Skip to content

“How to Assess Internet Cures Without Falling for Dangerous Pseudoscience”

Science writer Julie Rehmeyer discusses her own story:

Five years ago, against practically anyone’s better judgment, I knowingly abandoned any semblance of medical evidence to follow the bizarre-sounding health advice of strangers on the internet. The treatment was extreme, expensive, and potentially dangerous.

If that sounds like a terrible idea to you, imagine how it must have felt to a science journalist like me, trained to value evidence above all. A decade ago, I never would have believed I’d do such a lunatic thing.

But I was desperately, desperately ill. . . . So I took a deep dive into the murky world of untested treatments. The incredible thing is, I found something that brought astonishing improvements, even if not quite a cure . . .

The whole article is worth reading in that she provides advice to people trying to evaluate advice from uncertain sources. It’s a challenge. In a world where experienced journalists can get fooled by junk science in Lancet, PNAS, etc. (as Rehmeyer puts it, “‘Authoritative’ sources also need evaluation and scrutiny”), how much more challenging must it be for an individual patient with a personal stake in the matter to make sense of conflicting recommendations. Rehmeyer’s advice is sensible, and it points to larger questions of how researchers and policymakers can evaluate claims.

This often comes up when I discuss pop-science claims based on peer-reviewed articles that get hyped by Freakonomics, NPR, Ted, etc. If the experts get it wrong, what can the person on the street do?

P.S. Rehmeyer also wrote a book about her experiences, which I reviewed here and here.


  1. Paul says:

    Western medicine is often very broad-brush and average-oriented. My wife has been told by multiple doctors that experiences and symptoms she has *actually had* are impossible because some study said that side effect X is uncommon. I am eagerly waiting for us to stop using machine learning only to make people click on ads, and start using it to develop more complex, individualized understandings of how people respond to treatments.

    • Kyle C says:

      I recently learned from a nephrologist that kidney pain doesn’t exist, because kidneys have no nerves. OK then! Maybe someone actually did stab me in that exact area with an ice pick?

      • Adam S says:

        Many of our internal organs do not have pain nerves in them, but their capsules do and are exquisitely sensitive to distension or nearby inflammation. For instance, a liver or kidney could be filled with tumors and have almost no symptoms, but when one develops near the surface and pushes on the capsule, there is intense and unrelenting “cancer pain.” The nephrologist is technically correct but did not communicate what was actually meant, and probably just meant it as a “medical factoid.”

        I’ll leave my thoughts on the article if I get enough time to write out a full reply.

      • Clyde Schechter says:

        Well, your nephrologist is right that kidneys have no nerves. If you are feeling pain in the location of your kidneys it is likely arising from an adjacent structure that does have nerves. The primary pathology could even be in the kidney, and it can cause irritation in an adjacent structure through pressure (if the kidney is swollen) or through diffusion of chemical irritants if the surface of the kidney is not disrupted. But strictly speaking, the pain is not coming from the kidney. A distinction without a difference, perhaps.

        • Clyde Schechter says:

          Sorry, that should read …if the surface of the kidney IS disrupted…

        • Kyle C says:

          Thank you, I didn’t doubt that he was right about the nerves, but “kidney pain” is such a common phenomenon that it seems bizarre to dismiss. He and his assistant tried to convince me I may have pulled a muscle, when I had the classic symptoms of a stable kidney stone.

      • Jeff Walker says:

        kidneys have a rich set of nerves, just not any carrying pain. There are two types of nerves: 1) sensory fibers in the renal pelvis, which is the part that collects the urine from all of the kidney and drains into the ureter (the tube exiting the kidney with the urine), which sense pressure due to fluid levels. I don’t think this rises to consciousness as “pain”. and 2) sympathetic (autonomic motor) throughout the kidney, specifically, to all the smooth muscle in the walls of the arteries/arterioles, including the juxtaglomerular cells. These are nerves carrying information from the brain/spinal cord to these muscles to signal them to contract and to secrete Renin (in the case of the JG cells).

        • Kyle C says:

          Wow, interesting. So the $64K question then: what explains the oft-reported experience of stabbing pain exactly in the kidney area, which persists and radiates, over a period of hours, into the abdomen and down the urinary tract, before gradually fading? Lay people call this “kidney pain.”

