How Doctors Harm their Patients by Prescribing Placebos

When it comes to human attempts to improve their health, placebos are among the most commonly used treatments. “Alternative medicine” consists of nothing but placebos. For example, acupuncture doesn’t actually work; instead it makes the patient feel better, because of the placebo effect—they expected that there must be some improvement, thus they feel like they have gotten better.

Unfortunately, it’s not just practitioners of alternative medicine who routinely prescribe placebos. Even in legitimate clinics you can get placebo prescriptions from people who are supposed to be legitimate doctors. Since placebos can make a patient feel better, somebody might ask what the harm is. Why can’t legitimate doctors prescribe some placebo to a worried parent whose child has a cold? After all, it will make the parent feel better while they wait for the cold to end. Unfortunately, carelessly prescribing unnecessary pills can result in serious negative consequences.

When I was a child, each time I got a cold my doctor prescribed homeopathic medicine. My mother took me to a legitimate clinic expecting evidence-based medicine. My doctor wasn’t just some self-proclaimed healthcare practitioner, she was a real doctor who had gotten real education in medicine. Nonetheless, my doctor misinformed my mother and explained to her that homeopathy was a milder form of natural cure with no side effects. Where was the harm? Maybe the fact that telling a poor single mother to waste her money on a placebo is immoral, because it causes her to pointlessly loose what little money she has.

Years later, as a teen, I got access to the Internet and I learned that homeopathy was a placebo with no proven health benefits. I was furious. If my doctor hadn’t already retired by then, I would have wanted to get her fired. If she truly believed that homeopathy actually works, then she was uneducated and unfit to be a doctor. If she knowingly prescribed an expensive placebo to an impoverished single mother, then she was an asshole.

As a result of my childhood experiences with homeopathy, I started to mistrust doctors and question everything they told me.

But there is also a different potential outcome. The doctor prescribes a placebo, and the patient gets better, because they have a disease that tends to improve without any treatment. Alternatively, the doctor prescribes real medicine and a placebo simultaneously, and the patient gets better. Afterwards, the patient imagines that the placebo is what cured them. They start routinely using alternative medicine and avoid real doctors. Until one day they get a cancer and try to cure it naturally with coffee enemas and a diet consisting of fruit juices. And then they die.

Some people tend to claim that placebos make patients feel better, thus they are useful. If the problem was pain, then feeling better because of a placebo really would mean a tangible improvement. But for most other health problems, feeling better for a day after getting some placebo treatment isn’t really solving the underlying problem. If you have a cancer, feeling better after a coffee enema won’t make the cancer disappear.

Where I live, many vets use antibiotics as a universal placebo. If they don’t know what exactly is wrong with some animal or they know that they are dealing with some disease that will probably go away on its own after a while, they prescribe antibiotics. A few years ago my dog was coughing, so we took him to a vet, who examined my dog and prescribed antibiotics. Here’s the conversation that followed. Me: “But isn’t this probably a virus?” Vet: “Yes.” Me: “Antibiotics kill bacteria, why are you prescribing them for a virus?” Vet: “When a dog gets sick from a virus, it weakens his immune system. Thus it becomes easier for bacteria to infect him. Antibiotics are meant to preemptively prevent that from happening until your dog gets better.” WTF? “Let’s use some antibiotics just in case” is a problematic attitude. Unnecessary usage of antibiotics also kills gut bacteria and contributes to the emergence of superbugs. By the way, in about a week my dog got better on his own. My opinion about this vet, however, never recovered.

I suspect that some vets treat antibiotics as a placebo, because they want to pretend that they are doing something. If a pet owner goes to a vet who says, “No treatment is necessary, your dog will probably get better on his own in a few days,” then the pet owner will wonder why they are even paying for this vet consultation. On top of that, next time their pet gets similar symptoms, they probably won’t go to a vet at all. Some less than ethical vets seem to want to ensure that clients keep coming back again and again, so they prescribe antibiotics as a universal placebo for everything.

And it is not just placebos, often doctors prescribe medicine that isn’t truly necessary. Back when I was in my late teens, I had to go to a doctor every time I got a cold. My school demanded a piece of paper signed by a doctor to prove that I truly was sick and wasn’t just skipping lessons. Each time my doctor prescribed me a long list of various medicines. Nasal sprays, pills, and so on. As she was giving me all those prescriptions, I just nodded and pretended to listen. The moment I left my doctor’s office, I deposited all those prescriptions directly in the garbage bin. The main reason why I didn’t use all the medicine my doctor prescribed was the fact that this stuff was expensive and my family was poor. The second reason was that I didn’t feel like I needed them. For me cold symptoms were always relatively mild. I got three days of slightly sore throat, and I had to regularly blow my nose for a week. That was it. Of course, medicine can alleviate various symptoms caused by having a cold, but personally I didn’t feel like I needed all that. If you treat a cold, it will be gone in seven days. If you do nothing, it will take a whole week. For me doing nothing and just staying at home, chilling out, and getting plenty of rest worked just fine.

