Nukemen, slight return
Further evidence that it would only be possible to have a sensible debate about the role of nuclear in the UK's future energy strategy if we first got rid of nearly everyone currently working in nukemanship, on both the industry and policy sides.
It was always a certainty that, although completely non-replicable in the UK (both due to the design of reactors used, and the fact that we don't have earthquakes or tsunamis) - Fukushima would provoke an emotive response from people who don't understand about the topic but saw what radiation does in comics and are worried Dr Manhattan will blow up the world.
ReplyDeleteThe debate on UK nuclear should be on "is a new round of nuclear investment good value for money given peak oil and AGW, but also given massive historic cost overruns?", not on "OMG THE NUKES ARE GOING TO KILL US" - because the first is a sensible question, whereas the second is ignorant paranoia. For the government to try and shift public opinion on *any* topic away from ignorant paranoia is both remarkable and welcome, and to be encouraged.
A fine sentiment, of course, but I can't help but suspect that there's a hidden assumption here that when the debate is reconvened, after the righteous dispatching of the nukemen, that somehow the military viewpoint will be completely annihilated in the blast radius that obliterates the commercial interests.
ReplyDeleteFun fact: in the military, nuclear reactions are used in things other than warheads.
Government departments deciding to carry out PR campaigns that departed from the facts, and to make statements they couldn't possibly back up in the name of "controlling ignorant paranoia" is the same managerialist dickery that gave the world the MMR scare.
ReplyDeleteI'm afraid I'm a bit ignorant when it comes to immunization dumbfuckery, having never taken an interest in the dumbfuck side of things, so you'll have to spell that one out for me.
ReplyDeleteDan's suggesting that people's MMR fuckwittery arose as a response to overly reassuring government reassurances about vaccine safety (ie "the government is telling me vaccines are 100% safe, but they made my kid ill, so therefore I'll believe purveyors of fruitloopery rather than the government"), rather than as an example of government failure to control ignorant paranoia.
ReplyDeleteI disagree with him on this, but it's a reasonably mainstream/respectable view of how the crisis arose.
Rushing out an "it's all perfectly safe" campaign is the sort of thing that just makes people more paranoid.
ReplyDeleteThe nature of a radiation release is that it's very hard to quantify who's been harmed by it. It's also very persistent and impossible to clean up once it's got out of the bottle.
It's in the nature of humans to mislead their bosses about how well things are going, as uncovered in the Challenger disaster. That makes me skeptical of the "nothing can possibly go wrong" view; there's bound to be some other unforseen suprise which causes problems. The potential downside of those problems is the loss of hundreds of square miles of land. Can we afford that in the UK?
Can we afford that in the UK? Maybe not. But France is bigger.
ReplyDeleteRemember as well that the MMR bland reassurances of perfect safety came hard on the heels of similar campaigns with respect to BSE and salmonella, and so by this point the government had basically fucked its credibility. It's the attitude of mind that causes the problems rather than each specific instance.
ReplyDelete(Actually, I think if you want to make nuclear power "work" politically, you should take a lead from the naval uses and develop offshore nuclear. It would have the advantage of a guaranteed supply of cooling water and the ready ability to dump any failure into the sea. Nobody cares about the huge amount of nuclear material that went into the sea at Fukushima, and contaminating European waters to the extent of rendering fish inedible would probably be good for fish numbers on balance.)
ReplyDeleteWhat I remember about MMR was the way that the denials preceded any actual evidence. It was pretty obvious that their response would have been identical if there had been a problem. A stopped clock has to be right occasionally.
ReplyDeleteI think the mathematicians among my readership might be grateful for official confirmation that I don't intend to revisit the issue of whether a stopped clock has to be right occasionally.
ReplyDelete"BSE and salmonella" - both of which were scandals primarily featuring media paranoia rather than actual mass harm.
ReplyDeleteBut both of which were scandals in which government spokespeople began by saying that there was no problem and everything was fine, and then had to really embarrassingly walk this back when it became clearer that the scientists wouldn't back it up.
ReplyDeleteAlso, salmonella poisoning in eggs does cause actual mass harm.
I think your memory is at fault; the salmonella crisis actually started with a government spokesperson, Edwina Currie, saying "Most of the egg production in this country sadly is now infected with salmonella", and the scientists then refusing, 4 million dead chickens later, to back her up.
ReplyDeleteI'm with this pieceon eggs & salmonella. There was a problem; Currie talked it up rather than down - and generally, the job of governments is to talk problems down rather than up.
ReplyDeleteOverlapped with Ajay; he was right; my comment was a waste of time. GET A BLOG, MAN.
ReplyDeleteOffshore power also has the advantage of being tsunami-proof. Tsunami, like any other wave, pile up as they approach the shore; if you'd encountered the 11 March tsunami at sea, you wouldn't even have noticed it, because it would have been a swell with a wavelength of hundreds of metres and a peak-to-trough of maybe a few cm.
