A recently published ethics paper addresses the issue of whether or not it is proper for crowdfunding sites, like GoFundMe, to allow campaigns to fund dubious medical treatments. This question is also part of a more general issue – how tech companies have replaced traditional industries and institutions thereby bypassing existing mechanisms of safety, justice, and quality control. On the medical issue, the authors write:

Recent studies have shown that many individuals are using crowdfunding to finance access to scientifically unsupported medical treatments. Recently, GoFundMe prohibited campaigns for antivaccination groups on the grounds that they “promote misinformation about vaccines” and for treatment at a German clinic offering unproven cancer treatments due to “the need to make sure people are equipped to make well‐informed decisions.” GoFundMe has not taken any additional actions to regulate the much larger presence of campaigns seeking to fund unproven medical interventions on the platform. In this article, we make the ethical case for intervention by GoFundMe and other crowdfunding platforms.

The basic principle is that tech companies still retain an ethical and legal responsibility for how their platforms and technology are used. Most applications and social media outlets begin as an unregulated peer-to-peer environment, just facilitating an individual exchange between two private citizens. In a way this is a Libertarian nirvana. However, as such applications scale up the downsides that have already had to be dealt with in the traditional industries they are supplanting begin to resurface.

We can take any such app as an example, such as Uber. The Uber app, which I use, is very convenient. They have definitely made a better mousetrap. But as Uber has grown huge, we begin to question what responsibility they have to their drivers and riders. How much do they have to vet drivers to protect riders? What kinds of protections and benefits should they offer drivers? Did they just replace a regulated industry with an unregulated one? The same questions have arisen with Air bnb, which critics warn is being used to simply create de facto hotels that skirt regulations.

There are two principles here. The first has to do with the role of regulations in general to protect the public from exploitation of various sorts. I don’t want to go down that rabbit hole entirely, but just summarize my position as this. I support carefully considered and monitored regulation to keep society functioning optimally and prevent exploitation, externalizing costs, unfair competition, and the like. The only truly free market is a regulated one, because an unregulated market will become increasingly distorted over time as the powerful use their power to obtain more power (rather than play fairly).  At the same time we have to be humble regarding unintended consequences, which is why regulations need to be minimalistic and monitored for their effects. If you buy some version of this basic premise, and are not an anti-regulation purist, then it should be concerning that effective regulations are being nullified by an app. This is happening without any elected representatives of the people making any decisions – without any public representation. In the extreme this can evolve into a tech oligarchy.

The second principle is fairness. If one industry that is regulated is competing with another that is unregulated, that gives an unfair advantage to the unregulated one (regardless of what you think about regulations in general).

Continue Reading »

Move over CRISPR – enter Prime Editing.

Maybe. What we can say is that the pace of technological advancement in genetic editing is advancing so quickly it’s hard to keep up. Now a new study, published in Nature, details a new method for editing called prime editing. The authors write:

Here we describe prime editing, a versatile and precise genome editing method that directly writes new genetic information into a specified DNA site using a catalytically impaired Cas9 fused to an engineered reverse transcriptase, programmed with a prime editing guide RNA (pegRNA) that both specifies the target site and encodes the desired edit.

So actually this is built off of CRISPR technology, using a Cas9 component, and then pairing it with a bit of RNA (pegRNA) that both targets the bit of the genome you want to edit and also has the new code you want to insert. Insertion is accomplished by the enzyme reverse transcriptase. How does this compare to existing methods for gene editing? The authors again:

“Prime editing offers efficiency and product purity advantages over homology-directed repair, complementary strengths and weaknesses compared to base editing, and much lower off-target editing than Cas9 nuclease at known Cas9 off-target sites. Prime editing substantially expands the scope and capabilities of genome editing, and in principle could correct about 89% of known pathogenic human genetic variants.”

It is more precise and has fewer errors, and is able to target 89% of known genetic diseases. It cannot fix errors where a gene is entirely missing, or where there are too many copies of a gene. They tested the method on two human genetic diseases in human cells, Tay Sachs and sickle cell anemia, with success.

