Dr John Campbell, the BBC, and Ivermectin

 

Dr John Campbell is a retired nurse who has built up a considerable Youtube following with his almost daily videos on the evolving coronavirus pandemic. His approach is ‘evidence-based’; he is pro-vaccine, but also pro-ivermectin, an effective antiparasitic which is being actively investigated for ‘off-label’ use in three principal settings relating to Covid: ‘prophylactic’ (preventative), ‘early-stage’ (keeping mild disease mild) and ‘late-stage’ (helping sick patients avoid the worst outcomes).

Ivermectin use in these settings is, to say the least, controversial. A highly vocal group of physicians is convinced by the evidence to date – in all 3 settings – while others, particularly academics, consider the evidence gathered to date to be weak – many small trials with methodological flaws and growing evidence of outright fraud. The demarcation between ‘physician’ and ‘academic’ is not a hard one – many such academics have medical qualifications, and many physicians have a good grounding in the sometimes esoteric world of trial evaluation and composite analysis. The pro-ivermectin lobby have chosen to enlist the assistance of the lay public in their fight to get this drug available for clinical use, and it has become highly politicised. Amateur advocates on both sides (including, let me be perfectly clear, me!) have developed opinions on this scientific controversy. However, this lay public is often unaware of the details of specialism – the possession of the title ‘MD’ or equivalent is no guarantee of expertise in the complexities of clinical trial conduct and evaluation. It requires a good grasp of statistics, and an awareness of the bewildering number of possible sources of bias.

The BBC ran an article highlighting the work of a group of independent researchers who have been casting a critical eye over the research to date, which had given some eyebrow-raisingly positive outcomes. Eyebrow-raising particularly in view of the fact that other research has found no effect. If a treatment really were 90% effective, one would expect the bulk of trials to show a positive outcome, rather than a few outliers. We also need to be aware of publication bias – the tendency not to publish null results – to know whether the published picture accurately reflects the underlying reality.    

And so to Dr Campbell. He released a video on his channel eviscerating the BBC and the researchers. Because this is a video medium, tone and visual cues become significant.  His curmudgeonliness is part of his appeal, but also, saying ‘make up your own minds’ while gurning furiously adds a layer that belies the objectivity suggested by the bare words!

1)      0:54 Heavy sarcasm because the BBC article that ‘debunks ivermectin’ was written by two mere journalists. However, those journalists are simply reporting what the scientists they mention – Jack Lawrence, Gideon Meyerowitz-Katz, Dr James Heathers, Dr Nick Brown and Dr Kyle Sheldrick - have found. They are not themselves offering a personal opinion on efficacy or the quality of the research. He lays it on with a trowel: you wouldn’t be keen to have a journalist operate on you – but likewise, I would not be keen to be operated on by a retired nurse! Point being: credential mongering is a fun game that anyone can join in on! (My own credentials? Biochemistry graduate, 2 years postgrad research, IT professional. I'm nobody.).

2)      2:45 “Is it true that many campaigners are anti vaccine activists? They give no evidence for that” Yes. Of course it is. It is astonishing and worrying that Dr Campbell claims to be unaware of this. His own comment sections are an absolute hotbed of anti vaccine activism, with the same people frequently promoting ivermectin. Such comments garner hundreds of ‘likes’. He cites himself, pro-vaccine, as a refutation, but it is not a statement that can be refuted simply by bringing out a ‘black swan’. Scroll down to the comments below his video. I’ll give you a couple of  minutes to find half a dozen clear pro-ivermectin anti vaxxers. Don't read slowly just to prove me wrong!

3)      3:25  “Even advocating formulations used for animals...”. He hints that this is a deliberate attempt to smear the drug, but it is entirely true. There is a substantial market in veterinary ivermectin. They aren’t wrong to mention this on the grounds it might smear ivm, they are absolutely right to mention it to discourage the practice.

4)      3:50 “Most of the BBC’s conclusions appear to have been based on this study here" This is a Nature Medicine correspondence by the individuals named in the article, calling for greater transparency in the availability of supporting data for clinical trials. It does not actually have much to do with the investigations of the group into specific clinical trials, and Campbell is wrong to say this is what the BBC piece is based on.

