Some medical care may not help. Alternatives may not either.

‘Why Doctors Still Offer Treatments That May Not Help’, by Austin Frakt in the New York Times, August 27, 2019, is a well-done article that, despite its title, is fairly optimistic. He reports that while about 50% of current medical treatments no not have good evidence supporting them, only 3% have evidence showing them to be harmful, and another 6% unlikely to be helpful. This means about 41% (or only 41%, if you prefer) have good evidence that they are helpful.

Frakt cites some of the research on health interventions that do not work, or may even harm, quoting leading researchers Vinay Prasad and Adam Cifu (although he does not mention John Ionannidis, arguably the father of the field). He discusses the important “Choosing Wisely” campaign, initiated by the American Board of Internal Medicine Foundation, that asks each medical specialty society to identify at least 5 things often done in their specialty which usually should not be done. There is not, however, always consensus. For example, among its 20 recommendations (a high number that continues to grow, perhaps because of its discipline’s broad scope), the American Academy of Family Physicians (AAFP) recommends against imaging (including MRI and CT) for uncomplicated back pain, as does the American College of Physicians (Internal Medicine) (see ‘Medicine should not be primarily a business: choosing appropriate care for all, not excess testing for some, August 19, 2018).  Yet, neither the American Academy of Orthopaedic Surgeons, or the various groups of radiologists who perform these procedures, make this recommendation. Of course, both stand to benefit financially from doing these procedures.

It is concerning to think that, at least in some cases, financial benefit may influence the selection of some medical societies’ “Choosing Wisely” recommendations. Sadly, it is not a surprise, and financial benefit is, I believe, the reason for the use of many unproven treatments. Another important reason is an understandable reluctance for providers to abandon therapies that they have used for many years and believe to be successful. After all, if 50% of treatments do not have good evidence that they help or harm, many probably often do help, even if most of the evidence is anecdotal. Much more malicious, however, is the intensive marketing of new drugs and therapies to physicians by drug and device makers with a strong financial profit motive.

Frakt provides a litany of therapies-since-found-not-to-work-or-even-do-harm, including hormone replacement for post-menopausal women. The only drug he specifically names is Vioxx®, the “miracle” anti-inflammatory pulled from the market when it was found to cause heart disease (at least publicly found; the manufacturer actually knew it before it was released!). He also mentions “tight blood sugar control in critically ill patients” (and actually, probably most patients). This certainly relates to drug. The plethora of new, expensive, and marginally advantageous diabetes drugs is a testimony to their profitability. Ironically, it is often the same physicians who are unwilling to give up on treatments that are unproven or even proven to be of no benefit or of harm who are the first to begin using new (and, redundantly) more expensive drugs heavily promoted by the manufacturers.

Of course, it is not just diabetes drugs that are expensive. Compared to some of the newer drugs for rare diseases they seem like a bargain. In “The $6 million drug claim”, Times writers Katie Thomas and Reed Abelson discuss a woman who has $1 million in drugs to treat her condition in her refrigerator! While in her case they were paid for by her husband’s union, the cost can still be backbreaking: “At one point in 2018, for every hour that one of the union’s 16,000 members worked, 35 cents of his or her pay went to Alexion to cover the Pattersons’ prescriptions.” And what about those who have poor insurance, or no insurance? Or a union that goes bankrupt paying for them? Alexion, of course, does well, though.

A new law allows “gravely ill” patients the “right to try” drugs that the FDA has not approved. It sounds good; as a friend told me “I figure if I’m going out and they want to try something that might kill me I haven’t got much to lose. At best I’m cured. At worse, I go out but knowledge is gained.” Except for maybe dying sooner or more painfully, or possibly going bankrupt. It is not uncommon for doctors to suggest something new, unproven, or even a treatment that has already failed, one more time. Not everyone in this country has good insurance, and if your insurance company appropriately refuses to pay because there is no evidence of efficacy (yes, they are sometimes right!), you are on the hook for the bill. You may be dying, but your family may need the money you were saving for retirement. Thus, there may be a great deal to lose.