    • Z says:

      Yes, enough western medicine, time for natural treatments extracted from the leaves of the random forest

    • Dzhaughn says:

      Machine learning is broad-brush and average oriented. So is science.

      And do you really find that those ads are so sharply focused? I sure don’t. But then I block most of them.

      • Martha (Smith) says:

        Science isn’t necessarily broad-brush and average oriented — but a lot of science (especially of human beings) is conducted in ways that only look at broad-brush and average phenomena. The move toward “personalized” (I would say “individualized”) medicine is one exception. N of 1 experiments are also exceptions.

        • Dzhaughn says:

          What does an N=1 experiment do other than falsify or fail to falsify “broad brush” theories?

          • Martha (Smith) says:

            N = 1 experiments have at least some potential for discovering possibilities.. They can be especially useful in deciding between possible treatments for individuals. And, when conducted with several individuals (individually, of course) have potential for forming “medium brush” theories focused on specialized populations.

  2. Dzhaughn says:

    Can anyone recommend an article that explains how to enchance the effectiveness of a placebo?

    For instance, this article illustrates the sound advice of starting a treatment when the condition is much worse than average, to leverage regression to the mean. Also, make your treatment coincide with a vacation.

    I am glad the author feels better. That is not evidence of learning. (Hey that’s another good placebo enhancer: try to gain wisdom from your condition and/or treatment. Having delivered its message, the ghost can leave.)

  3. Pancake, a bloody one says:

    My god, did that read like something off the shopping channel. With all the appeal to emotion and links to buy her book… as Dzhaughn, I’m glad that she feels better, but this whole thing just seems ridiculous. There seemed to be some surface level “be skeptical!” message, but it felt really insincere and tacked on. I was glad to see some proper skepticism in the comments.

    Not to say that there wouldn’t be problems with the way medicine is practiced, but indeed, these trivial observations about the problems are constantly used to sell us all sorts of oils from a variety of snakes–or indeed books about conquering the medical community or whatever.

  4. Thanatos Savehn says:

    I don’t think we’re ready to tackle the problem of how to infer fake causes for fake diseases. That having been said, as a non-doctor whose Dad was killed by doctors who misdiagnosed his illness, gave him a powerful fluorinated steroid and then failed to timely diagnose the obviously resultant rhabdomyolysis – (despite the fact that I, though an idiot, thanks to PubMed almost immediately figured out that this had been a recognized sequela of the treatment for 60 years; as confirmed, take my word for it, by his discharge diagnosis and Death Certificate) – I’m now a firm believer in reading the literature and thinking (and deciding) for yourself rather than merely submitting to the unthinking assembly line medicine that pervades current practice.

    • Martha (Smith) says:

      “I’m now a firm believer in reading the literature and thinking (and deciding) for yourself rather than merely submitting to the unthinking assembly line medicine that pervades current practice.”

      I agree. Some additional suggestions Ive learned from experience:

      1. If it seems to involve a muscle or tendon and seemed to occur from using that body part, ask for a referral to a physical therapist. My experience is that they are usually better at diagnosing (technically ,”evaluating”) the problem, and in some cases can help you learn to avoid or mitigate the problem in the future. (e.g., a physical therapist trained me to use my traps and lats, which reduces strain on my gimpy pectoralis minor tendons.)

      2. If a prescription drug has any chance of producing mental symptoms, look it up on a “harm reduction” website before using. Often such drugs are used “recreationally” and the harm reduction websites are likely to give information about side effects that is not on package inserts.

      3. If the physician does not give options for treatment, ask about them.

  5. Ian Fellows says:

    My favorite quote:

    “And when it came down to it, I was out of reasonable ideas—only unreasonable ones were left.”

    I am hyper-evidence based when it comes to my health, but what to you do when there is no evidence to evaluate? Just lay down and die on the best advice of your doctors?

  6. Tony Boyles says:

    [Eliezer Yudkowsky’s new book]( was (partly) inspired by a similar experience treating his spouse’s Seasonal Affective Disorder. He generalizes the advice well beyond the scope of medicine, to good and interesting effect.

Leave a Reply