ReplyDeleteOn the downside, I suspect it would be a lot more expensive, simply because it costs a lot more to put something on a ship than to put it on the ground, because ground is there already and ships have to be built. The Russians are looking at it, but with reactors of less than 100MW which is tiny compared to an AGR.
On the gripping hand, it would be nice for the Clyde or the Humber yards to have contracts for fifteen or twenty very large power barges over the next five years.
So what is your policy recommendation?
ReplyDeleteI submit that "There's nothing wrong with it but please, go ahead and be terrified" or words to that effect is nothing anyone would mistake for a clear, convincing line-to-take and would have been universally reported as OFFICIAL: JABS WILL FUCK YOUR MORTGAGE, and further that "There's nothing wrong with it, but we must recognise your Very Real Concerns" is both the kind of thing you rightly despise in all other contexts and also pretty patronising in itself.
I think my policy recommendation is a) "don't pretend to have better evidence than you actually do" (note that global warming science has always been very honest from day one about the uncertainty and imprecision of their estimates, and have been massively successful), and b) DO NOT, ever, hang on to a clearly unsuccessful campaign. With both MMR and nukes, the government committed the cardinal sin - not being wrong, but staying wrong. The "nothing is wrong and therefore no other approaches need be tried" version gives you the psychological comfort of being able to blame the meeja and ignorant plebs for your failures, but has no other advantages.
ReplyDelete(It's worth noting that Ben Goldacre, among others, still thought that Wakefield's MMR investigations were good science as late as 2004. In such an environment, was it ever likely that "there's nothing at all wrong and only an ignorant idiot would ever think so" was going to work? All those Daily Mail (and Private Eye) papers were able to quote actual doctors. About a third of GPs thought that there might be something up with that vaccine, and people talk to each other. Maybe offering single vaccines wouldn't have helped, as they are more difficult to deliver. Maybe differently phrased advice wouldn't have either. But nobody ever even checked. You wouldn't launch a new brand of biscuits without testing, and you wouldn't launch a vaccination campaign in the third world without testing either).
On the nuke thing. My own take on it is not so much that I'm anti-nuke (I'm pretty neutral on it), but that in the UK (and for that matter the US) I don't trust the people running, or planning to build them. They're greedy, secretive and corrupt. Always have been, see little to change my opinion of that.
ReplyDeleteNor do I trust the government to regulate them. I mean to take one tiny example. Have there been serious studies done on the effects of global weirding on the location of these new nukes they're going to build?
Properly run nukes; well maybe then we can have a discussion.
I get the impression that the people here who are pro-nuke think my position is naive, or anti-science, or something. Am I wrong about that, or is there something flawed about my position?
I looked for evidence that Goldacre thought Wakefield's MMR investigations were "good science as late as 2004" but all I found was this piece from 2003, which doesn't really fit the bill—unless, I suppose, by "good science" you mean something like "wrong, but no compelling evidence for fraud".
ReplyDeleteBy 2004 there had been plenty of retrospective epidemiological studies that detected no link, e.g. Gillberg and Heijdel 1998 (Sweden); Taylor 1999 (London); Kaye 2000 (UK); Dales 2001 (California); Madsen 2002 (Denmark).
Also, in 2004 Brian Deer's first articles on Wakefield appeared, revealing that he had been paid to help with a lawsuit against the MMR manufacturers (though the full scale of these payments didn't come out until 2006).
So I think you may be misremembering the sequence of events.
Cian -- who is this "they" you describe as "greedy and corrupt"? Actual names and descriptions of banker-class lifestyles, please. Nuclear power in the UK (and most countries AFAICT) is a tedious industrial enterprise that pays typical energy-sector salaries and makes an unspectacular return on capital.
ReplyDeleteYes, what I meant was "most likely wrong, bit not fraudulent and so can't be completely ignored", which is what nearly everyone (in my opinion reasonably) thought until Brian Deer's articles started being published.
ReplyDeleteAnd it's the nature of epidemiology that it takes a long time to provide anything like definitive answers - think of how long it took to establish something as definite as the connection between secondhand smoking and lung cancer (or for that matter, think of one of Goldacre's favourite examples, hormone replacement therapy where the epidemiology literature ended up doing a 180). I don't think I have got the sequence of events wrong and indeed I think a lot of other people are very much reading back the 2004+ state of knowledge to justify courses of action taken 6 years earlier.
Even if we jad known in 1998 what we know now, though, it would still have been wrong to stick with a vaccination strategy that wasn't working. It was the dogmatism that was the really bad idea.
While I'd prefer not to feed the wet dreams of xenophobia with national stereotypes, I'd prefer to have the dirigiste-and-prepared-to-think-the-theoretical-through-to-practice French running my nukes rather than the throw-it-together-and-we-can-patch-it-as-we-go-along Brits. Nukes do not prosper in a bodgery environment.
