Artificial intelligence, grassroots rebels and the pharmaceutical cuckoo

By Jerome Burne

Theoretically, Artificial Intelligence (AI) applied to Evidence-Based Medicine (EBM) is a marriage made in digital heaven. AI’s algorithms, founded on reliable RCT evidence that’s updated as new findings come in, will whip through mountains of data, leading to improved diagnosis, more accurate screening and better testing protocols; all guaranteed to improve performance and cut costs. 

Mmmm, well, yes. EBM is certainly a good idea but the notion that it is ready for such a union involves large dollops of optimistic unicorn fantasy – to borrow a useful Brexit shorthand. And not just the obvious issues around garbage in leading to garbage out or the malign effects of human bias, which are regular worries about its implications for officially sanctioned treatments. 

AI could also have a damaging effect on efforts to develop new approaches to the big chronic killers – such as obesity, diabetes, cancer and Alzheimer’s. Ones which don’t just involve researching and prescribing an increasing number of drugs. Drugs, of course, have their place, but their massive financial muscle has given them the power of a giant cuckoo, able to shoulder all other solutions out of the medical nest on the grounds of lack evidence. 

Various grassroots revolutions by independent clinicians and expert patients have recently been having considerable success identifying the failings of official treatments and promoting the importance of lifestyle and other approaches. These include the treatment of low thyroid, chronic fatigue or ME, nutritional advice and mineral and vitamin supplements. Just bolting on AI in areas like these is likely to boost cuckoo power with little benefits for patients

The grassroots revolt over the prevention and treatment of diabetes – rejecting the long-running advice to follow a low-fat high carb diet and treating it by lowering carbohydrates and eating more fat – is one of these and shows why change is needed and the threat posed by AI. 

Can AI confront and sort out corrupt RCTs?

Recently FDA-approved an AI algorithm called DreaMed Advisor Pro which monitors patients’ glucose levels and then recommends the insulin dose they need. It would be based on the result of trials and likely to be more precise. So what’s not to like?

But what if you are a clinician or patient, unimpressed by the strategy of endlessly ramping up insulin on the grounds that it does nothing to treat the underlying disorder and besides, the results coming in for the low carb diet are consistently impressive? 

The app could be modified so that it takes the new data into account. But there are far more trials involving the insulin boosting approach; that’s what clinicians are familiar with, and besides who is going to pay for a modification to track eating less? The change in eating habits is often combined with group sessions and maybe other lifestyle interventions that don’t lend themselves to standard trials. 

Even without the tidying up that would come with AI, clinicians advocating the low carb approach to diabetes have already been hounded by committees of dieticians for giving dangerous advice, while experts in cardiovascular medicine, who challenge the wisdom of prescribing statins significantly to cut the risk of heart disease, have had their Wikipedia pages arbitrarily removed.

Medical orthodoxy cheerfully ignores EBM rule-breaking

Given the growing tendency to dismiss or block challenges to medical orthodoxy, modifying apps to reflect grassroots challenges seems unlikely. Easier to ignore them.

And ignoring evidence-based challenges to treatment based on a trial widely accepted to be high quality and a basis for Nice guidelines, is something the medical establishment is prepared to do. This is precisely what happened when a group of patients with a condition involving chronic fatigue, known as ME/CFS, re-analysed the data gathered in a trial that supported the orthodox treatment. 

Known as the PACE trial, published in 2011, it had reported that 22% of patients getting treatment- a combo of psychotherapy plus gradually increasing amounts of exercise – recovered, compared with just 7% of those who got no therapy. The re-analysis carried out by senior, independent clinicians in 2016, found that the results had been massaged to produce a favourable result. Without the fancy statistical manipulation, just 3% recovered and 7% showed some improvement. 

It’s a long-running and complicated story but it shows not only a readiness to play fast and loose with the data but also an ability by the psychiatric establishment to totally ignore a carefully conducted and devastating critique. The authors continued to assert that the treatment was beneficial and the Lancet – the journal that had published the PACE trial – ignored all calls to withdraw it or even to publish letters about it. 

SSRI drugs and playing fast and loose with evidence

There are also many examples of similar fiddling with the evidence from drug trials by psychiatrists, especially those involving antidepressant SSRIs and heavyweight tranquilizers known as antipsychotics. The risks were largely identified by a few courageous whistleblowing psychiatrists and researchers and fines of billions of dollars, sometimes, eventually followed, but none of this had the slightest effect on prescribing. In fact, SSRI prescriptions doubled over a ten-year period according to a recent report

An appalling case involved promotion of the antidepressant Prozac as safe enough to be used on children.  The profession is clearly intensely relaxed about disregarding EBM principles. AI that relied on official RCTs would improve nothing and close down whistle-blowers would show a reckless disregard for patients.

