Post No.: 0530
Any ‘health guru’ who wants to make a name for him/herself must ideally come up with some kind of proprietary USP or original piece of insight. So while passing genuinely healthy nutritional advice such as ‘eat a wide variety of fruit and vegetables’ won’t get you in the commercial or media spotlight – coming up with something like an extreme or odd diet or ‘detox’ regime will… at least until the market one day gets bored of a particular trend/fad I suppose.
In science, most researchers personally want impactful and flashy findings that’ll grab the headlines over something that has more substance. (This might even lead a few to commit research fraud.) An extra layer of hype or the over-extrapolation of findings is often added too.
The media constantly craves novelty too – something new, original, unusual, interesting and hype-able to report on. Some of the most tried-and-tested, reliable and sensible advice isn’t headline-worthy, whereas something extreme, odd or controversial that sounds plausible has a far better chance of being picked up on; even though it might not be conclusive, completely safe or relatively affordable.
As a result – exorbitant ingredients imported from some remote island, pretentious procedures inspired by some ancient practice, and the like, can make living healthily seem far more complicated, expensive and inaccessible than it really is. So blame this industry for making health messages seem complicated and for making eating healthily seem impossible on a tight budget!
Marketers often want to make things seem more difficult than they really are so that they can justify selling an overpriced solution (e.g. hydration is ‘hard’ because plain water apparently isn’t enough for most people most of the time, hence an expensive drink is what consumers ‘need’).
Due to commercial interests being their primary motive – it’s ultimately about them making money more than about improving your health. The interests of a business and the interests of their customers are thus not always (long-term) aligned. What’s best for the shareholders isn’t always best for the consumers’ interests in health, or it certainly isn’t always best for their purses.
The motivation for profits and the provision of evidence for advertised claims don’t always align either. This includes many, although not all, of the PR claims made by the complementary and alternative medicine (CAM) industry (that provides treatments that fall outside of conventional healthcare), the ‘wellness and clean eating’ industry and naturopaths, as well as occasionally the big pharmaceutical companies themselves. Journalists and the mainstream media, and consumers of social media, are regularly being misled by dubious claims. There are sometimes outright lies about scientific studies even existing! Post No.: 0332 revealed more dodgy scientific practices.
Most drug trials are funded by and conducted by the drug companies themselves, but they’re of course biased in wanting to show positive results for their own products. This is why such experiments must ideally be replicated by independent researchers to those original experiments – by scientists who don’t have a stake in the results one way or another.
Replicating studies isn’t as glamorous a task as producing original studies but is utterly vital for science. Like those ‘health gurus’, any scientist who wants to make a big name for him/herself must come up with some original research that teaches us something novel and makes a major contribution to their field. But scientists who replicate other people’s research to independently verify them are the unsung stars. You won’t make a name for yourself by repeating someone else’s research, most scientific journals have less interest in publishing repeated studies because they’re not original, and funding agencies are also less interested in funding them – yet being able to reproduce research is a major cornerstone of the scientific process. Woof!
A ≥95% confidence that a result wasn’t down to statistical chance means that there’s a <5% possibility that the result was down to mere chance – or up to 1 in 20, even faultlessly conducted, studies will have results that are down to just pure chance (and not every study is faultlessly conducted). But if an experiment is repeated, the chances of a false result occurring twice in a row will be 5% of 5% (assuming that independent datasets or samples from the original trial are used) i.e. 0.25% or 1 in 400. And so forth with further replications of a study. This therefore massively improves the reliably of the findings if the findings concur with the original study.
Note that peer reviewers for scientific papers don’t really check whether the data gathered in a study or experiment is truthful or accurate (e.g. they don’t double-check to see if respondent A really gave the response he/she did). So an academic paper being published by even a reputable scientific journal is not a verification of a study’s findings – it’s just a verification of a study’s editorial quality and that it meets certain scientific standards (e.g. that the study is well-designed, its claims logically accord with the apparent evidence provided, it acknowledges and builds upon other work in its field). This means that peer-reviewed work isn’t necessarily correct or conclusive – it just means that it meets certain criteria to be able to be called academically-worthy science.
Scientific knowledge is cumulative because it builds upon the work of others, thus people citing and relying on the unreliable findings of others is hugely problematic. It’s like a whole pillar of what we thought we understood can crumble because the foundation was iffy. Reliability and trust in science is generated by peers critiquing and figuring out ways to solve problems in each other’s work, incorporating and building upon each other’s work, and replicating each other’s work.
And this is why replication is so critical – if someone else follows the exact same methodology and gets essentially the same results then this double-checks the results of the original study.