5)      4:10 “well no actually, he‘s a student there”. As a bald statement, this is unexceptionable, since it is true. He is a postgraduate student. But Campbell delivers this with considerable theatre – bait and switch, meaningful look to camera, look away with pursed lips, pause for effect. The audience is clearly being invited to dismiss everything subsequent on these grounds alone. Even though only a nonverbal moment, I feel compelled to delve into this in detail, because - even if only by implication - telling a lay audience that students can’t do good work is damaging, and rather bad form.

a.      Students publish all the time! Even undergraduates. They are often doing original work. My own 3^rd year project, using HPLC to measure adenosine levels in blood, could have been written up had it occurred to me to do so – several years later, I saw a paper describing exactly my method!

b.     Every distinguished academic was a student once. We don’t dismiss their early work because they were ‘just a student’ then.

c.      John leaves himself wide open to being dismissed as ‘just a nurse’. I would not dream of it of course; it would be academic snobbery and poor academic etiquette...

d.   (Edit). It was not written by Jack and rubber-stamped by the rest. Nick Brown, one of the letter's authors (see comments) has emphasised to me the Author Contribution statement: "All authors contributed equally to the writing of this manuscript". This is where one goes for information on apportionment. Or, failing that, the Corresponding Author. Don't just guess.

6)      5:30 “Who debunks the debunkers?” Likewise, who debunks the debunkers of the debunkers? Who monitors YouTube experts for quality and accuracy?

7)      6:20 “The 26 studies they critique are not in this article”. No, they aren’t, because the Nature piece isn’t the supporting evidence. It’s not even mentioned in the BBC piece, and John has just decided on his own initiative that this is what it’s based on, because it happens to be by the same people. A reasonable criticism of the BBC article might be that it does not provide a reference that would prevent people barking up the wrong tree, but then, this is quite common in journalism. How many newspaper articles are referenced? It is not a scientific paper. If John wanted to know more about the underlying work, he could, as a matter of due diligence, have contacted the BBC authors. But he seems to prefer Google.

8)      7:38 “This international group of scientists is a Twitter group”. I see nothing wrong with that. Twitter and WhatsApp are part of the means by which scientists in different continents communicate (I so want to add ‘grandad’!). Collaboration does not require physical proximity – particularly when data driven. Also, it’s ironic coming from an influential YouTuber to, seemingly, denigrate people over their choice of medium. This segment is terminated by the odd disclaimer ‘nothing wrong with that”, which is rather at odds with the well-poisoning tone of what goes before!

9)      8:35 “They give no references. It’s pathetic. I did some digging around...”. I actually agree that we do need to see more of this evidence. But again, it’s not a scientific paper in itself. If more info is required, John has already noted that it’s a Twitter group. So they have an online presence which, if he cared to, he could use to gain information. The Nature piece has a Corresponding Author. Again, this seems a failure of due diligence to me.

10)   9:05 “A few studies are wrong, but the rest are presumably OK”, and “Some of the data in this paper might be wrong, but the rest of it’s presumably OK” (both paraphrases). This is rather surprising. If there are signs of fraud, it really should raise a massive red flag. But Dr Campbell seems to want to paper over the cracks.

11)   14:50 “The BBC is quoting unpublished evidence”. I’d certainly like to see publication myself, and I would concede that the disparate means by which this has been disseminated don’t look good at first glance. But, it is work in progress. And note that a lot of the studies critiqued, that often form central pillars of meta analyses, are themselves ‘unpublished’ preprints. One reason to put things onto preprint servers is indeed to solicit review, under which rubric this work would fall. It would be absurd to insist that a critique of an unreviewed preprint should itself require formal publication. I think Dr Campbell is misrepresenting the scientific method to a lay audience, here.

12)   16:10 “Why would they reference a paper that has been retracted?” This is an extraordinary complaint. It is an example of a paper that has been shown (by this group) to be highly suspect, if not fraudulent. They got it retracted. Why would they not reference it?

13)   17:20 “I’ve been reading meta analyses for about 30 years now”. I have no reason to doubt Dr Campbell on this, but I would imagine it uncommon that nurses (or even doctors) spend much time on the nuts and bolts of meta analysis. It’s a highly technical area and somewhat specialised. Abstract, Discussion, that’s usually it.