The on-going battle between ME/CFS patients and psychiatrist over evidence and treatment has similarities with the struggle some patients diagnosed with low levels of thyroid hormone have with the endocrinologists.  Both groups suffer from chronic fatigue but while ME/CFS patients don’t believe in psychotherapy they are offered  (with good reason) these thyroid patients know the drug treatment that may help (known as T3) but the orthodox position has been that few should be given it.  

Postcode lottery based on belief

Recently, following a remarkable 5000% hike in the price of T3, the confusion over the orthodox position has created a bizarre postcode lottery. Patients who have greatly improved on the drug can have it taken away because they live in a Local Health Authority area where consultants openly say they don’t ‘believe’ in it. (Read more). One driver for the ban has been as a cost-cutting exercise.

A roughly equal number of Authorities, however, have taken the view that if the drug might help, then it is worth a try and if it is already helping, taking patients off is not a clinically sound strategy. Drug or no-drug depends not on trials or evidence but on where you live. About as far from EBM as it is possible to get.

An estimated 10% of low thyroid patients (that’s around 300,000 people, mostly woman) are at the mercy of this lottery. The obvious solution, you might think would be to run RCT’s to see which patients do best – with T3, without T3. Some trials have been done and found not much difference, which patients claim – familiar story – doesn’t fit with their experience. They are now being supported by a group of senior clinicians who have identified serious shortcoming in the trials claiming no benefit.  

Ignoring failures of evidence has led to vaccine crisis

These are just a few examples that I have looked at in some detail where the bland reassurances about the reliability of EBM are highly misleading. Not only can treatments be based on shaky to non-existent evidence but when this is pointed out a favourite official response is to ignore it. This situation makes a mockery of current attempts to dismiss critiques of any official policy as the medical equivalent of fake news. 

This is at its most hysterical in the labelling of sites or people raising concerns or criticising vaccines an “antivaxers”. The implication that they object irrationally to all vaccines rather than having a specific and often technical concern. It might make sense if EBM was the norm and if results of drug trials were trustworthy – unfavourable findings not hidden or massaged. But they obviously aren’t. 

If, however, you pretend they are then you could certainly link vaccine data up with AI and use it to display the best schedules and doubtless link that in with households that weren’t complying. Scary. 

So far, we have been looking at existing medical areas where EBM and AI could well turn out to be a toxic pairing for patients and whistleblowers. But new research is throwing up some serious challenges to a couple of orthodox policies where AI could well be very useful; not as a way of enforcing the status quo but undermining it in ways that could prove very beneficial to patients. 

Healthy eating advice on the way out

The first involves dismissing broad-brush healthy eating advice – eat low fat, eat low carbs, for weight loss eat less and move more, calories in need to balance calories out. These slogans and others like them, don’t tell you anything useful about what you personally should eat because we are all different. 

That sounds very banal but the research it is based on, done by Professor Tim Spector at King’s College London, throws up some remarkable surprises and points in the direction of a far more effective nutritional science. 

Spector, a geneticist who has built up a large database of identical twins to tease out genetic links, ran nearly a thousand twin pairs through a detailed program of monitoring the way their bodies reacted to different foods in real-time. The first surprise was that each twin in a pair could have very different responses to the same foods – a chocolate brownie, rye bread or butter. In fact, a twin pair was barely more alike than any two people picked at random. 

The big and surprising implication of this is that genes can have almost no effect on whether your blood glucose stays flat soon after eating the likes of a chocolate brownie or shoots right up as we all expect. Spector himself found that while his glucose climbed rapidly after eating pasta, it barely moved after munching on grapes. His wife’s response, however, was the opposite.

So, personalised healthy eating advice should soon be much more available and accurate when the app Spector and his team are working on becomes available. AI will certainly be put to work organising the mountains of data already gathered – the monitoring also tracked the immediate effects on gut bacteria – but the potential for fudging the data for commercial ends seems much smaller. 

And there is another app in the pipeline that is set to change the way we think about drug safety and mineral and vitamin supplements. It’s based on just published research that really does seem a game-changer. It’s well known that supplements can interfere with the effects of a prescription drug. Your GP may have warned you about such a link. 

Prescription drugs need supplement support

What your GP won’t have told you is that the drugs you are being prescribed can affect your ability to absorb or use various minerals and vitamins effectively. In other words, if you are taking half a dozen drugs – not uncommon in older people – their collective effect could be to make you deficient in, say, vitamins B6 and 12, vitamin E, and the minerals zinc and magnesium. Over several years – this is very likely to have a damaging effect on your health.

