Not Treating – A Neglected Option

One of the criticisms of modern medicine is that doctors prescribe too many pills. That’s true. Patients and doctors sometimes get caught up in a mutual misunderstanding. The patient assumes that he needs a prescription, and the doctor assumes that the patient wants a prescription. But sometimes patients don’t either need or want a prescription.

I’ll use myself as an illustration. I get occasional episodes of funny, blurry spots in my visual field that gradually expand to a sparkling zigzag pattern and go away after 20 minutes. They are typical scintillating scotomas, the aura that precedes some migraines. I am lucky because I never get the headache. My doctor said we could try to prevent my symptoms with the same medications we use to prevent migraine, but there was no need to treat them from a medical standpoint. Nothing bad would happen if we didn’t treat. I told her I didn’t want them treated. They are a minor annoyance; I can carry on with my normal activities, even reading, throughout the episodes, and I have no desire to take pills with potential side effects and with the cost and the hassle of remembering when to take them.

If it had been a typical patient and a typical doctor, the sequence of events might have been very different. The patient might have been more frightened by the strange phenomenon than I was. (I thought the weird tricks my brain could play on me were fascinating and fun to watch, not scary.) The patient might have desperately wanted those threatening symptoms to go away without understanding how insignificant and non-threatening they really were. The doctor might have assumed the patient wanted them to go away. The pills might have been offered and accepted with little thought.

A woman goes to her doctor complaining of nausea and vomiting. She wants to get something to stop it. She finds out she is pregnant, and it’s only morning sickness. Her whole attitude changes; now she is happy to put up with the symptoms. She wouldn’t want to take anything that might harm her baby.

When a patient goes to a doctor, he doesn’t necessarily want treatment. He really wants answers to these questions:

What is causing my symptoms?
Is it serious?
How long is it likely to last?
How can it be treated?

He may ask about the side effects of treatment.
He is less likely to ask one other crucial question: what would happen if we didn’t treat at all?

Here’s a scenario that plays out in doctor’s offices thousands of times every day: a patient has cold symptoms. Maybe he’s afraid it might be something more serious than a cold, like a sinus infection or pneumonia. Maybe he thinks the doctor can cure his cold with antibiotics or some other new remedy. The doctor examines him and reassures him that it’s only a cold. He knows there is no treatment that will make the cold go away any faster, but since the patient has gone to all the trouble to come in for an appointment, the doctor feels a little embarrassed about telling him it’s trivial and sending him away with nothing. He gives him a decongestant and some Tylenol. He knows they won’t do a thing to affect the course of the cold but hopes they may help relieve some of the symptoms. He has the feeling that in this social interaction he is expected to give the patient a token, in this case a prescription. The patient has the same feeling: that he ought to get something to take home. Both the doctor and the patient think the doctor ought to “do” something. (These are not the bad doctors; the bad ones prescribe antibiotics or homeopathy or something equally nonsensical.)

What if every doctor explained that there was no effective treatment, that the symptoms would subside in a week with or without pills, and that symptomatic treatment wouldn’t help much and carried a small risk of side effects? The patient would then have the chance to say, “I want something to take even if it’s no better than placebo” or “I don’t really need to take anything. I just wanted to be sure it wasn’t something serious.”

When a patient has a high LDL cholesterol and other risk factors for heart disease, especially if he has already had one heart attack, some doctors assume they should prescribe statins, and patients assume they should take them. The reality is that most patients who take statins will probably not benefit from them. Instead of a knee-jerk grab for the prescription pad, a thoughtful doctor could look up the NNT (number needed to treat) and NNH (number needed to harm) and tell the patient something like “For your particular risk group, 84 patients need to take statins to prevent one death from heart attack, and of those 84 patients, 3 will have serious but not life-threatening side effects.” (I’m just making up these numbers.) Some patients will think “I might be that one patient whose life is saved” and will want the pills. Others will opt not to treat, thinking they are more likely to be one of the ones with side effects and/or no benefit.

Neither course is unreasonable; the choice depends on things like the patient’s philosophy, his degree of optimism, his fear of death, and his willingness to take risks. One way to look at it is that statistically a group of people on statins will live longer than a group not on statins, so the doctor should offer them to the group. Another way to look at it is that if 84 patients chose not to take the pills, one would have guessed wrong and would have a heart attack and the other 83 would have guessed right. (Of course, the situation is really a bit more complicated, because statins may have good or bad effects on other parts of the body besides the heart, and there are other things to consider like cost and interactions with other medications the patient might be taking and the patient’s ability or willingness to work on reducing other risk factors.)