A certain percentage of people will take all this to mean that mainstream medicine is not to be trusted. This is definitely an overreaction, although a healthy skepticism is warranted, especially when the treatment is incredibly expensive and it is obvious that somebody (or some corporation) is making a killing on it. Some people will decide to pursue only what used to be called “alternative” medicine. These practices, often with hundreds or thousands of years of experience, are often called “complementary” and are now frequently offered along with mainstream medical therapies in “integrative” practices. But the same cautions must apply; because something is different, or traditional, does not make it necessarily either effective or safe.

For example, while people sometimes believe it is safer to choose only “natural” medicine, because a therapy grows in nature does not mean it is necessarily safe. Earlier in my career, medicine virtually abandoned the use of digitalis, made from the natural plant foxglove, to treat heart disease, after it had been used for decades. The key point is that if a substance works biologically and has “good” (i.e., desired) effects it can also have “bad” (i.e., undesired, or “side”, effects). This is true if it is straight from the plant, modified and standardized, or completely created in the laboratory.

Another real danger is what Frakt calls “wishful thinking”. Being optimistic and hoping that things will turn out well is good, and often useful when confronting serious illness. But when this transitions into the realm of “magical thinking”, being convinced that something good will happen to you because you want it to (or vice versa, that something bad will not happen because you don’t want it to), or believing a treatment will work because it is “natural”, or because your doctor recommended it in the absence of evidence, it is a real risk. “Magical thinking” is normal in three year olds, but dangerous in adults.

So what to do? Be open to new treatments, but do not reject the old, whether “traditional” or medical. Ask for evidence for treatment, such as when just changing a drug from an inexpensive standard (say, metformin for diabetes) to a new, costly one. Learn to understand probability, and ask for numbers. Do not reject anything out of hand, but do not believe that something will work just because you want it to.

Stay skeptical but not intransigent. Look for the evidence. And look also, when something is expensive, for the profit motive. Cui bono? It may not always be you.

"Eggs Benedict" and "Choosing Wisely": often the best thing to do is nothing

Years ago (OK, decades ago), when Woody Allen was a standup comic, he did a short, famous bit called “Eggs Benedict”. The narrator (let’s say “Woody”) is having pain in the chestal area (a great formulation which has long been popular in my house). He is pretty sure it is heartburn, but is worried and doesn’t want to pay the $25 to see a doctor (I told you it was a long time ago). Luckily his friend, Eggs Benedict, is having the same kind of pain. He figures if he can get Eggs to go to the doctor, he can find out what it is and save money. It works, and Eggs finds out it is heartburn. Two days later, he discovers Eggs is dead. He immediately checks into the hospital, has all kinds of tests, and discovered he has – heartburn. The bill is $110 (it was a really long time ago!) He goes to see Eggs’ mother and asks if his friend suffered much. “No,” she says, “the car hit him and that was it!”

I was thinking about this story when I was asked to call into a local radio talk show to comment on “Choosing Wisely”, an initiative of the American Board of Internal Medicine (ABIM), which calls on medical specialty societies to list 5 tests or treatments frequently performed in their specialty that they recommend not be done, or not be done on most patients. In the recently released report, 9 medical specialty societies are represented, for a total of 45 recommendations, which can be found here. The 5 for my own specialty, Family Medicine, listed by the American Academy of Family Physicians’ (AAFP), were:

1.  Don’t do imaging for low back pain within the first six weeks, unless red flags are present.

2. Don’t routinely prescribe antibiotics for acute mild-to-moderate sinusitis unless symptoms last for seven or more days, or symptoms worsen after initial clinical improvement.

3.  Don’t use dual-energy x-ray absorptiometry (DEXA) screening for osteoporosis in women younger than 65 or men younger than 70 with no risk factors.

