$3.7 million to study quack medicine at a leading cancer center

Sometimes I'm not sure whether the best response to pseudoscience is to ignore it, or to patiently try to explain why it's wrong, or to get mad.

This week I'm mad.

My anger and frustration was triggered by a tweet from Memorial Sloan-Kettering's Integrative Medicine account, shown here:
For those who don't know, Memorial Sloan-Kettering Cancer Center is one of the world's leading cancer centers, both for treatment and research. If you are diagnosed with cancer, MSK is one of the best places to go.

But not everything at MSK is world class. Unfortunately, they have an "integrative medicine" center that offers a mixture of therapies ranging from helpful to benign to useless. One of their biggest activities is acupuncture, which they claim offers a wide range of benefits to cancer patients.

The MSK tweet shown here was boasting about a new, $3.7 million study funded by NIH to study the effect of acupuncture on pain that cancer patients experience from chemotherapy and bone-marrow transplants.

Here's why I'm mad: cancer patients are extremely vulnerable, often suffering the most frightening and difficult experience of their lives. They are completely dependent on medical experts to help them. When a place like MSK suggests a treatment, patients take it very seriously–as they should. But they really have no choice: a cancer patient cannot easily look for a second opinion, or switch hospitals or doctors. Even if they have the money (and cancer treatment is extremely expensive), switching hospitals might involve a long interruption with no treatment, during which they could die, and it might also involve traveling far from their home.

Offering these patients ineffective treatments based on pseudoscience–and make no mistake, that's what acupuncture is–is immoral. Now, I strongly suspect that the MSK's "integrative medicine" doctors sincerely believe that acupuncture works. Their director, Jun Mao, is clearly a true believer, as explained in this profile of him on the MSK website. But that doesn't make it okay.

I've written about acupuncture many times before (here, here, here, and here, for example), but let me explain afresh why it is nonsense.

Acupuncture is based on a pre-scientific notion, invented long before humans understood physiology, chemistry, neurology, or even basic physics, which posits that a mysterious life force, called "qi," flows through the body on energy lines called meridians. As explained in this article by MSK's Jun Mao:
"According to traditional Chinese medicine ... interruption or obstruction of qi was believed to make one vulnerable to illness. The insertion of needles at specific meridian acupoints was thought to regulate the flow of qi, thus producing therapeutic benefit."
Today we know that none of this exists. There is no qi, and there are no meridians. In that same article, Jun Mao continued by admitting that
"the ideas of qi and meridians are inconsistent with the modern understanding of human anatomy and physiology."
And yet this is what they offer to patients at MSK.

Just to be certain, I read one of the latest studies from MSK, published early this year, which claims to show that acupuncture relieves nausea, drowsiness, and lack of appetite in multiple myeloma patients who were going through stem cell transplants.

It's a mess: totally unconvincing, and a textbook case of p-hacking (or data dredging). The paper describes a very small study, with just 60 patients total, in which they measured literally dozens of possible outcomes: overall symptom score at 3 different time points, a different score at 3 time points, each of 13 symptoms individually, and more. I counted 24 different p-values, most of them not even close to significant, but they fixated on the 3 that reached statistical significance. The two groups of 30 patients weren't properly balanced: the sham acupuncture group started out with more severe symptoms according to their own scoring metric, and Figure 2 in the paper makes it pretty clear that there was no genuine difference in the effects of real versus sham acupuncture.

But they got it published (in a mediocre journal), so now they point to it as "proof" that acupuncture works for cancer patients. This study, bad as it is, appears to be the basis of the $3.7 million NIH grant that they're now going to use, they say, in "a larger study in 300 patients to confirm these previous findings."

And there you go: the goal of the new study, according to the scientists themselves, is not to see if the treatment works, but to confirm their pre-existing belief that acupuncture works. Or, as one scientist remarked on Twitter, "they already have a result in mind, the whole wording of this suggests that they EXPECT a positive outcome. How did this get funded exactly?"

Good question.

So I'm mad. I'm mad that NIH is spending millions of dollars on yet another study of a quack treatment (acupuncture) that should have been abandoned decades ago, but that persists because people make money off it. (And, as others have explained in detail, acupuncture is actually a huge scam that former Chinese dictator Mao Zedong foisted on his own people, because he couldn't afford to offer them real medicine. For a good exposé of Chairman Mao's scam, see this 2013 Slate piece.)