In my article on “Reading the Medical Literature with a Critical Eye” I discussed an article about t-PA treatment for stroke. “The bottom line is that one in eight patients is helped at three months, one in 17 is harmed, and although the randomized trial showed no increase in mortality, there has been a documented increase in death rates in patients who have received t-PA therapy outside of research trials.” There is room for disagreement, but many people would agree that not treating is the better option.

Early stage prostate cancer is another real dilemma. It may be slow-growing, and the patient may die of other causes before the prostate cancer can kill him. Should you try to eradicate the cancer with surgery or radiation, with a significant chance of complications like impotence? Or is conservative management reasonable (surveillance, watching for a rise in his PSA levels)?

Are your knee joints kaput? Joint replacement surgery is an option. Medical treatment with drugs like NSAIDS or narcotics is an option. Non-treatment is an option too – maybe you would rather give up sports, avoid stairs, and accommodate your lifestyle to your limitations.

Another personal example: when I had a breast biopsy, I chose local anesthesia because I thought it was safer than general anesthesia. The anesthesiologist agreed, but said he would use IV sedation along with it. I asked him why. He said to control my anxiety. I told him I wasn’t anxious and didn’t want it. We negotiated, and finally he agreed not to give me a sedative before the surgery and I agreed that he could give it at any time during the procedure if he thought I seemed anxious or my BP rose too high to suit him. It was probably the first time in his career that he had omitted routine IV sedation for a procedure like this. All went well; I wasn’t anxious, we chatted throughout the surgery, and my time in the recovery room was minimized.

IV sedation is a good idea for the average patient, but not every patient needs or wants it, and it’s nice to have the option of no treatment. My anesthesiologist would never have even considered that possibility if I hadn’t spoken up.

Of course, non-treatment is not always a reasonable option. If a patient has severe diabetes and doesn’t get insulin, he will die. I’m not talking about those cases. But even in those cases, the principle of autonomy gives the patient a choice. Every surgeon has had an occasional patient who refuses life-saving surgery with the full understanding that it means he will die. If the patient is mentally competent, he has the legal right to refuse any treatment.

Think of the many self-limiting minor illnesses where non-treatment IS a reasonable option. Things like colds, common backaches, headaches, earaches, sore throats, gastroenteritis… many of the most common reasons for doctor visits. If the doctor rules out a serious condition, there is no need to take pills for symptoms that are expected to subside on their own. The patient may prefer to try simple comfort measures at home. I’m rather fond of the old remedy for a cold that says to go to bed, put your hat on the bedpost, and drink whiskey until you can’t see the hat. That’s a joke; but seriously, “tincture of time” is a treatment with no side effects: it’s the safest and cheapest remedy of all.

I have had lots of patients tell me “It doesn’t really bother me that much. Now that I know it’s nothing serious I’m happy to put up with it. I don’t need to take anything.”

So this is a plea to doctors and patients alike: when you’re considering treatment options, keep non-treatment on that list of things to consider. Let’s destroy the myth that the doctor always has to “do” something.

Posted in: Science and Medicine

Leave a Comment (24) ↓

24 thoughts on “Not Treating – A Neglected Option

  1. weing says:

    Great post. Non-treatment is always an option and the patient has the right to decide. I also tell my patients to let their families in on their decisions. This avoids problems in the future.

  2. overshoot says:

    In my own very limited experience, anaesthesiologists are the hardest to negotiate with. Bummer — I have a (literally) dangerously high pain tolerance but don’t tolerate anaesthesia at all well. If I’d had my druthers, my appendectomy would have been under local, just like the c-sections that delivered all of my kids.

    Knees? Run in the family [1]. My orthopod (bless him) has informed me that one will require replacement — someday. In the meantime, exercise is working as well for me as for my brothers and at least two of theirs are worse.

    Your point about the “informed healthcare consumer” is so well worth repeating that I would suggest it should be part of the regular high-school curriculum; it would fit in so well with other existing health topics that I don’t think it would actually displace anything.

    [1] Sorry ’bout that.

  3. pec says:

    Thank you Harriet for a heck of a rational post. I absolutely agree.

  4. Stu says:

    But Harriet, what about Saint John’s Wort?

  5. overshoot says:

    But Harriet, what about Saint John’s Wort?

    There’s research showing that duct tape works.

  6. DLC says:

    Good article, thanks.
    I was just wondering though: Let’s say Joe goes to the doctor with a headache, runny nose, nasal congestion and body aches.
    Doctor tells him it’s the flu (and for purposes of discussion it is).
    Joe, not satisfied with being told to go home, rest and get plenty of fluids goes to the local sCAM clinic/store/hole-in-the-wall and buys a couple boxes of homeopathic sugar pills. Joe takes the pills and in 3-4 days is feeling much better, and so ascribes his ‘cure’ to homeopathy. Would spending time telling Joe about the flu, and about how there’s no cure for it but time be advisable ?
    (this assumes Joe is otherwise healthy and not in a high-risk group)

    Personally, I also get those visual symptoms, but unfortunately they domean a migraine is coming.
    Oddly, I have never asked my doctor for any special drugs for this, as I only get one about 1-2 times per year.

