Hard News: An interview with Ben Goldacre
57 Responses
First ←Older Page 1 2 3 Newer→ Last
-
"I’m not aware of any longitudinal studies that would make a fair comparison over time to say if they’ve got better."
We can do a quick and dirty look at google trends for news stories over time, comparing "miracle cure" with a baseline of the more boring "parliamentary debate".
Against the baseline of frequency of parliamentary debate news, neither better nor worse.
-
Moz, in reply to
Interesting plot, that one.
And thanks Russell and Ben for the fascinating interview/article.
Also, I think that the further you get from seeing the meat hit the grinder the easier it is to forget that in the end that's what you're supposed to be working on. It's absolutely essential to have the "how best should science communicators work to convey the summary of research into a possible solution to a problem identified by epidemiologists working for a committee of political appointees" people working, but they both depend on the rest of the chain, and their ultimate goal is still "less meat into the grinder".
At the extreme I think it's really useful when the political decision makers have to face the people whose lives they effect and say face to face "it's better that you die than these other people I have identified". My fear is that too often one pretty, stricken figure gets the media attention so the easy political solution is to let 100 poor, elderly people freeze to death away from the media eye so that the politician can fund treatment for that one person. Which is a whole other part of the "is this social science or politics" question :)
-
Russell Brown, in reply to
My fear is that too often one pretty, stricken figure gets the media attention
Which is what we see when drug companies push for Pharmac funding by taking those cases to a willing media.
-
Bart Janssen, in reply to
My fear is that too often one pretty, stricken figure gets the media attention so the easy political solution is to let 100 poor, elderly people freeze to death away from the media eye so that the politician can fund treatment for that one person.
That was the reason Pharmac was kept outside the political system. Up until the point that National decided to buy a voting group by selling a Pharmac decision down the river.
These decisions are one of the reasons I work on plants. I'm just not emotionally equipped to make those calls, even though you can argue that improving the food supply is important it just isn't that immediate.
It's also the reason I defend so strongly those people who are willing to sit in a committee and try and make those very calls. Yes they will get it wrong sometimes but shit that job is a tough one and anyone willing to take it on deserves respect for their effort.
-
Bart Janssen, in reply to
Which is what we see when drug companies push for Pharmac funding by taking those cases to a willing media.
This too.
If I was king (a really bad idea) drug companies would be banned from making any media statement or advertising at all.
-
David Hood, in reply to
drug companies would be banned from making any media statement or advertising at all.
At which point they get round both by doing what they do now- funding patient advocacy groups to call for funding. So "donating to a charity".
-
Thanks, Russell. What a pity he's not coming to Wellington. We ethics-nerds would love to hear him.
Interesting what he says about the anti-vaxxers. I think Wakefield is still very much alive for many people and not talking about him becomes another conspiracy.
-
Moz,
Hmm, I'm a fan, but I don't know that I'm $90 a fan (Sydney talk). Will have to ask the lady of the finances for permission.
Asked a kinda-friend if she wants to go, made myself not say "the social ethics aspect may be too pointed for you" because friend has very much sold out and is making good money managing forced council amalgamations. Those are being done for a combination of long-term ideology, and short-term "administrators we appoint will support the motorway we're forcing through"... most affected councils have court cases in progress against said motorway. But the pay is excellent!
-
Bart Janssen, in reply to
At which point they get round both by doing what they do now- funding patient advocacy groups to call for funding.
We're talking about me being King here so basically I'd just rule that treason and chop their heads off - I did mention me being king was a really bad idea.
-
Stephen R, in reply to
It's also the reason I defend so strongly those people who are willing to sit in a committee and try and make those very calls. Yes they will get it wrong sometimes but shit that job is a tough one and anyone willing to take it on deserves respect for their effort.
My (second hand) understanding is that Pharmac has a way of classifying medicines on a scale that roughly comes out as $/quality life year, and they try to rank all the stuff they could fund in order, and then they work down the list until they run out of money.
If pharma companies want to improve their chances of having their drug funded, they can come to the party on price (moving their $/QLY up) or release drug studies that show that it's more effective than they thought.
From what I hear, one of the things that's getting up some noses at Pharmac are the companies who don't release all their study results, just the ones that make them look good...
-
B Jones, in reply to
Interesting what he says about the anti-vaxxers.