But I'm even more upset that doctors at one of the world's leading cancer centers are telling desperately ill patients, who trust them with their lives, that sticking needles into their bodies at bogus "acupuncture points" will relieve the pain and nausea of chemotherapy, or help them with other symptoms of cancer. I'm willing to bet that most MSK doctors don't believe any of this, but they don't want to invest the time or energy to try to stop it.

(I am somewhat reassured by the fact that MSK's Twitter account has nearly 75,000 followers, while it's integrative medicine Twitter account has just 110.)

Or perhaps they are "shruggies": doctors who don't believe in nonsense, but figure it's probably harmless so they don't really object. To them I suggest this: read Dr. Val Jones's account of how she too was a shruggie, until she realized that pseudoscience causes real harm.

And finally, let me point to this study in JAMA Oncology from last year, by doctors from Yale, which looked at the use of so-called complementary therapies among cancer patients. They found that
"Patients who received complementary medicine were more likely to refuse other conventional cancer treatment, and had a higher risk of death than no complementary medicine."
And also see this 2017 study from the Journal of the National Cancer Institute, which found that patients who used alternative medicine were 2.5 times more likely to die than patients who stuck to modern medical treatments.

That's right, Memorial Sloan-Kettering: patients who use non-traditional therapies are twice as likely to die. That's why I'm mad. This is not okay.

Gluten-free diet has no effect on autism, according to a new study

Parents of autistic children are constantly seeking new treatments. Autism spectrum disorder, or ASD, is a developmental disorder that causes problems in social interaction and communication, ranging from mild to severe. It's an extremely challenging condition for parents, and as of today there is no cure.

However, there are plenty of websites that offer treatments for autism, many of them unproven. One of the more common claims is that autistic children will benefit from a gluten-free, casein-free diet. There has been some weakly supportive evidence for this idea, such as this 2012 report from Penn State, but that study was based entirely on interviews with parents. Interviews are notoriously unreliable for scientific data collection.

Perhaps because so few effective treatments are available, many parents of autistic children have tried gluten-free diets, in the hope that perhaps they might work. (One can find entire books dedicated to this diet.)

The science behind the idea that gluten or casein causes (or worsens) autism has always been sketchy. The push for diet-based treatments has its origins in the anti-vaccine movement, beginning with the fraudulent 1998 study (eventually retracted) in The Lancet led by Andrew Wakefield, a former gastroenterologist who lost his medical license after his fraud was discovered. Wakefield claimed that autism was caused by a "leaky gut," which somehow allowed vaccine particles to make their way to the brain, which in turn caused autism. That chain of events was never supported by scientific evidence. Nonetheless, it morphed into the hypothesis that gluten (or casein) somehow leaks out of the intestines and causes some symptoms of autism. There's no evidence to support that either.

(Despite losing his medical license, Wakefield has become a leading voice in the anti-vaccine movement, making speeches and even movies to try to convince parents not to vaccinate their kids. Many journalists and scientists have written about him and the harm that he's done, but that's not my topic today.)

Another hypothesis, according to WebMD, is that autistic children have some kind of allergic reaction to gluten. There is no good evidence for this either.

Surprisingly, virtually no good studies have asked the question, do gluten-containing foods actually cause the symptoms of autism? Now we have a carefully-done study that provides an answer.

The new study, just published in the Journal of Autism and Developmental Disorders, is the first randomized, well-controlled study of gluten-free diets in children with autism. The scientists, all from the University of Warsaw, Poland, recruited 66 children, and assigned half of them at random to a gluten-free diet. The other half were given a normal diet, with at least one meal a day containing gluten, for 6 months. The children ranged from 3 to 5 years old. After 6 months, the scientists evaluated all children using multiple standardized measurements of autistic behavior.

The results were very clear: the study found no difference between the diets. None of the core symptoms of ASD were different between children in the two groups, and there were no differences in gastrointestinal symptoms either. As the study itself stated:
"There is no evidence either against or in favor of gluten avoidance for managing symptoms of ASD in children." 
This study should put to rest all of the claims that a gluten-free diet can somehow improve the symptoms of autism. It doesn't provide an easy answer for parents, and the medical community still needs to do much more work to find better treatments. But let's hope that parents get the message: don't feed your autistic child a restricted diet.