ReplyDeleteAnd it attracts too many of the Brit mad-professor-wingnut-monomaniac stereotype as well.
To take one tiny example. Have there been serious studies done on the effects of global weirding on the location of these new nukes they're going to build?
ReplyDeleteYes. See the documents here and here.
"is there something flawed about my position?"
I'm wildly conjecturing here, but I suspect your position may not be wholly evidence-based.
To expand on my earlier remarks, it's worth knowing that:
ReplyDelete1) Even pre 1998, MMR had quite poor takeup compared to other commonly given vaccines - hanging around 90%, it was actually less popular than whooping cough. No obvious reason why; pre-MMR, uptake of the single shot measles vaccine was even worse. I think it's because the measles vaccines require two visits and are inconvenient; there may also be an effect from the popular (and for nourished children in the first world, not wholly incorrect) belief that unlike whooping cough, tetanus or tuberculosis, measles is a minor childhood ailment not worth bothering with.
2) Relative to that baseline, the drop in takeup was surprisingly small. When the whooping cough vaccine got a scare in the 70s, takeup declined to below 70%. The worst that MMR got was 80%.
3) There appear to be two problems in MMR takeup. The Wakefield-related one was a phenomenon of the upper and middle classes, and was about half offset by use of privately procured single-shot vaccines. Extremely interestingly, this was largely a word-of-mouth phenomenon; as far as researchers who have looked into this can tell, newspaper coverage largely followed the phenomenon, it didn't lead it. An anti-vaccination group of parents of autistic children had already got themselves organised enough to have meetings at Number 10 before the original 1998 Wakefield paper.
4) The middle class non-uptake phenomenon was at least partly associated with not particularly great levels of information about measles on the part of healthcare practitioners themselves. Plenty of GPs and nurse practitioners were suspicious of MMR, apparently; there were also suspicions that some GPs were compromised on their incentives if they had a sideline prescribing single shots. The approach of accusing parents of putting their children at risk was probably counterproductive as it just caused them to start ignoring public health messages. Nearly everyone who writes on this notes that a large problem was the generally low level of trust in public health communication as a result of vCJD.
5) More or less entirely separate from this, there was a problem of very low uptake in some London boroughs, which appears to have little to do with the media and everything to do with the general extreme difficulty of making the NHS work for poor patients with low resources.
6) The middle class uptake problem has been more or less completely reversed; vaccination rates are below the level needed to confer herd immunity but they always were. The London problem is still there, but progress is being made.
7) In NHS reviews of MMR strategy (nearly every PCT seems to have one online), the measures that consistently work are a) education of practitioners, and b) parent-led discussion groups. Occasionally c), the distribution of a DVD about a boy in Dublin who died of measles.
What this says to me is that:
1) This was a quite difficult problem of dealing with a viral (yes yes) phenomenon. Viral marketing isn't very well understood, and dealing with a PR response to this sort of problem is something that PR people are just beginning to learn about.
2) But the "Brick Wall Denial" is a PR strategy that is known to not work well in any situation at all, and that's the strategy that was followed from 1998 to about 2003/4.
3) Since 2004, lots of government departments and local healthcare trusts seem to have been doing a much better job, as evidenced in their results. Decentralisation seems to have worked here too.
[ctd]
ReplyDelete4) Blaming this all on the media isn't accurate as far as I can tell - in fact the Daily Mail has been pro-vaccination for quite a couple of years now. A lot of the problem appeared to be that the medical profession wasn't and isn't all that good at communicating the results of research to practitioners on the front line, a known problem that doctors are always setting up working parties to do something about. A lot of the rest of the problem was indeed one of communication - it failed to recognise that the majority of parents who were deciding against vaccination were in fact doing so because they had more information about medical matters than the median, not less, and were doing their best to make an informed decision for their child. The communication strategy involved treating them as if they well ill-informed and in many cases effectively accusing them of being bad parents.
But I'm not a marketing man (although I've studied it at business school and worked with some very good ones). So, since I am not a big fan of "detailed alternative propositions", my answer to Alex's question would be a) "something" (as opposed to a brick wall denial and persistence in a plan that wasn't working, and b), potentially "what was actually done, five years earlier".
"Yes, what I meant was "most likely wrong, bit not fraudulent and so can't be completely ignored", which is what nearly everyone (in my opinion reasonably) thought until Brian Deer's articles started being published.
ReplyDeleteAnd it's the nature of epidemiology that it takes a long time to provide anything like definitive answers"
Of course the Wakefield paper, even if not fraudulent, didn't really provide any evidence of an MMR-autism connection. It raised the hypothesis, but that is rather different. And that's part of the problem - a hypothesis based on pretty much no evidence is hard to disprove because you can't focus on the evidence presented (because there isn't any) and you haven't studied the hypothesis properly because there's no real reason to have done so. The proliferation of later negative studies can be seen as a massive waste of money.