    I looked up migraines on google, found the mayo clinic website section on it, and they list sCAM treatments available
    Migraine: Complementary and Alternative medicine
    The Mayo Clinic just lost 1000 respect points with me.

  7. mckenzievmd says:

    Excellent essay! After a few years of practice, when I stopped being so afraid of clients, I started telling them when I thought the best choice was to do nothing, and I was amazed at how many of them really liked that option. Sure, if things to progress then some will be mad you didn’t do something right away, but most seem happy with your honest opinion. I was surprised, though, at how often the nurses seemed suspicious of my not offering treatments they had seen other docs offer, and it takes a while for them to get to know me before they stop wondering if maybe I’m just lazy. :-)

  8. Michelle B says:

    Pec wrote: Thank you Harriet for a heck of a rational post.

    Yes, it is so rare that Harriet writes rationally! *Sarcasm off*

  9. daedalus2u says:

    Dr. Hall, is there a way to send you a private message?

  10. wertys says:

    In my field (pain medicine) I routinely offer the option of no treatment, as by the time people come to see me they have often reached a working arrangement with their chronic symptoms which demands a good reason to change. For many chronic stable problems, reversion to the natural history is a safe but sometimes difficult to choose option. Its a good topic to bring up..

  11. apteryx says:

    DLC: There is a difference between “listing” *CAM* treatments and “recommending” them. Since your link did not work, I Googled this page. It is a consumer education page that simply mentions commonly used treatments and the state of science surrounding them. For example, their sole comment on chiropractic is: “There are no scientifically valid studies that prove that chiropractice or other spinal-manipulation treatments are effective in the treatment of migraine.” However, they acknowledge that biofeedback can reduce migraine pain, and that there is “some scientific evidence” that feverfew may prevent migraines. Those statements are true, DLC. If you do not want to use those modalities, that’s your prerogative; but please don’t condemn Mayo for giving people straight facts about things they have already probably heard of elsewhere.

  12. Grapmag says:

    The worried well are responsible for over 1/3 of office visits to primary care in my unscientific estimation. My colleagues think it’s even higher.

    Imagine how many pounds of hydrochlorothiazide we could provide to impoverished hypertensives with the resources wasted on these visits.

    But that would be logical. Applying logic to the American health care system will give you scintillating scotomatas followed by a pounding unilateral headache. Don’t go to the ER though – its to busy being the US national health care system to help you.

  13. overshoot says:

    The worried well are responsible for over 1/3 of office visits to primary care in my unscientific estimation. My colleagues think it’s even higher.

    Imagine how many pounds of hydrochlorothiazide we could provide to impoverished hypertensives with the resources wasted on these visits.

    You can’t have it both ways. If patients only come to be seen when they have spurting arterial wounds but “don’t want to bother you over something minor” you won’t see the man with a case of “indigestion” or his wife who is “acting funny.” (Says //me, whose father died of an MI that he didn’t think was serious enough to interrupt a fishing trip.)

    Barring perfect diagnostic skills in every member of the general public, you will have either Type I or Type II errors. That being so do you prefer to see people who are well but concerned about something that might be an early sign of something serious, or do you prefer that they stay home with what could have been treatable if caught earlier?

  14. Grapmag says:

    >>You can’t have it both ways. If patients only come to be seen when they have spurting arterial wounds but “don’t want to bother you over something minor” you won’t see the man with a case of “indigestion” or his wife who is “acting funny.” <<

    Please, you are smarter than that.

    The point is NOT that it is not perfect (everyone who makes an appointment has a serious problem) but the current ratio creates an incredible amount of waste. We are NOT helpless when it comes to improving the utilization of resources.

    The medico-industrial complex has done a great job of terrifying the public into useless testing and inefficient use of resources. Doctors are willing participants in the process – and in fact are incented to see as many patients and perform as many procedures as possible.

    If you give a doctor a patient ….

    He’s gonna want to do a test

    and if you let him do a test

    the test is going to come back with a trivial abnormality

    and if there is a trivial abnormality,

    he’s gonna want to do a test ….

    Actually I am surprised that you have time to answer this post.
    Don’t you know that the gurgling in your stomach could be due to a tumor of your duodenum that is partially blocking stomach outflow? When’s the last time you had an unproven triple ultrasound screening? I don’t like the way you look – I am going to have to suggest a PURE scan to check the calcium in you heart arteries – can’t ever be too safe, now can we? You know the protein level on your chem 21 is a little bit high, we better repeat that and do an electrophoresis ….