Data on this effect is buried in the medical literature but almost no doctors know anything about it, except for a few familiar ones, such as the ability of metformin and the stomach acid-blocking drugs to cut the availability of B12, vitamin D and magnesium. I’m being deliberately vague because I’ve just written about it but the feature hasn’t been published yet. 

There are plenty of medical silos where EBM is a worthy aspiration rather than a fact. AI is the kind of disruptive technology that just could trigger a major upgrade, but the pharmaceutical cuckoo’s shoulders are awfully broad.

Jerome Burne

Jerome Burne

Jerome Burne is the editor of HealthInsightUK. He is an award-winning journalist who has been specialising in medicine and health for the last 10 years and now works mainly for the Daily Mail. His most recent book “The Hybrid Diet” was written with nutritionist Patrick Holford, published 2018. Award: 2015: Finalist for 'Blogger of the Year' Medical Journalists' Association.

6 Comments

  • It’s the psychiatric drugs with sensors in them that are going to be most dangerous. The sensor alerts the Dr. if you’ve missed a dose and when they make ‘taking your medicine’ mandatory as they’re now doing with vaccines what happens if you missed a ‘dose’ because of crippling side effects, severe adverse effects or the mental torture from being thrust into continual withdrawal when Dr.s abruptly stop one drug to start another? Or experiencing psychiatric drug induced depression, anxiety, insomnia or akathesia? How about drug induced sucidality or homicidality? What then. ‘Take your medicine’ they’ll say or we’ll force it. No one mentions the poor children, toddlers, babies & elderly drugged the most who have no voice. Psychiatry is nothing more the culling of humanity while big pharma laughs all the way to the Bank – all in the name of their pseudoscience.

  • The only thing I disagree with here, is the idea that Brexit has anything to do with the subject you discussed, or is even useful as an analogy.

    It is important to realise that AI underwent huge hype starting in 1980. Those of us who are old enough can remember endless articles about the massive job losses, danger of computers taking over control from us, the danger/benefits of using AI in defence systems, blah blah blah. By 1989, this hype was dying and those involved were embarrassedly trying to escape to other field. This is just an abstract from back then, but it gives you the flavour

    https://www.sciencedirect.com/science/article/abs/pii/S0736585384800313

    This is happening again. This time the hype about AI centred particularly on the idea of driverless cars, and it has taken a number of people to lose their lives for that enthusiasm to begin to curdle. Marrying this with Evidenced Based Medicine (as if doctors ever admitted to practising any other kind of medicine) will have one horrible consequence. Some people will be fooled (perhaps) for a while that medical science is coming out with something new, and try to follow it.

    Perhaps the only relevance of Brexit here, is that those in power with a lot to gain from maintaining the status quo will twist any argument to come to the conclusion that despite a nationwide vote to leave the EU, we should either remain, or dilute the effect of leaving to homeopathic levels.

    • Editorial

      David hi – first thanks for the background on AI and the link;very useful.
      Re Brexit: the reference to unicorns in the post had nothing to do with Brexit per se. It indicates wild optimistic-cum-magical thinking that has been applied to Brexit by its critics but it is perfectly legitimate to use it to describe almost any such project. I mentioned Brexit to clarify the possibly puzzling reference to unicorns, since it has frequently appeared in that context in the media recently – not to say anything specific about Brexit. I would, respectfully take issue with your analysis of the Remain position in the final paragraph.

  • Hi
    Re The dietary advice, I read the Tim Spector report, but I couldn’t see that insulin levels had been tested for, which I think would explain some of the differences between patients. I have just read Joseph Kraft’s book ‘Diabetes Epidemic and You’, published in 2008. Well worth a read. Kraft did over 14 thousand 5 hour assays of both insulin and glucose identifying distinct 5 patterns. The point has to be that the insulin response is affected by the history of carbohydrate intake before the standardised eating for the trial was implemented.

    We could definitely do with getting rid of the current Public Health England dietary advice which was based on very poor science and is arguably at the root of current chronic illness epidemics. But, having been brainwashed with the low fat mantra, the public do need some help to get on the right track.

    • Editorial

      Confess I didn’t read full report but they certainly measured glucose level which would give an indication of insulin, although i know that the two can diverge.Real tme monitoring interesting area that needs more research. Didn’t know about Kraft but dreadful that his work,done a decade ago, has had no impact on likes of the “experts” at PHE. Spector did track fat clearance rate in the lab but i understand there isnt yet monitors suitable for home use.He linked fat clearance with effect on gut flora. Slow clearance inhibits a class of bacteria that act as a kind of clean up squad.

Leave a Reply


WP-Backgrounds by InoPlugs Web Design and Juwelier Schönmann