4.  Don’t order annual electrocardiograms (EKGs) or any other cardiac screening for low-risk patients without symptoms.

5. Don’t perform Pap smears on women younger than 21 or who have had a hysterectomy for non-cancer disease.

Each is backed by the evidence, and is accompanied by a summary of the reasons. For example, for #2: “Most sinusitis in the ambulatory setting is due to a viral infection that will resolve on its own. Despite consistent recommendations to the contrary, antibiotics are prescribed in more than 80 % of outpatient visits for acute sinusitis. Sinusitis accounts for 16 million office visits and $5.8 billion in annual health care costs.” Big money for a treatment that doesn’t work, and can cause bad side effects (allergies to the antibiotics and increased resistance of bacteria to antibiotics, for two).

The talk radio hosts had questions, especially about treatment of cancer. One of the American Society of Clinical Oncologists’ (ASCO) recommendations is “Don’t use cancer-directed therapy for solid tumor patients with the following characteristics: low performance status (3 or 4), no benefit from prior evidence-based interventions, not eligible for a clinical trial, and no strong evidence supporting the clinical value of further anticancer treatment.”  That seems like a no-brainer to me, but they wanted to know “What if a patient wants treatment for cancer anyway?” I said I thought it was the responsibility of the doctor to point out that the treatment would not help, and would not only cost money but would have a lot of toxic side effects; I said that I thought most people, if they knew they were going to die from their cancer and the treatment would not help, would not wish to spend their last days and weeks nauseated, losing their hair, and being unable to interact comfortably with their loved ones.

The radio hosts also said “We are always told patients should be self-advocates. What if advocating for ourselves we say we want antibiotics for our sinusitis?” I said that self-advocacy was really important, because no one but the patient and their family could really identify what their health goals were. But this was different from deciding what treatment you wanted; going to the doctor is not going to the grocery store with your shopping list that may include both healthful and bad-for-you foods. It is critical that you identify where you want to go and work with the doctor, or other provider, to come up with a plan that is mostly likely to get you there safely and effectively.

Coincidentally, the same week I facilitated the discussion about futile treatment at the end of life for a group of third-year medical students in their medical ethics course. They had just received a lecture from a distinguished medical oncologist, who presented the four core principles of medical ethics: Autonomy, Beneficence (do good), Non-maleficence (do no harm), and Justice. They also read several articles documenting cases in which end of life decision-making did not go smoothly or well, and the article “How Doctors Die” by Ken Murray, that I have previously discussed (How people die, and how should we?, April 7, 2012). One of the cases involved a patient with terminal cancer who had already failed treatment, and for whom further treatment would be futile (although not without adverse effects). He had agreed to hospice, and to comfort care, until a relative (a physician!) came to town and demanded additional treatment for him, and convinced him to as well.

The medical students, who had  all been through clinical experiences and most of whom had been part of teams that confronted dealing effectively with dying patients, were in agreement with the physician in the article, and with each other, that the treatment was futile and would be a bad idea. However, several felt that, if the patient demanded it, the principle of Autonomy required that it be given. Others noted that this might violate the principle of Non-maleficence. I pointed out that that there were many areas in which we do not allow the patient to pick his/her treatment of choice. For example, we do not allow people to walk into a pharmacy and buy narcotic pain relievers without a prescription, no matter how much they might want them. Certainly the effects of chemotherapy poisons were at least as great.

These are not always easy decisions, but it is one of the reasons being a doctor is hard, respected, and well-paid. If everything were a simple algorithm and one could just memorize the right answer as these medical students were expected to do so often on their multiple-choice tests, it could be done by someone with much less training. Sure, it is possible that someone getting antibiotics for viral sinusitis to get better – indeed they usually do, with our without the antibiotics. Temporal association is not cause. Thus, even when you do the right thing, based on the evidence, and do not get a test or treatment, it is still possible that you will end up worse.

As Eggs and Woody discovered.