You mention the autism-MMR link preceding Wakefield (indeed people seeking vaccine compensation provide some of the motivation behind his paper, the other part being his pre-existing obsession with measles and gut disorders) - I think that has its origins in the thiomersal (mercury preservative) - autism link (obviously the belief that vaccines cause various diseases is longstanding, and sometimes true).
a) "something" (as opposed to a brick wall denial and persistence in a plan that wasn't working, and b), potentially "what was actually done, five years earlier".
ReplyDeleteAn ad hominem campaign against Wakefield might have done the trick. It would probably have thrown up the Legal Aid funding much earlier - and it was that IIRC which really cut the heart out of the anti-vax narrative.
It't also worth noting that almost anyone having their children vaccinated in the late 1990s grew up in an era where hundreds of thousands of people got measles every year.
ReplyDeleteIt was only the institution of a second jab in the late 1980s which made the number of cases drop off a cliff. (I don't have the numbers to hand but IIRC you are seeing 100k cases sporadically throughout the 1980s, and about 10k in the early 1990s).
OTOH, the number of people who -die- after catching measles hasn't really changed since the 1950s - one in every few thousand (in fact, I believe that the ratio of deaths to cases is higher now than it was c.1960-90).
All of which means that it's hardly surprising that people felt that taking the chance of their kids getting measles wasn't the worst thing in the world.
a hypothesis based on pretty much no evidence is hard to disprove because you can't focus on the evidence presented (because there isn't any) and you haven't studied the hypothesis properly because there's no real reason to have done so. The proliferation of later negative studies can be seen as a massive waste of money
ReplyDeleteI don't agree with this; because vaccines are dangerous things (and sometimes they *do* get launched with serious side-effect problems that aren't noticed until they are in large-scale application) it's always worth doing big epidemiological studies on them - after all, this is why the data was being captured which allowed the MMR/gut/autism studies refuting Wakefield to be done.
A lot of the problem was that a lot of GPs don't keep up to date with research, as I say, and so (I would guess) they also were made aware of the MMR issues via the Daily Telegraph and Channel 4. Which is not to blame the GPs per se - my brother is a surgeon and so it was naturally to him that I turned for advice, and despite the fact that he reads the journals diligently, he wasn't in a state of knowledge where he could honestly reassure me sufficiently that there was nothing going on (and so it was that my first two kids got single-dose vaccines for their first shots; by the time second shots came around it was post 2004 and they got MMR). What was really needed here was a government campaign targeted on doctors (health warning: I am now breaking my own edict and just monday-morning quarterbacking, but there you go, as long as nobody mistakes it for anything other than recreation no harm will be done).
All of which means that it's hardly surprising that people felt that taking the chance of their kids getting measles wasn't the worst thing in the world
ReplyDeleteYes I think that's right - this might be why some of the PCTs found that DVD of the kid who died such a useful marketing tool. Which of course is IMO the worst kind of shame-n-blame paternalism and completely goes against my underlying thesis, but since this is a science blog, I posted it anyway.
"I don't agree with this; because vaccines are dangerous things (and sometimes they *do* get launched with serious side-effect problems that aren't noticed until they are in large-scale application) it's always worth doing big epidemiological studies on them - after all, this is why the data was being captured which allowed the MMR/gut/autism studies refuting Wakefield to be done.
ReplyDeleteI don't disagree with the value of post-licensing surveillance - but the point I'm making is that there was such epidemiological data available, but it didn't look explicitly at the MMR-autism link (and why would it), so it turned out to be of little comfort in countering what was basically a piece of speculation by Wakefield. Medics and scientists in general are far from averse to such things, but are normally kept in check by their peers, in this case it seemed to take on a life of its own.
"A lot of the problem was that a lot of GPs don't keep up to date with research, as I say, and so (I would guess) they also were made aware of the MMR issues via the Daily Telegraph and Channel 4. Which is not to blame the GPs per se - my brother is a surgeon and so it was naturally to him that I turned for advice, and despite the fact that he reads the journals diligently, he wasn't in a state of knowledge where he could honestly reassure me sufficiently that there was nothing going on"
GPs are regularly confronted by bollocks medical stories from the Daily Mail so they should, in theory, have been quite capable of dealing with this the same way they deal with similar things every other day of the year. The paper itself is transparently useless. Of course no one could say that there definitely wasn't an MMR-autism connection because they didn't have any evidence that there wasn't (but no reason to believe that there was) - this sort of open and honest response to the public (and to the sort of morons who do interviews on the Today programme and Newsnight) is easily twisted into saying 'well, there must be something to it'.