I had a conversation once with someone expressing a view opposing vaccines (she wasn't anti-vaccine, but...), who cited Ben Goldacre in support of her position. "You can't trust anything researchers say, they're all employed by big pharma, Ben Goldacre says how they hide all the data," etc. It wasn't how I thought the conversation was going to go.
-
Bart Janssen, in reply to
From what I hear, one of the things that’s getting up some noses at Pharmac are the companies who don’t release all their study results, just the ones that make them look good…
Which is exactly the point Dr Goldacre is making.
Pretty much every scientist hates the way pharma (big and middle sized) play games with data. Same is true with almost every big industry - do you really believe the environmental impact numbers from Toyota or VW?
But we (the public) have all collectively decided that we are happy to allow free market capitalism to manage these industries. Which works well for some things but when it comes to health it looks increasingly dodgy.
But the alternative is to use public funds to discover, and more significantly, develop new medicines - then we can apply the rules we (society) want. Or we go head to head with pharma and demand they adhere to our rules about data - anyone want to guess how well that conversation is going to go.
But despite the gloom about data - things are starting to change - there are a lot more open source journals with open data. There are efforts to improve reproduciblity standards. And even in the Pharma industry there are noises that what was done before is not really acceptable any more. Things are getting a bit better.
All of that would be dramatically helped by an increase in the general population's understanding of scientific method ... if the public was more willing and able to be critical of claims maybe we could get past the paternalistic "trust me" and into the more inclusive "here's why we think it works, what do you all think?"
-
Funny, a lot of what he says could be applied to other aspects of living, life's, and umm... other things. good I/V
driven in the public eye by trust rather than evidence.
we are so grotesquely under-using the information we already have
people in very elevated social positions who seem to have forgotten that their principal role is serving the public
-
Russell Brown, in reply to
One thing I knew there wouldn’t be time for in the interview was a discussion about the study in 2008 that found the real trial data – published and unpublished – on SSRIs was underwhelming.
That got promptly written up in the papers (and shouted from some blogs) as the drugs don’t work (example from the Herald, which led with that angle).
The reality of course is that as I wrote at the time it’s a bit more complicated than that.
-
Fascinating interview! Enjoyed it very much.
I have Ben Goldacre's I think you'll find it's a bit more complicated than that on my bedside table. He makes the same point about assorted British quacks in the book as he does about Andrew Wakefield in your interview - that it's not really about individuals, it's about how the systems around them allow them to operate unchallenged. Like Ken bloody Ring and his radio appearances here. -
Grant Taylor, in reply to
The reality of course is that as I wrote at the time it’s a bit more complicated than that.
Yes. Actually, that applies to nearly everything covered in the interview.
Ben’s right, I think, in most of the points he makes but reality is simply way more complicated than the evidence-based medicine/healthcare database (even if were full utilised), research methodologies and budgets can get around currently – possibly ever, but who knows what might be possible as AI develops.
His views of how things ought to be are actually pretty simplistic; conversely, things are not really as bad as they might seem in light of his valid criticisms. For example, the researched medicines industry is not without self-serving bias and frank corruption – but I am very grateful for researched medicines and for the billions that Big Pharma has invested in developing treatments, which although not perfect and probably not as effective as they are ‘sold’ to be, are a damned sight more efficacious and safe than so-called “alternative” medicines. Cognitive-behaviour therapy may have been shown to be as effective as SSRIs in a range of studies of questionable ecological validity – but SSRIs are a great deal cheaper, more accessible and socially acceptable than CBT, which is why they are more often prescribed.
Also: we have no option but to “trust the experts” – in most aspects of an increasingly technology-driven world. Life would be untenable otherwise.
I have written a little myself on these questions elsewhere, if anyone is interested.
-
We've been trusting experts since we let electricians put wires in our houses. But electrician skills don't have the same sort of social and economic cachet and scarcity that scientific expertise has, so there isn't an alternative electrical industry*. You can train as an electrician in a relatively short space of time, compared with research or medicine. A lot of the criticism of science comes from a place of (justified or unjustified) suspicion of the power that people who wield it have. Socially marginalised groups have trouble with it, but so do social elites who think that they're just as entitled to wield that power as those who have worked for it.
*also, staff turnover would be pretty high.