14)   18:20 “They are saying all existing meta analyses without IPD should be considered  at high risk of bias”. He may have simply misspoke, but it is the individual studies that are being argued as at high risk of bias, not the meta analysis itself. The biases afflicting meta analyses are distinct from those in trials. 

15)   19:00 Having poured scorn on the piece, he then agrees it is a ‘brilliant idea’!

16)   20:04 “Presumably we need to rip up every meta analysis ever...” This is a hyperbolic and obtuse reading of the recommendation. No change to practice calls for the complete abandonment of all prior work, and this one is no different. 

17)   20:40 “Trials are conducted by people who are not professional researchers”. That is no excuse. If the work on which advocacy is based has elevated potential for methodological flaws, then it should be regarded as less influential for public policy than ‘gold standard work’ with large amounts of money behind it. It may seem unfair, but it also seems reasonable.

18)   21:30 'ivmmeta'. This is where things really fall off the cliff. I don’t think a reputable scientist should associate themselves with this site. I say this as someone with no particular expertise in this area; even I can see how bad it is! It is anonymous, not peer reviewed, not ‘published’ in exactly the sense he criticises these researchers for, and operates on a cryptic algorithm unavailable for examination. Its inclusion criteria are also dubious, and lack rigour – no assessment of bias, for example, a critical part of conventional meta analyses. “To avoid bias in the selection of studies, we include all studies ...” Conventional, published meta analyses assess each study on a ‘risk of bias’ metric, usually semi-formal, to avoid giving undue weight to methodologically suspect work. Yet what they appear to do is examine papers for the effect that puts ivm in the best light, and add it to the pot – even if that outcome was not the registered intent of the trial! That is a truly awful way to go about things. It means that random noise, if positive, can add weight when the overall result is neutral. Their supposedly unbiased ‘all-in’ approach is heavily biased, in favour of something, anything, that shows ivermectin having an effect. 

There is also a ‘lumping’ issue – some studies simply do not belong in the same bucket, others mix therapies. For example, a paper promoting a 4-drug ‘IDEA’ therapy (Ivermectin, Dexamethasone, Enoxaparin, Aspirin) adds its weight simply by virtue of mentioning Ivermectin – even if the effect is entirely due to (say) Dexamethasone, and IVM is merely piggybacking.

The extraordinary benefits in all 3 clinical settings, for multiple end points, reported by ivmmeta are not replicated by any other meta analysis - except that produced by an ivermectin advocacy group, which rated the fraudulent Elgazzar paper as at low risk of bias! They refuse to re-analyse pending a response from Elgazzar, still awaited 3 months on. Dr Andrew Hill, on the other hand, withdrew his previously favourable meta analysis and revised his opinion based on the fraud uncovered by the group in the BBC article. Roman et al conclude "IVM [...] is not a viable option to treat patients with COVID-19. The Cochrane Library, regarded internationally as a 'gold standard' for trial evaluation, likewise report no benefit.

Dr Campbell mentions none of these; his championing of ivmmeta in their stead is disappointing.

So, in sum, while I agree with Dr Campbell that we need more meat on this, I don’t share his eagerness to dismiss this group out of hand. The red flags in the Elgazzar paper, causing its retraction (even from ivmmeta!), were noticed by ‘just-a-student’ Jack Lawrence.  Ivmmeta has also dropped the Samaha paper, another target of this group, while Niaee is reportedly being withdrawn from publication, and Carvallo has been seriously called into doubt. So while it’s certainly possible they have nothing to back up their stronger claims, I think it would be hubristic to be dismissive of them, just on their results to date.  Something is clearly up. Why is there any fraud at all? Could it be that enthusiasm for the treatment has got the better of people? Everyone should be concerned about fraud. If ivermectin works, the fraud devalues legitimate research. If it doesn't, unjustified belief in it has probably cost lives. Either way, it undermines science.  

Listen to two of the group, Gideon Meyerowitz-Katz (epidemiologist) and Kyle Sheldrick (medical doctor) discussing their ‘sleuthing’ here. It's long, but full of detail on exactly why they are making the bold claims that they are. Worth 90 minutes of anyone's time!