Nonetheless, I never dared to tell my doctor that I simply needed the paper for my school. I lied that I was dutifully using all the stuff she prescribed. I feared that if I told the truth, she would criticize me for not following her recommendations and being careless about my health.

This left me with a nasty dilemma—on one hand, I am not a doctor and I am not qualified to determine, which medicines are necessary and which ones are optional. On the other hand, I couldn’t explain to my doctor that I wanted her to prescribe me only the medicine that was necessary and skip all the optional stuff. I was afraid to confront and question a doctor who knew about medicine much more than I did. Simultaneously, I felt that her recommendations for me were wrong.

The other problem that annoys me is doctors’ reluctance to explain what some medicine would do and why exactly I would benefit from taking it. Doctors probably feel that it is too much hassle to try to dumb down the explanation so as to make it comprehensible for the average person. Yet I feel that not knowing what happens with my body bothers me. Besides, I am naturally curious. While I do trust doctors’ advice, I still want to know what is going on. For example, after a surgery my nurse gave me various pills and injections without even telling me what each of those contained until I started asking “what is this?” all the time.

I understand why some doctors feel compelled to prescribe something and why they often don’t bother explaining to the patient what exactly some treatment will do to their body. Patients are different. There are some patients who read scientific journals for a hobby (me). There are also patients who don’t even know what a randomized controlled trial is, nor do they know even basic facts about science and human biology. There are patients who are naturally curious and want to know everything about their treatment (me). There are also patients who don’t want to know the details and prefer the doctor to simply tell them what to do. There are patients who prefer not to use any medicine that isn’t necessary (me; only because pills are expensive and I am sort of poor). There are also patients who want to try various things hoping to find something that can alleviate their symptoms.

I have heard about doctors struggling to deal with patient expectations. A patient expects some prescription and they don’t want to be told to go home and wait for their problem to go away on its own. Or a parent feels stressed and wants to help their child even when it’s just a cold and there’s nothing they can do about it. So they prescribe something. Hopefully something that actually helps rather than some useless homeopathy.

Still, we need better communication between doctors and their patients. And prescribing placebos can have harmful consequences.


  1. says

    John Morales @#1

    From the linked text: “To summarize, there is a place for placebo therapy, as long as: its use is transparent and the patient’s autonomy and right-to-know is respected.”

    Yeah, but in practice there is little transparency in real life. More importantly, nobody gives patients placebos for free, instead they are charged money for them. Selling an expensive placebo is fraudulent. Besides, if later a patient finds out that they were defrauded and sold a placebo, they will get very angry (as I did, I never forgave that doctor who prescribed homeopathy to my then impoverished mother).

  2. John Morales says

    Sure, Andreas. I don’t disagree with anything you wrote.

    As is my wont, I immediately thought of edge cases.

  3. xohjoh2n says

    Just a point: you keep equating the effect of placebo with the patient “feeling better”, but the placebo effect is more complicated than that. Sure, it’s not going to cure your late-stage cancer. But in many cases it does indeed provide objectively measurable improvements over and above just “feeling better”. And then there is its evil twin, the nocebo effect, whereby patients given placebo experience negative side-effects as if they were on a real drug too.

    There is a reason why good clinical trials are placebo controlled – the control group is not the do-nothing untreated group, versus the real drug, because even if the real drug had no direct effect on the condition in question, it would likely still show a statistically significant improvement over the untreated group. For a new drug to be accepted as a valid treatment, it has to do better than placebo, not better than nothing.

    Ben Goldacre used to write about this a fair bit. Where he (as a registered GP) drew the line in terms of what he could see a doctor legitimately prescribing was honest informed consent, and he spent a lot of time going after homeopaths because they didn’t. (Luckily, as he pointed out, there are several trials that show the placebo effect can be maintained to an extent even if you tell the patient up front “this is a sugar pill, it contains no actual medicine at all, would you still like to try it?” He also pointed out that the nocebo effect puts a pretty big hole under the waterline of those defending homeopathy by saying at least it does no harm.)

  4. says

    I was listening to TWiV the other day and they were discussing a Remdesivir trial with one of the subjects who – as is appropriate – does not know if they are in the test group or the control group. Apparently, they gave them a different/irrelevant intra-muscular vaccine as a placebo, so they couldn’t tell if they were in the placebo group because of muscular soreness at the injection site. That’s pretty interesting.