Continue Reading »

In the US we have semi-compulsory vaccination. Keeping up to date on the vaccine schedule is required in order to attend public school. However, families always have the option of attending private schools, which can determine their own vaccination policy, or they can home-school. Also, each state can determine its own exemption policy. Medical exemptions are uncontroversial – if you medically cannot get a vaccine, that is up to your doctor, not the state. However, some states have religious or personal (philosophical) exemptions. These have been under increasing scrutiny recently.

Other countries have different systems. In the UK, for example, there is no compulsory vaccination to attend school.

These policies are all being reexamined in the wake of the burgeoning anti-vaccine movement and the resulting return of previously controlled vaccine-preventable illnesses. Let’s take measles, for example. As you can see by the chart, measles cases have increased by greater than 20 fold from 2010 to 2019 in the US. But if we pull back even further, in 1980 measles caused 2.6 million deaths worldwide, with cases in the US alone measured in the hundreds of thousands. By 2000 routine use of the MMR vaccine had prevented an estimated 80 million cases of measles.

Also by 2000 measles had been eliminated in the US – that means there was no circulating virus in the wild. All cases came from outside the country. Outbreaks were also very limited, because there was no fertile ground for the virus to spread. The antivaccine movement has changed that.

The rate of children getting MMR has not changed much in the US due to compulsory vaccines, ranging from 90-93% over the last two decades. It has dipped to around 91.1% recently, but the overall number remains high. However, this statistic hides what is really going on. Vaccine refusal clusters, by neighborhood and by school. If a private school does not require compulsory vaccination, then anti-vaxxers will cluster there. These pockets of very low vaccination rates then serve as potential locations for outbreaks, and that is exactly what happens. In recent years these outbreaks have been large enough to spread into the general population.

Continue Reading »

A recent systematic review and meta-analysis of studies comparing humans to artificial intelligence (AI) in diagnosing radiographic images found:

Analysis of data from 14 studies comparing the performance of deep learning with humans in the same sample found that at best, deep learning algorithms can correctly detect disease in 87% of cases, compared to 86% achieved by health-care professionals.

This is the first time we have evidence that the performance of AI has ticked over that of humans. The authors also state, as is often the case, that we need more studies, we need real world validation, and we need more direct comparisons. But at the very least AI is in the ball park of human performance. If our experience with chess and Go are any guide, the AI algorithms will only get better from here. In fact, given the lag in research, publication, and then review time, it probably already is.

I think AI is poised to overtake humans in diagnosis more broadly, because this particular task is right in the sweet spot of deep learning algorithms. Also, it is very challenging for humans, who fall prey to a host of cognitive biases and heuristics that hamper optimal diagnosis. A lot of medical education is focused on correcting these biases, and replacing them with clinical decision-making that works better. But no clinician is perfect or without blind-spots. Even the most experience clinician also has to contend with an overwhelming amount of information.

There are a couple ways to approach diagnosis. The one in which human excel is the gestalt approach, which is essentially based on pattern recognition. With experience clinicians learn to recognize the overall pattern of a disease. This pattern may include signs or symptoms that are particularly predictive. Eventually the pieces just click into place automatically.

Continue Reading »

Acupuncture is defined as an intervention involving placing thin needles into specified acupuncture points in order to relieve symptoms or promote healing. What is special about the acupoints? They are supposed to be locations where the flow of life force (chi) can be manipulated. There are a few problems with the claims made for acupuncture. First, there is no place for vitalism – belief in a life force – in modern science. There is no evidence that it exists and no reason to hypothesize that it exists. It is worse than wrong. It is unnecessary. There is also no convincing evidence, after a century of research and thousands of studies, that acupuncture works for anything.

But also, an independently fatal flaw in the notion of acupuncture is that there is no evidence that acupuncture points exist. They are a complete fiction. They have no basis in anatomy, physiology, neuroscience, biochemistry, or empirical evidence. A recent study highlights this fact – Accuracy and Precision in Acupuncture Point Location: A Critical Systematic Review. This is a review by acupuncturists in the Journal of Acupuncture and Meridian Studies. The authors reviewed the literature for any studies that looked at the precision and reliability of the location of acupoints and found:

Considerable variation in localization of acupoints was reported among qualified medical acupuncturists. Variation in point location among qualified non-medical acupuncturists is unknown due to lack of any identified study. The directional method was found to be significantly inaccurate and imprecise in all studies that evaluated the method.