I don't know where you get your idea about many GPs being suspicious - the wikipedia article (I know) suggests 2-3% believing in an autism-MMR link (too many, but you can generally get 5-10% of people to agree to anything in a survey).
Also, claiming that the media followed rather than lead the concern is an interesting hypothesis - what is your evidence for that? Looking at, for instance, this figure it seems that MMR uptake dropped precipitously in just the period where the media furore was at its height. Now I'm sure much of the media coverage was driven by parental concerns (in the same way that much Diana coverage is driven by public interest) but that doesn't mean that there is not a reciprocal and causal role of the media.
Most of the relative decline in uptake also appear to have occurred during the early stages of the controversy when media attention was relatively low
ReplyDeleteIf you look at that figure (it's clearer in the similar figures in the paper linked above), you can see that 2001/2 was the peak year for media coverage, by which time nearly all of the peak-to-trough decline had happened. When you allow for the fact that there's a big lag in the uptake rate data (a child goes into the file upon reaching his or her second birthday with or without being vaccinated - obviously this can lag the actual decision by a substantial amount of time).
On GPs approval or otherwise, I'm sure WPis right on the proportion of actual anti-vaxers among British GPs, but, eg this study shows that about a third of non-MMR parents thought their GP "agreed" or "wouldn't mind" about their decision (a lot of these would have been going down the single vaccine route).
In general, I think my view on this can be summarised by David Ogilvy's maxim "The customer is not an idiot; she is your wife". Mrs Digest is not an unintelligent woman, not a believer in homepathy, not uninformed and not a reader of the Daily Mail. And yet, we got single vaccines for kids 1 and 2; I don't know what we would have done if this hadn't been an easy copout due to having the cash but I am not going to accept 100% of the blame for what I still believe was a rational decision based on the information available at the time, and I don't think it's the media's fault either.
According to those figures (thanks for the link) the peak of MMR distrust was 2002 with the trough of uptake in 2yr olds 2003 (fitting with a trough of uptake later in 2001 to later in 2002).
ReplyDeleteThe media coverage ramped up early 2001 and peaked early 2002 but there was obviously a fair bit of coverage (including front page splashes) in 1998 and it continued through 1999 and 2000 including a further claim of a link by Wakefield in 2000.
The sharp fall off in MMR rates starts in 2000 (fitting with the original publication date in 1998). So, while you can't blame volume of media coverage alone, I don't think we can say that this data shows it didn't contribute significantly - in particular high media coverage rates correspond to the peak of MMR distrust and the original media coverage of Wakefield seems the most likely origin of the initial drop-off in MMR vaccination. That media coverage quantity lags the drops in MMR coverage does not mean that media coverage did not have a causal role because, as other studies have suggested, the coverage was highly partial and failed to highlight the speculative nature of Wakefield's study or the evidence and medical/scientific consensus against it.
As to your own choices, that is entirely up to you, but I'm sure you're aware of the pressures of cognitive dissonance. I think I've argued with you before that your claim that early on in the controversy it was not clear whether there was an MMR-autism link or not (and thus that there was a scientific, as well as media controversy) is false when you look at the essentially anecdotal level of evidence presented by Wakefield et al.
Given how many speculative claims are made by scientists and medics in papers and at press conferences every day you have to wonder why this one took off? (A similar, popular finding was the Kirsch et al antidepressant meta-analysis which really caught the imagination of the media and the public - but was rather less obviously speculative than Wakefield's work).
This is interesting:
ReplyDelete"Publicity about the alleged dangers of the combined measles, mumps and rubella (MMR) vaccine may have damaged the immunisation programme in Wales.
Researchers have studied the percentage of children taken for an MMR jab in the principality and found a drop in the numbers coming forward.
Fears about the safety of the jab were raised in February following the publication of a study by doctors at the Royal Free Hospital in London. It suggested a link between autism and a newly identified bowel disease found in some children after having the combined vaccine."
So even in 1998 they were talking about adverse media publicity re: MMR. The difference may be that by 2001/2 it had reached epic proportions (certainly every tom-dick-and mel was writing about their ill-informed views by this point).
You've picked up regional variation there; the Swansea Evening Post had a massive campaign, very early on, and that is indeed the health authority that was driving the figures. But even in the case of Swansea, it's not at all obvious that the media was driving rather than following the story. And even if it was - isn't influencing the media story part of public relations? Journalists on the SEP, Daily Telegraph, etc, didn't pick up on the MMR story out of pure cackling evil, and many of them were not actually innumerate; there were plenty of specialist health correspondents on the wrong side of the story. If the NHS wasn't convincing them, is it any surprise it wasn't convincing anyone else? The main reason why brick-wall denial is such a bad PR strategy is that it loses control of the media narrative.
ReplyDelete"But even in the case of Swansea, it's not at all obvious that the media was driving rather than following the story"
ReplyDeleteSo Swansea had a grassroots anti-MMR campaign that was simply reflected in the media coverage? That sounds pretty implausible to me (for more references to concern about media coverage in a non-Welsh context try here or here).