    Jenny – make sure Mr Overshoot has an appointment in 6 weeks to go over everything we order today ….

  15. weing says:

    What the hell is a PURE scan?

  16. weing says:

    Oh. I know it as the Calcium score by EBT. Haven’t used it yet. Where would we be without the worried well? Frequently it’s a waste of time explaining to them that they don’t need those tests. They’ll find another doctor who will take them seriously. OTOH there is the saying “If you don’t do a test that’s not indicated, how will you ever make the diagnosis?”

  17. Joe says:

    Ah yes, the St. Louis Business Journal, where else would we learn medicine; our friend?

  18. As a layperson, I think the hardest thing is knowing when something is serious and treatable. Like I suddenly have black spots moving in my visual field. It is Saturday night and I have a 4 hour drive alone starting early Sunday morning. I live in a rural area and remember a friend telling the story of her retina becoming detached when she was driving alone a long way from home, but I don’t remember the details if I ever heard them. Do I go to the ER? I didn’t. I hit the road and got to the doctor when I got home, but my decision was not based on knowledge and I was worried driving.

    In my experience taking care of myself, loved ones and pets, if I’m dealing with an MD or DVM who I know well, I can phone and get the required information to make an informed decision. But in today’s legal climate that is impossible if the doctor doesn’t know you well.

  19. Grapmag says:

    rjstan – well said, reasonable and actionable patient information can be difficult to find. I certainly sympathize with doctors that don’t want to give advice over the phone in the current legal climate.

    I hope that everything came out okay for you. Did you end up with a CT scan of the head? Followed by an MRI? And a neurology consultation? told to follow up with ophthamology? Or did your physician pursue watchful waiting?

    I was referring to even simpler cases – colds for instance. Mild non-chronic musculoskeletal pain. Mild anxiety. Mild insomnia.

    Then we can talk about the diseases that the pharmaceutical companies have created, advertised and legitimized in order to sell a drug that didn’t have a disease to start with or had a disease with a narrow definition that the drug company was able to widen with expanded ‘off label’ prescribing.

    Medicalization: “To medicalize the human condition is to apply a diagnostic label to various unpleasant or undesirable feelings or behaviors that are not distinctly abnormal but that fall within a gray area not readily distinguishable from the range of experiences that are often inescapable aspects of the fate of being human.” Paul Chodoff, M.D.

  20. weing says:

    Are you referring to restless legs syndrome by chance?

  21. oderb says:

    Wonderful post, but still exhibiting the bias inherent in conventional medicine. If it’s not a phama developed and promoted product it can’t be effective and shouldn’t be recommended. One example for the flu is Sambucol, over the counter, developed by an eminent Israeli virologist in the 90’s, patented based on the elderberry, and shown is several RCT’s to reduce flu symptoms by 4 days with no side effects.

    Why isn’t every MD aware of this inexpensive, safe, widely available and reasonably well studied treatment?

    Because it’s sold in the health food store and not at CVS or not sold as a prescription? Because it’s not hawked by the drug rep, or mentioned at medical conferences or advertised in medical journals?

    That’s one of my problems with this site. The assumption that no medicine not developed by a mainstream pharma should be given consideration, or that if it’s not in the mainstream there’s no science behind it.

  22. Harriet Hall says:


    Saying that we don’t recommend products just because they are not pharma developed and promoted is a vile calumny. I have 3 criteria for recommending a product: (1) is there good evidence that it’s effective, (2) is there good evidence that it’s safe, and (3) is there a reliable source that is pure and can be trusted to contain the amount of active ingredient stated on the label. If an herbal remedy met those criteria, I would recommend it; if a pharmaceutical didn’t meet those criteria, I wouldn’t recommend it.

    I was only able to find one abstract of a clinical Sambucol study from 1995, and it didn’t say how many subjects there were, and it was published in an alternative medicine journal, so I didn’t find the evidence very convincing. You say there are several RCTs – perhaps you could give us the references. Are they good studies? Do they pass Bausell’s 4-point checklist? (See

    The Natural Medicines Comprehensive Database tries to review all the published literature and it only lists one study, and it rates elderberry as “possibly safe” and “possibly effective” – it doesn’t even rate their “probably effective” rating, much less an “effective” rating. The Desktop Guide to Complementary and Alternative Medicine: An Evidence-Based Approach by Ernst et al does not even list elderberry or Sambucol.

    And since most cases of influenza resolve without treatment, the option of not treating with Sambucol OR pharmaceuticals is worth considering.

Comments are closed.