So acupuncturists cannot agree upon where alleged acupoints actually are on the body, and the primary method to localize them was both inaccurate and imprecise. The simplest explanation for this fact is that acupoints don’t exist. Even if you want to hold out the illogical conclusion that acupoints do exist, even though they have no basis in theory and they cannot be detected by any objective measure, you cannot avoid the conclusion based on this evidence that acupuncturists don’t know where they are. So how, then, can they claim to be able to stick needles into acupoints? Short answer – they can’t.

Continue Reading »

This is a great example of the unintended consequences that can result from making decisions based on scientific conjecture and preliminary hypotheses. In this case, the hypothesis was never a good one. If you are a pet owner, you may have noticed the recent trend toward grain-free dog or cat food. The justification for this trend is the notion that since dogs are essentially wolves, and wolves are pure carnivores, then we should not be feeding our dogs grains. This is basically the paleo diet for dogs.

It should be noted, however, that many carnivores do need to get some plant matter in their diet. They may get this from the stomachs of the herbivores they eat. Wolves will also occasionally eat berries as a minor supplement to their diet. But sure, wolves don’t generally eat grains. But that is not the problematic premise here.

The problem with the grain-free diet claim is that while modern dogs are closely related to wolves, they are not wild wolves. They evolutionarily split from wild wolves 15-40,000 years ago. Over that time their diet has shifted significantly. They no longer hunt in packs, but live off the scraps of human civilization. And they have adapted.

So the grain-free theory is flawed, and we should not base dietary recommendations on theory alone anyway, but actually gather evidence to test those theories. Perhaps we already have, in an unintended ecological experiment. In 2018 the FDA noticed a spike in reported cases of dilated cardiomyopathy (DCM) in dogs. Previous years had 1-3 reported cases per year, in 2018 there were 320, and 2019 is on track to exceed this. This is still a small number compared to the millions of pet dogs in the US, but is likely massive underreporting. There is also likely some reporting bias here – one article on social media might explain the spike, rather than a true increase. But the FDA had to investigate the increase to see if it were real and what the cause might be.

Continue Reading »

Vitamins are, by actual definition, essential nutrients for humans. They are substances we need but cannot make ourselves from other nutrients, and therefore must consume. Each of the vitamins have a disease resulting from a deficiency of that vitamin. For some there may also be negative health effects from having an insufficiency (less than optimal levels, but not low enough to cause a deficiency state). This is why you should eat your vegetables, and have an overall well-rounded diet – to make sure you get enough of all the vitamins and other needed nutrients.

But of course if there is a way to exploit a fact to make money, someone will do it. In this case an entire industry has emerged to make money by convincing the public they need to take vitamin pills, the more the better. In 2017 the supplement industry in the US reached $36.1 billion a year. Unfortunately, the path to better health is not so simple, and the science simply does not support the vitamin industry hype.

Yet another meta-analysis was just published again demonstrating that routine supplementation is mostly worthless, and may even be harmful. This one is focusing on vascular outcomes, but since heart disease is the #1 killer, and stroke the #3 killer, vascular risk has a major impact on life expectancy. In 2018 I wrote about a meta-analysis published in the Journal of the American College of Cardiology. This study found:

• Multivitamins, vitamins D, C, A, B6, E, calcium, β-carotene, zinc, iron, magnesium, and selenium had no benefit or harm for vascular disease or all-cause mortality.
• Folic acid and B-complex (Folic acid, B6 and B12) reduced stroke risk
• Antioxidants and niacin increased all-cause mortality.

Continue Reading »

Many outlets are covering the story of Mari Lopez, a YouTuber who claimed, along with her niece, Liz Johnson, that a raw vegan diet cured her breast cancer. Johnson recently updated the videos with a notice that Mari Lopez died of cancer in December 2017. She has refused, however, to take down the videos.

The story, unfortunately, is a common one. When people are diagnosed with cancer it is understandably a psychological shock. They face not only the grim possibility of their death, but also the prospect of surgery, chemotherapy, and/or radiation therapy – none of which are pleasant. It is a life-altering event. It also makes people vulnerable – in that situation, who wouldn’t want an escape hatch, a way to get back to their normal life? That is what cancer quackery offers – forget surgery and chemotherapy, just engage in this mild intervention, like a diet change, and your cancer will be gone.