"If the NHS wasn't convincing them, is it any surprise it wasn't convincing anyone else? The main reason why brick-wall denial is such a bad PR strategy is that it loses control of the media narrative."
As yorksranter above intimates - it isn't particularly clear what an alternative strategy would be. The government reiterated that (a) there was no real evidence MMR caused autism, (b) evidence suggesting it didn't and was safe, and (c) that they couldn't totally rule out that there was a connection but there was no reason to believe that there was one. They later argued that (d) further evidence suggests that it is highly unlikely that there is a connection.
Since the government adopted an open an honest response that was also, in hindsight, true, I'm not clear what you would prefer. You say: "don't pretend to have better evidence than you actually do" (note that global warming science has always been very honest from day one about the uncertainty and imprecision of their estimates, and have been massively successful), and b) DO NOT, ever, hang on to a clearly unsuccessful campaign.".
I think you are just wrong on the level of evidence question, and on the question of campaign tactics, apart from advocating the provision of single vaccines I don't really see any alternative strategy. If all your point comes down to is that single vaccines should have been offered, I think that is a pretty narrow (and debatable) criticism of the government approach. I also wonder how many other scares could have been usefully handled by basically saying 'you're right, fuck the evidence, better safe than sorry'?
PJ, three points:
ReplyDelete1) You and Alex both seem to be regarding communication here as nothing other than a man in a suit doing a piece to camera. As I've discussed above, there was a lot of work on training GPs and on communicating parent-to-parent that could have been done. You can bet that when BP has an oil spill, the crisis PRs don't consider their job to have been done when they've drafted a press release.
2) This is a problem of risk communication and frankly, your four points look like a bit of a mess to me - I can tell from other posts that you understand the underlying issues, but your a) b) c) and d) look like half of them are repeating the same point and half of them are contradicting each other. And the thing is that your four points probably actually reflect an improvement on the standard of communication actually used at the time. Which relied hugely on the word "safe", which in context is basically a weasel word and is always a terribly misleading and confusing way of communicating risks. It has the one advantage of being considered to not give away any legal hostages to fortune, but it nearly always miscommunicates, and gives a strong impression of not taking the problem seriously.
3) As BenC identifies above, the government strategy was also based on some seriously, and IMO probably intentionally, misleading risk communication about the dangers of measles.
The government did not communicate remotely accurately about the relative risks of MMR vaccine and measles, on the basis of the evidence it had at the time; it was hung up on the useless word "safe" and it exaggerated the dangers of measles (it also talked far too much about herd immunity without explaining the concept). I don't accept that this view can be summarised as "basically saying fuck the evidence".
And my fundamental point is still - they took far too long to learn that what they were doing wasn't working. By 2004/5 the NHS had begun to get a sensible communication strategy and was reaping the benefits of having done so. I think it's not coincidental that this was when the original Wakefield parents' group lost legal aid and dropped their case and strongly suspect that before this point, the brick-wall denial strategy was being influenced too much by litigation risk.
I don't see any contradiction or repetition in pj's first three points:
ReplyDeletea) there is no evidence for X
b) there is some evidence for not-X
c) the first two points are not strong enough to rule out X completely
d) is a later revision of b): there is *good* evidence for not-X (although c) stands).
I'm not sure if they should have been saying anything different (setting aside the question of whether they should have been saying it different*ly*).
In a) I don't understand what the word "real" is doing there; whether the evidence for the gut-autism connection was real or not is just the same question as whether there was a gut-autism connection.
ReplyDeleteIn b) I am not at all sure what "evidence suggesting [...] it was safe" is meant to refer to, and it has that problem word "safe"[1]
In c), "there is no reason to believe in one" has the same annoying ambiguity as "safe". There is no reason to believe in an invisible spaghetti monster; there is no reason to believe that measles vaccine causes autism via gut inflammation - clearly these two propositions for which there is "no reason to believe" are not on the same logical footing.
You are right that you can work out something like your summary if you sit down, think about it and extend a modicum of interpretative good faith, but this is meant to be public communications. I would have gone for
"Some very small case studies have claimed that MMR can cause gut complications, and as yet untested theoretical claims have been made that gut disease is related to autism. However, long term studies of MMR vaccine do not show a measurable increase in autism itself among infected children. Even though the risk of measles is itself small, we can say with a high degree of confidence that your child has a much greater danger of being harmed by measles if unvaccinated".
I think that's something that could have been honestly asserted at any time in the program, and it has the advantage of being specific about risk language, not using weasel words and not talking down to people. (For the reasons in the RHUL study which showed that vaccination refusal was led by informed customers, I am not at home to "people wouldn't understand that).