-
Russell Brown, in reply to
We’ve been trusting experts since we let electricians put wires in our houses.
And in most cases, we should naturally continue to do so.
Given a follow-up, I’d have put it to Ben that the same might apply to him: on some level we trust him because he’s witty, articulate and convincing. None of which are guarantees that he is right.
-
Bart Janssen, in reply to
Also: we have no option but to “trust the experts”
Yes and No
I am very much aware of the mountain of knowledge I've had to absorb to become an "expert" in my very narrow field. I know that mountain is just as high in every field. So no we can't all have that knowledge and we do need some trust.
But you don't have to be an expert to ask the basic questions. Was the data reproducible (or is this just one experiment in one narrow system)? Was the data peer-reviewed (or is this a press release or worse a pay-to-publish journal)? Do the authors acknowledge the rest of the field (or do they claim special insight)?
These are standard things that indicate bullshit. The really frustrating thing is the media should be asking all these things BEFORE they go to press. If our media modeled this kind of critical questioning then I think the public would get used to doing it as well.
Every scientist I know can cite examples of science news stories where the "evidence" was obviously questionable.
So yeah if you want an expert on fluorescent bugs you talk to Siouxsie but most times you don't need Siouxsie to recognise that the well and good section of Stuff is mostly unsupported bollocks.
-
One feature that does make Goldacre more trustworthy is that he's generally quick to acknowledge where evidence countering his own position exists. Bad Science is in large part a series of case studies illustrating how pseudoscientists behave when their views are challenged; unsurprisingly, Goldacre actively tries not to be that person!
-
Nothing about water fluoridation
-
Russell Brown, in reply to
But you don’t have to be an expert to ask the basic questions.
This is true. Towards the end of Hard News’s life as a radio rant in (script posted to the internet) I committed to sussing out the science on GMOs and I did get to the point where I felt I understood basically what was going on even if I didn’t get every word. (i.e.: about where I’d aspire to be on a marae). I certainly started to see the patterns.
But it did take a degree of commitment. I think there’s still a role for journalists, it’s just that journalists need to be better.
The ongoing credulity over the “meth house” panic doesn’t entirely fill me with optimism though. The eagerness to ascribe authority to businessmen who bought testing kits pretty much fits Dr Ben's observations.
-
Grant Taylor, in reply to
These are standard things that indicate bullshit
Yes and no. Bullshit comes in shades of grey. Most current knowledge is more or less bullshit and, importantly, there is a difference between bullshit and bullshitters. I am always on the look out for bullshitters (and as Russell has indicated, this is a social judgement not an intellectual one; the “truthiness” concept of a few years ago comes to mind) but I am prepared to believe the word of an expert non-bullshitter as the best access to the truth I am likely to have, even if that word is almost certainly to some extent bullshit*.
The challenge is to figure out who we believe when our ability may be limited to even properly parse and understand the answers to the questions you suggest, Bart. You are right that it is reasonable to ask such questions but it may be foolish to believe that (a) we actually know what are right questions to ask in the first place (incidentally, asking the wrong questions is one of the things that creates a tremendous amount of bad science), (b) we are able to comprehend the answers we are given, and (c) we can accurately judge to what extent these answers are incorrect.
There ain’t no getting past trust, IMO. Trust is what the world runs on.
*PS: One of the Deans of Auckland Medical School many years ago (I think it was Cole) addressed a year 1 class saying that 50% of what they were about to learn in the next five years will be wrong - but we wouldn't know which 50% for another ten years).
-
Rich of Observationz, in reply to
there isn't an alternative electrical industry
Not many homeopathic software debuggers either.
-
Rich of Observationz, in reply to
Or, you can try learning about stuff from basic principles. Want to know about (e.g.) whether fluoridation is a good idea?
Learn some basic chemistry, like why different compounds of the same elements can have very different properties. Learn some biology around how substances get metabolised and used by the body. Understand some maths so you can deal with quantitative stuff like concentrations and doses. Then get on to more advanced things like the mechanisms of action, epidemiological data and the differences between outcomes and proxies (Ben Goldacre’s books will help you here). Once you’ve done that, you can read papers, look up citations and understand whether the presented findings are likely to be of significance.
No need to take things on trust if you can develop the right filters.
Post your response…
This topic is closed.