    Controlled studies with a placebo are the only situation in which placebos are appropriate. Homeopathy and other scams are a “rip off” not a “placebo” – you’re paying for medicine and they’re giving you sugar.

  5. says

    xohjoh2n @#4

    I have read Ben Goldacre’s “Bad Pharma.”

    Of course you need a control group in an RCT. Often patients tend to get better after some time passes anyway. Hell, even going to a doctor and taking about your problem can potentially make a patient feel better, because they feel like they have done at least something. Especially if the doctor reassures the patient and promises them that they will get better.

    I could potentially imagine some scenario where prescribing a placebo is ethical. For example, (1) the patient’s problem is pain; (2) the doctor gives them the placebo for free; (3) the doctor uses carefully worded language in order to not mislead the patient. In real life something like this pretty much never happens. Prescribing expensive homeopathy or antibiotics for a patient who is coughing should not be perceived as ethical.

    Marcus @#5

    Controlled studies with a placebo are the only situation in which placebos are appropriate.

    It’s possible to argue that, whenever possible, RCTs should have a test group on which you test the new medicine versus a control group which takes whatever is the current best treatment. This way you are not testing whether the new drug is better than nothing, instead you are testing whether the new drug is better than currently used drugs.

  6. xohjoh2n says

    @5,6 Yes, RCTs have a control group. The question is what exactly is the control group. Yes, for some conditions and/or some patients they will get “better after some time”. That isn’t the point. They sometimes get better faster or betterer *on placebo* than untreated. (I’m sure then you remember Ben’s comments that 4 sugar pills are better than 2, a saline injection is better than sugar pills, and both have improved efficacy if administered by someone in a white labcoat.) That is why the base control is placebo and not untreated, and why placebo is *not* equivalent to no treatment. And that is *not* a matter of just “feeling” better, or being reassured or anything like that. There are studies that show objective measurable physiological changes under placebo.

    I think Ben’s take on ethical prescription of placebo was more nuanced, I believe on the order of: if we don’t have any better proven treatment, and given the placebo effect appears to be a really real thing, and assuming informed consent, can it be entirely ethical to discount it as an option, to deny the patient even the limited benefit it might bring?

  7. says

    xohjoh2n @#7

    Yes, for some conditions and/or some patients they will get “better after some time”. That isn’t the point.

    When it comes to pseudoscience, there have been plenty of trials without any control group at all. Just take a bunch of people with cold, give them some vitamin dietary supplement, they get better after a week, announce that the trial was a success, because everybody got better after taking the vitamin. I mentioned the necessity for a control group, because advocates of alternative medicine routinely try to downplay the fact that often patients can get better even if you do nothing.

    I’m sure then you remember Ben’s comments that 4 sugar pills are better than 2, a saline injection is better than sugar pills, and both have improved efficacy if administered by someone in a white labcoat.

    So what? How does this justify defrauding patients by selling them placebos?

    And that is *not* a matter of just “feeling” better, or being reassured or anything like that. There are studies that show objective measurable physiological changes under placebo.

    Yes, sure, placebos have noticeable effects. This justifies a doctor telling their patient to pay countless dollars for a placebo? Hell no. That’s fraud.

    Several doctors and vets have tried to prescribe me placebos in my life. Homeopathy, vitamin C and antibiotics for colds, all kinds of weird and ridiculously expensive dietary supplements for my dogs—I have been told to use all kinds of medicine that was clearly unnecessary. Not a single doctor who told me to use these things actually said: “Here’s a free sugar pill, it’s a placebo but it might make you feel better anyway.” Instead they told me to buy the damn thing and pretended that whatever they recommended would have actual benefits. Poor people hate being defrauded of what little money they have. And I hate being lied to.

    As a patient, I should not be obliged to try to decipher when my doctor is telling me lies and when they are telling me truth. If doctors routinely lie to their patients (something I have experienced from more than one doctor and vet), then patients lose trust in doctors. More importantly, as a teen I was afraid to argue with my doctor and ask her questions, thus I got into habit of throwing in the trash bin every single prescription she gave me. Like this, I could have failed to take some medicine that I actually needed. When I assume that most of the pills I get prescribed are placebos, I won’t take a single one of them, thus I can fail to take the one pill I actually needed for real.

    if we don’t have any better proven treatment, and given the placebo effect appears to be a really real thing, and assuming informed consent, can it be entirely ethical to discount it as an option, to deny the patient even the limited benefit it might bring?

    Some people have little money and they are greatly harmed when financially defrauded. So far in my life I have never met a single doctor who would offer me sugar pills for free, instead every single attempt to defraud me required me to pay my own money for the damn placebo.

Leave a Reply