Drug trials usually take a long time and are very expensive however, hence most drug trials wouldn’t be carried out at all if it weren’t for being funded by the pharmaceutical companies that also manufacture and sell the products. This means that we must take their results with our own level of caution or trust, or desperation or luxury to wait until independent studies are conducted (e.g. an observational study of the efficacy and side-effects of a treatment after it has hit the market).
There’s generally a lot of research fraud in science, or at least a gross lack of reproducibility of the results of many studies for whatever reason(s). Scientists often cannot even replicate the results of their own original studies(!) This could be due to the (perceived) pressure on scientists to ‘publish or perish’ in order to bring attention onto themselves in their fields, which leads to failing to adhere to good scientific practices. Publication bias could also be a factor. New research from around the world is being published at an unprecedented rate, and if science becomes more about profits/seeking funding or making the headlines rather than simply the search for the truth then there’s a fundamental problem.
In some fields, such as the social sciences, there is, as of posting, a reproducibility crisis. The fanfare that an original study receives may mean that its conclusions become the lasting one, in both pop culture and serious academia, even when its results cannot seem to be replicated. But science is supposed to be self-correcting – not full of dogmas that cannot be overturned. Examples of experiments that struggle under replication even include big ones like the Stanford Prison Experiment, the Stanley Milgram experiments on obedience to authority, social priming, the facial feedback effect, and the marshmallow test. Their conclusions might not necessarily be completely disproved but have at least been overstated. That’s why we must never stop learning and questioning what we thought we knew (myself included).
Some partly blame a lack of reproducibility in the biomedical sciences on machine learning algorithms that directly look for patterns in big data, where a lot of these patterns are only spurious artefacts in the data rather than in the real world. (If you really want to find any pattern, you’ll very likely find at least a pattern out of chance alone – just like when looking at clouds for a while, if you’re looking to find any shape, you’ll eventually find at least a perceived shape eventually, such as a face, figure or fluffy beast of some kind.) So better statistical tests are probably needed to assess the chances of chance in machine-learning-found patterns.
When the science isn’t on their side with any positive evidence to support their claims – parties with self-interests to protect may instead try to attack the scientifically-backed opposition via a strategy of casting doubt. Nothing in science is ever absolutely 100% certain (not even whether the Sun will rise tomorrow), hence theories can never be proved once and for all – only falsified once and for all. So such parties will attempt to cherry-pick, focus on and drive a wedge into the <5% or <0.001% of uncertainty in order to try to widen it. That means that we can practically always cast some doubt on anything.
But if we’re rational, we should follow an argument that is supported with a ≥95% confidence rather than an argument with a <5% confidence. (I would also encourage people to study scientific courses rather than just news or social media coverage if people really want to understand a subject, like climate change, from a scientific perspective too.)
Claimed experts or even some scientists or professors themselves may not be impartial to tell the complete truth – due to commercial contracts, sponsorships, gifts or money they may have with, or received from, certain companies, which therefore creates a (hidden or undisclosed) conflict of interest. The pharmaceutical industry spends far more on sales and marketing (e.g. marketing to doctors who prescribe drugs to patients) than on actual research and development! So don’t believe it when they say they must make huge profits in order to cover their R&D costs – it’s more to cover their invasive and pervasive marketing, to ultimately maximise their profits.
Consumers themselves generally over-trust the marketing claims of products that’ll allegedly treat or cure what ails them. So drug companies would rather market directly to consumers than to more knowledgeable doctors (in places where this is allowed). Or drug companies will sponsor advocacy groups and exploit the media to serve their own campaigns to try to get their own drugs funded by a country’s national health service – they might make it seem like the national health service is the villain for not funding their questionable-efficacy, over-hyped and overpriced treatments; and the media and public might fall for this, with cries of ‘why is the NHS withholding this drug from us?!’ They might promote the public misunderstanding of science to their own advantage. It’s like if there really was a sports drink that was marketed to say it’s ‘got what plants crave’ because ‘it’s got electrolytes’ – at least some people would believe it because it sounds like science(!)
Companies sometimes hire doctors to promote their products just so that they’re able to say ‘recommended by doctors’ in their marketing. But what is that doctor a doctor of? There are so many different branches of medicine. Someone can also be called a doctor if they have a PhD in any field – including a totally unrelated field! Professors have their specific areas of expertise too, even if they’re broadly in the relevant field. Therefore such titles don’t imply expertise in all areas and we must ascertain what area(s) a doctor or professor works in, rather than automatically assume that ‘he/she has got a fancy title hence we must always defer to his/her opinion’.
Woof! In the end, you’ll need to say something original or at least unusual if you want to improve the chances of making a big name for yourself within an industry. However, extraordinary claims require extraordinary evidence.