chemotherapy for advanced cancer has been tested multiple times. There are even have meta-analyses (where multiple studies are combined to assess the benefits of a given treatment). While some may disagree about the magnitude of benefit, most therapies given to advanced cancer patients (especially common ones like breast, lung, and colon cancer) have been tested in randomized, double-blind, placebo controlled trials, already. These treatments generally meet the most rigorous endpoints: they make people live longer and feel better.


My suggestion is Spinal Fusion Surgery for Low Back Pain. The logical rationale for this surgery is shaky, it causes dysfunction at the segments above and below the fusion, and has been shown in some studies to be only minimally better than a wait and see approach. Improvements in non-surgical treatment of back pain, ie. Manual Physical Therapy, and alternative surgical fixes such as total disc replacements will make spinal fusion seem archaic and a thing of the past.


How about C-sections and epitomizes? Though not unnecessary at times they are probably an example of medical exploitation of fear and ignorance.


Why stop at individual procedures?

The entire medical profession is a distorted market morass of ethically compromised actions and reactions. The reason is the lack of true market competition being replaced by insurance nonsense.

The good news is that it's all falling apart and we will soon have government health care. The bad news is that it's all falling apart and we will soon have guvment health care.


The whole debate about stenting amazes me. There are ways of managing heart disease without drugs OR surgery. We should have gotten a lot more data about this since Ornish published his protocol, but it seems to be flying under the radar. Drugs are more invasive than diet and lifestyle changes, but the incentives for doctors aren't there.

Nutrition and the role of intestinal flora might be next. When will doctors face sanctions for prescribing antibiotics for common colds?


I think the real scandal is how unscientific nutrition and principles of general health are. Probably half of all health problems in the US are related to poor diet, muscular atrophy, and lack of exercise. And yet we rely on fad diets, self-help books and infomercials to deal with the problem. For my money it's more ridiculous than blood-letting.


Hi, I live in Canada, so health care's free. But drop-in clinics, run by general practitioners (GP), suffer from gross failures concerning the proper diagnosis of people and often fail to give them the necessary medical attention. Since these practitioners receive wages in part based on the number of visits, they have an incentive to rush you through the visit and will usually prescribe some kind of antibiotic or some other quick-fix. For example, I didnt get a proper medical diagnosis until I went to a hospital, instead of a GP.


I'm more appalled by heath insurance practices than the medicine being practiced.


So Dubner tosses a softball to the anti-physician crowd, even to go as far as to try to discredit one of the most researched, evidence-based fields in all of medicine (oncology)? Nice.

For one thing, to lump "cancer" as one disease is pretty telling of the vein of this thread. Chemo protocols are highly specific, and rigorous clinical trials evaluate benefit to both quantity AND quality of life improvements. Sometimes the risk and benefits, well-quantified, are so close that individual patients have to make tough decisions.

To suggest that oncologists are throwing chemo at dying folks for no reason demonstrates a glaring inability to use MedLine or UpToDate. Cancer sucks, chemo sucks, and we dump a lot of resources into research into balancing the two.


Yes, using Chemo is basically poisoning the cancer and the patient's heathly cells. But it is one of our best weapon right now. I agree with Dubner in that one day grandchildren will look back and be horror at the fact that we poison our cancer patients.


Ok billypilgrim37, I'll give you your little fairy tale. And raise you reality.

Here's how it really goes down. Doctor tells cancer patient that his/her cancer has returned. But good news, there's a new treatment out for just this case. A real miracle drug that has shown great promise. Would you like to have it?

Patient says, "well, hell yes!". Patient gets new drug. Dies in 2 months.

Turns out the new drug extented the life of such patients by 1 to 3 months in clinical trials. More often than not the trials were made statistically significant by cherry picking the data.

I'm with Dubner here. Pretty much a scam. When your data shows complete remission for 5 years in even 5 out of 10, then you may have an argument for the drug. You may argue for 3 out of 10. Otherwise, get back to work.

Connie H.

What the medical/insurance hierarchy isn't doing properly: adequate support of diabetics, especially daily maintenance. Do you know many insurance plans only pay for one blood sugar test strip per day? Since the long-term damage from diabetes can be caused by widely fluctuating blood sugar levels, it would seem that the wisest and most cost-effective support would be at least three tests a day to adjust insulin as needed.

Don't even get me started about making syringes and needles prescription-only, while we're on the subject of stupid "medical" practices.


egretman, the absolute vast majority of chemotherapeutic trials right now are not for new drugs. They're for combinations of chemo agents and dosing regimens trying to optimize outcomes.

I'm not sure where you derive your supposed knowledge of oncologist behavior, but I'm fairly certain that your characterization of docs pushing supposed miracle drugs is absolutely irrelevant to reality. It certainly wasn't reality for my family member that died of colon cancer back in February.

Cardiologists have tremendous financial incentives to perform stenting procedures, but those same financial incentives simply don't exist in oncology on any similar scale.

Evidence-based palliative chemotherapy is a good thing when a patient stands to gain a balance of quantity and quality of life. And its existence doesn't affect an oncologist's bottomline significantly enough to bolster your anti-physician conspiracy theory.

"When your data shows complete remission for 5 years in even 5 out of 10, then you may have an argument for the drug."

Umm, late stage chemotherapy is typically for palliative purposes only. An extra 3-6 months of quality life would be a pretty noble goal. If that's the rigor you demand of cancer therapeutics, then we might as well throw most cancer patients out the window of a 20 story building and save you the trouble.



Although medicine (oncology or otherwise) may be one of the "most research, evidence-based fields," who is funding this research and what are the incentives for finding "evidence" that points in a certain direction.


Improved imaging techniques will soon make routine screening colonoscopies (talk about invasive) a thing of the past. Likewise, though we have a ways to go at present, vaccines against HPV will hopefully make PAP smears (and the necessary pelvic exams for symptom-free women) unnecessary. Diabetics will be happy to hear that trials are underway for inhaled insulin. Islet cell transplants will one day make type 1 diabetes a thing of the past.

However, the biggest leaps in medical therapeutics will come when we begin to harness stem cells for targeted tissue regeneration. For example, people suffering from heart failure may receive doses of autologous stem cell that have been pre-treated with surface proteins which targets heart muscle. The result: regenerated heart muscle.


--Imaging colonoscopies won't remove the need for biopsies of suspicious lesions, which are pretty dang common. They'll be a great advance and will probably save money and trouble, but they won't make the mechanical procedures rare by any stretch.

--Gardasil only targets the HPV subtypes responsible for about 70% of cervical cancers. That's a LOT of cervical cancer, but I'd be very surprised to see the pap go anywhere during our lifetime.

--Pfizer's Exubera (inhaled insulin) is already on the market. We don't have sufficient data for long-term pulmonary safety, and for some strange reason, Pfizer decided to issue the med in metric units rather than the more familiar IUs. Docs and patients are underwhelmed on the whole.

I only jumped on Dubner for making a really bad choice in his selection of a therapeutic intervention that will go away soon. I certainly don't give a blank check to drug and device companies.