But my main point is that however well-drafted, a talking point to be read out by a middle-aged man in a suit was never an acceptable media strategy.
--
[1] To expand: "safe" in context and to someone talking rigorously means "posing an acceptably low level of risk compared to the alternatives" (a stuntman might consider his rocket bike to be 'safe'). But talking loosely and out of context, it means "no risk". It's a bad word to use.
"And my fundamental point is still - they took far too long to learn that what they were doing wasn't working. By 2004/5 the NHS had begun to get a sensible communication strategy and was reaping the benefits of having done so."
ReplyDeleteOf course trust in the MMR had started to recover by then suggesting something was working beforehand. How did the strategy change in 2004/5, and how much of that change was influenced by the increasing availability of studies undermining the MMR-autism link?
(also, as well as "people wouldn't understand that", not at home to "but that's basically what they did say". Nuance matters here - it's marketing! - and the point about my version (as well as being written in a style meant to pastiche the actual risk disclosures the NHS uses these days) is that it doesn't try to pretend that the Wakefield study didn't exist, is specific about what kind of evidence it is using rather than blustering "all the evidence" and doesn't exaggerate measles. I am still not happy with 1) using "measurable" as a substitute for "statistically significant", which I think I can justify but is still talking down to people a bit and 2) using "degree of confidence" in a way that fairly invites the question of whether it's just as weasel as "safe".
ReplyDeleteHow did the strategy change in 2004/5
ReplyDeleteI think it might have started to change earlier than that (the paper on Brighton that I linked above was based on work carried out in March 2004, so there were clearly initiatives to be assessed; I can't find any documents on MMR communication strategies dated earlier than that but there might have been some).
Certainly you're right though that a number of the key studies came out in 2003, which supports the RHUL thesis that the whole thing was being driven by educated consumers. But note that 2003 was (ex the Q1 02 peak) actually nearly the biggest year for sustained media coverage of MMR.
(Did you see, btw, there's an article on the Guardian today where James Randerson picks a few nits with respect to one of Ben Goldacre's articles!)
I think there's also a certain amount of political background which is important to the whole thing - obviously there was a new government in 1998 which had never handled a major health scare before.
ReplyDeleteThis government was made up of ministers who almost to a man had never been in office before, and who went native as soon as they had run through their top-line manifesto items. So while a Tory government might have learned from salmonella, vCJD, etc. Labour defaulted to the blanket denial. Qv. also FMD and, to go full circle, nuclear power.
it also talked far too much about herd immunity without explaining the concept
ReplyDeleteIt didn't help that "herd immunity" is a spectacularly patronising phrase (would it have killed them to say "community immunity" or similar?), and isn't really that easy to market as a concept - i.e. "you should vaccinate so that other people can skimp".
As for the scare-video, it's striking that they needed to go to Ireland to find a suitable case...
"(Did you see, btw, there's an article on the Guardian today where James Randerson picks a few nits with respect to one of Ben Goldacre's articles!)"
ReplyDeleteHeh, yeah, although it is actually an academic article of Goldacre's he's critiquing. Slightly ironic that most scientific claims are reported quite uncritically (or flat out misleadingly) while ones criticising journalists are subject to a decent level of scrutiny. I doubt you can conclude much from a study that only looks at thirty odd articles but Goldacre is right that Randerson sets up a massive straw man: "James Randerson’s main argument is that we are foolish in wanting to apply the WCRF grading criteria for evidence on health advice to ALL science and engineering stories."
Still, good to see a bit of reflexive practice on the part of science journalists for a change, and Goldacre was definitely due some of the attention he has been dishing out.
"whether the evidence for the gut-autism connection was real or not is just the same question as whether there was a gut-autism connection."
ReplyDeleteAs I've said before, I think we differ over whether the Wakefield paper counts as 'evidence for...' rather than 'a hypothesis that...'. And certainly there is a massive distinction between whether there is evidence for X and whether X is true.
"I am not at all sure what "evidence suggesting [...] it was safe" is meant to refer to, and it has that problem word "safe"[1]"
It is meant to say that there was epidemiological/surveillance evidence of no association between MMR-autism (but this was not overwhelming earlier on because no one had looked specifically at the proposed link). In this context 'safe' means of very low risk and refers to surveillance evidence of all risks associated with MMR (not just autism). You may not like the word 'safe' but what would you replace it with? 'There is evidence it poses no risk of autism and no evidence it poses a risk of autism'? That's probably more confusing for a public we already know has massive trouble understanding risk. You may "not [be] at home to "people wouldn't understand that"" but I bet they wouldn't. I also think you want to acknowledge that there is evidence of risk, however small, while government rightly didn't want to acknowledge any risk (because of the lack of evidence that there was any, and because it would likely be seized upon as an admission of risk - cf. the Blairs vaccinating their offspring).
"clearly these two propositions for which there is "no reason to believe" are not on the same logical footing."