Those who get lured in by this siren song experience one of several possibilities. First, many may still undergo some initial treatment, such as surgery, to treat the cancer, but then refuse adjunctive therapy like chemotherapy or radiation. Some of those patients may, in fact, be cured by the surgery. Often the purpose of the chemotherapy is to reduce the risk of recurrence. Others may forgo any science-based treatment.

Following the initial diagnosis, with or without some intervention, there is what we call the honeymoon period. In this phase the cancer may be relatively asymptomatic. This is especially true if it was discovered because of some screening test, like an X-ray or blood test, and not because it had already become symptomatic. This period can last for months and even years, depending on the stage and type of cancer. This is the phase where those who pursued some form of cancer quackery convince themselves that they are cured. They will also tend to credit the alternative treatment with their cure, even if they received surgery or some other standard treatment.

Continue Reading »

Last year a Chinese researcher, Dr. He Jiankui, announced that he had altered the germ line DNA of two babies using the relatively new and powerful gene-editing technique known as CRISPR. Dr. He is back in the news because of a new study looking at the effect of a mutation similar to the one Dr. He created on the life expectancy of those with the natural variant. The study finds that those who are homozygous for the gene variant (the delta 32 mutation of the CCR5 gene) have a 21% greater all cause mortality than those without the variant. What this means for the two children is unclear, but does raise concern.

Dr. He took it upon himself, without proper oversight or approval, to use CRISPR to alter the CCR5 (C-C chemokine receptor type 5) gene of embryos he then used for IVF (in-vitro fertilization) on his patient. The father who donated the sperm for fertilization (the patient’s husband) is HIV positive, so He sought to make a genetic change to the eggs to prevent HIV infection from the father. CCR5 is a protein on white blood cells that is used by HIV as an important gateway into the cell. Without it HIV infection becomes much less likely. There are other gateways, so it is not perfect immunity, but those with the naturally-occurring delta 32 mutation of CCR5 seem to be immune to HIV as a result.

He’s plan was to alter the CCR5 gene in the embryos in a way similar to, but not identical to, the delta 32 mutation. This was apparently successful in preventing HIV infection in the resulting babies. He announced what he had done after their live and apparently healthy birth.

He received widespread criticism for what he did for several legitimate reasons. First and foremost is his unsanctioned use of CRISPR on humans. He essentially conducted illegal human research. Human research is carefully regulated, with international standards, in order to protect the rights of people from harm and exploitation. He bypassed these regulations and was acting as a rogue researcher. That in and of itself is a career-ender.

Further, the specific application that He chose was not necessary. There are already effective proven treatments to minimize the chance of HIV infection from infected sperm used in IVF. Using an experimental treatment instead of a proven standard treatment is also considered unethical.

Continue Reading »

Some cultures routinely have a siesta after lunch. Is napping in the middle of the day good for you or bad? The short answer is – it depends. However a new study adds further evidence for a possible benefit to the mid-day nap, at least for elementary school children.

Let’s start with the concerns about napping, which has to do with “sleep hygiene.” Sleep hygiene refers to behaviors that optimize sleep quality, such as avoiding bright light late at night, going to bed with an empty stomach and bladder, and keeping a consistent schedule. One item on the good sleep hygiene list has been to avoid napping during the day. The problem with napping is that it makes it more difficult to fall asleep at night, which can result in a net loss of total sleep and sleep quality.

But it turns out the real answer is more nuanced. Some studies, such as this Spanish study in favor of siestas, show that a 30 minute light nap in the early afternoon is actually a net benefit, including improved memory, performance, and a reduction in stress. In 1995 NASA conducted a study and found that 26 minutes was the optimal time for their pilots to nap in order to optimize performance.

The features of a beneficial nap, therefore, include a limit on time to around half an hour. They also include (and this is probably a linked feature) lightly napping only – not falling into a deep sleep. The deep sleep is more likely to interfere with sleep onset at night. Also, the nap needs to be early in the afternoon, at least four hours prior to when you want to sleep at night. It should also be part of your routine schedule.

Continue Reading »