No, but neither are propositions for which the evidence would be considered sufficient to make any policy changes about. In the same way, many people have 'evidence' that the moon landings were faked or that homeopathy treats malaria.
So you actually do have an articulated critique! Kiss! You see, I had the impression it was just a slightly cynical bit-of-contrarian-for-the-dads. But the point about the real audience being the GPs is a genuinely good one.
ReplyDeleteAlso (as you know Bob) I have a very low opinion of GPs in general...
"But the point about the real audience being the GPs is a genuinely good one"
ReplyDeleteDo we believe that better targeting of GPs (who did receive advice from government, and who should have been able to assess the evidence themselves, and the vast majority of whom, apparently, didn't believe in an MMR-autism link) would have had a material effect on the controversy?
I don't know, but I don't get the impression that GPs were necessarily in the loop when these decisions were being made by families (to book for a jab I don't think you needed to go via a GP then, as now).
to book for a jab I don't think you needed to go via a GP then, as now
ReplyDelete... which means that the main touch point was either the community midwife or the practice nurse, and plenty of them believe a lot of very odd things (we had one that had an absolute fixation with squirting breast milk into newborn's eyes). The Brighton and the RHUL studies did suggest that a majority of MMR-refusing parents did have contact with the GP - they were middle class after all - though.
With respect to the BG/Randerson contretemps, I think that the "all science and engineering stories" element of Randerson's critique is a pretty silly strawman, but he does land two body blows, in that 1) ten newspapers for a week isn't actually all that massive a sample, and the fact that only two stories came from the Guardian and 31 from the Express suggests that the sample was quite likely to be unrepresentative, and 2) if the researchers aren't trying to say that stories with "flimsy" evidence shouldn't be published at all, then it is a big weakness that they didn't identify the context or caveats.
ReplyDeletebitheway:
ReplyDeletethe vast majority of whom, apparently, didn't believe in an MMR-autism link
there's a big gap between "having a correct belief" and "being able to make a convincing case for your correct belief to a sceptical party".
"the fact that only two stories came from the Guardian and 31 from the Express suggests that the sample was quite likely to be unrepresentative"
ReplyDeleteI think that was the number of 'health claims' - so quite possibly representative but a very small number of overall articles.
"... which means that the main touch point was either the community midwife or the practice nurse, and plenty of them believe a lot of very odd things (we had one that had an absolute fixation with squirting breast milk into newborn's eyes)."
ReplyDeleteOh totally, but I'm not sure how you change their minds (there was a similar issue with allied health professionals and swine flu vaccination) - these aren't strongly evidence based beliefs (milk into eyes, WTF?)
"The Brighton and the RHUL studies did suggest that a majority of MMR-refusing parents did have contact with the GP - they were middle class after all - though."
The RHUL one doesn't have any data on that.
milk into eyes, WTF?
ReplyDeleteFor conjunctivitus/ crusty eyes - not just for the hell of it I assume?
It's a thing. Works reasonably well, saves on hospital trips, makes parents feel involved. (Anecdotally: is sometimes used to persuade wavering mothers to breast-feed).
http://tropej.oxfordjournals.org/content/28/1/35.extract
"So Swansea had a grassroots anti-MMR campaign that was simply reflected in the media coverage? That sounds pretty implausible to me."
ReplyDeleteIts not really, local papers are a lot easier to get coverage in for any group. When the group concerned is 'concerned parents' then all that has to happen is one of the parents has the confidence to be able to pick up a phone.
The local media are always after these stories, Swansea evening post even more so given the standard type of stories in that newspaper are of the level of 'local man writes to MP complaining about bad traffic management'.
"It's a thing. Works reasonably well"
ReplyDeleteReally? I can't access that paper (on prophylaxis with colostrum in newborns, not treatment or using normal breast milk) but the methodology doesn't look massively sound - and I can't find any other studies. I can't say I'd be happy with mothers treating neonatal conjunctivitis with breast milk.
Grr, shouldn't start arguments when really busy.
ReplyDeleteJohn, from memory (seriously don't have time to check right now), those particular studies used figures for sea level rise which were seen as out of date at the time (the Dutch were using higher figures, for example, for their own contingency plans), and it was also criticised for not considering other factors relating to expected changes (including, if I'm remembering correctly, the Met). Not everyone thought that, but a number of people (not enviro-types, but scientists) seemed to think they had deliberately chosen to go with conservative data to make it seem better than it was.
Plucky Underdog: So you're saying that the only possible form of corruption is financial, and if the sums involved aren't vast, it doesn't happen. Well it's an argument I suppose.
Cian -- I'm genuinely curious as to what you mean, but so far you've chosen not to say. Sorry, I'm a morally obtuse INTJ, so it's not as obvious to me as it clearly is to you. Is it the connection to nuclear weaponry?
ReplyDelete