In Praise of Observational Evidence
asteriskmag.com
asteriskmag.com
This can be more subtly critical than it might seem, in that even if you can manipulate some proxy, often that proxy is insufficient in actually representing the phenomenon of interest, or the conditions under which they actually occur.
I often use the example of videogames and aggression. There were plenty of experimental studies of this but it was always questionable whether lab-induced anger is the same thing as, say, the sort of violence we generally are concerned about societally.
I generally have tried to teach students that experimental designs when done right provide powerful causal evidence of something, but often with limited generalizability; observational designs in contrast provide powerful generalizable evidence of some kind of association, but often with limited certainty about the causal pathways involved.
I've been in a department that was rabidly experimental in its focus and it always seemed sort of short-sighted, because people were idolizing RCTs with proxy manipulations that had questionable generalizability to the real-world phenomena they were trying to model.
Ideally you'd bring both experimental and observational evidence to bear on a question. Your conclusions should be robust to different types of designs.
OP suggests that alternative methods like target trial emulation, propensity scoring and double machine learning can be used to approximate the conditions of an RCT using existing data. In saying so he gives away the tell, which is that RCTs are the standard being aspired to.
Observational data may well be undervalued. But RCTs are still the gold standard.
_Parachute use to prevent death and major trauma related to gravitational challenge: systematic review of randomised controlled trials_
https://pmc.ncbi.nlm.nih.gov/articles/PMC300808/
"Advocates of evidence based medicine have criticised the adoption of interventions evaluated by using only observational data. We think that everyone might benefit if the most radical protagonists of evidence based medicine organised and participated in a double blind, randomised, placebo controlled, crossover trial of the parachute."
Nobody reasonable suggests that RCTs must be used for every problem. When the effect size is enormous, or there's obviously no alternative (as in the parachute example), it's silly to suggest the use of an experiment.
On the other hand, if you want to compare some subtle difference between two competing parachute designs, for example, then perhaps an RCT would actually be appropriate. This is much closer to what is actually being measured in real clinical trials.
[1] To be completely fair to the authors, it was intended as a joke, not an argument. The real joke ended up being that people took it seriously.
We could be much more flexible in our approach to things if we would un-ban things. For example after phase 1 trial or based on sufficient observational evidence, things can no longer be banned but have a higher standard for "insurance is now required to cover it".
Don't forget that RCTs are very far from perfect and issues -- sometimes literally fatal issues -- have later turned up via observational evidence in large cohorts. Vioxx, for instance. Many others.
I believe, without the tiniest shred of doubt, that the only trials drugs need to go through are initial safety/toxicity trials (phases 0/1) and that everything else would be much better left to access+observation.
Should we demand an RCT before we accept evidence? Of course not. At some point you do have to make a choice on things.
And it should be noted that most drugs do have early cutoff criteria if the evidence is strong enough that it is working. It isn't like people are wanting to withhold good treatments from the world. Adding controls and randomizing them, though, has proven to be highly effective at helping progress.
If you have enough data, you can smooth out individual fluctuations due to things like drug interactions, non-compliance, etc. (And indeed you might discover drug interactions!) Observational trials ultimately mirror how drugs are used in the real world.
> "Adding controls and randomizing them, though, has proven to be highly effective at helping progress."
I would argue just the opposite. Demands for increasingly byzantine trials have ballooned the costs associated with drug development, and have slowed things to a crawl. There's a reason the field's golden age was in the 1940s and 1950s, and it's not just "low hanging fruit." Today nobody in their right mind wants to work in drug development when they could work in tech or even finance.
Again, it is off to think that one is automatically superior to the other. Certainly to the exclusion of the other. And that is what feels off with the framing of the parent post. I am perfectly fine saying you should use both observational and controlled trials. But I think it is also wrong to think you don't have to build experiments to test interventions.
This is why you put metrics in your service code. So that you can observe them behave and look for things to change. This is also why you do test cases on your code, so that you can specifically target your change.
Now, I fully back the idea that just A/B testing something doesn't automatically mean you learn something true. But neither does observing a strong outcome on uncontrolled data.
"Large controlled experiments are costly and can hurt people who opt-in to informed consent. Instead, we should do significantly, significantly larger experiments, with undefined success/failure conditions, and no informed consent."
Insane opinion
Yes, that was because of things like
https://en.wikipedia.org/wiki/Tuskegee_syphilis_experiment
https://en.wikipedia.org/wiki/Stateville_Penitentiary_Malari...
https://en.wikipedia.org/wiki/Operation_Sea-Spray
https://en.wikipedia.org/wiki/Nazi_human_experimentation
I understand that certain people are salivating at the thought of a return to those times; I'm not one of them.
Wow.
A flippant dismissal in an HN comment does not actually negate reality.
This is silly.
FDA has essentially one requirement: prove that your drug is safe and effective.
The reason trial designs get more and more byzantine is because the drugs themselves work less-and-less well. They're far more nuanced and precise. The experiments have to be extremely well-controlled, and then this has to balance against cost/timeline of the trial, and that's why sponsors choose to use byzantine trial designs.
This is especially great because it puts anyone who actually wants to actually make a working drug at a significant disadvantage. It'll take them longer to get to market, cost them a billion dollars more, then their medicine gets to sit next to a thousand variations of "Vitamin C for Leukemia" that all cost a lot less.
There would be virtually no incentive for anyone to make an actual drug.
> Postmarketing surveillance can easily determine what's effective and what's not, and medical orgs can adjust.
"Easily" is doing a ton of work. Postmarketing surveillance can sometimes give low-confidence signal as to what's effective and what's not.
Dumb question: Wouldn't the market discover what works and what does not automatically, like any other product?
Step 1. Ensure extremely widespread use to ensure that effects are noticed relatively quickly (i.e. within a few decades)
Step 2. Hope someone notices correlations between that drug use and the positive/negative downstream effects.
Step 3. Hope someone who has access to the data required to actually conduct an observational study actually wants to run the observational study. By the way, you better hope all the info you need is available in insurance claims data (it's not) because that's really the only data you'll have at high scale. For most of these fraudulent drugs, you won't even have that! Because why would you go through the hassle of gaining reimbursement when you can just sell bullshit for cash? So in practice the required real-world data on these drugs and their effects literally will not exist in most cases. Where the data does exist, it'll be confounded by the unrecorded use of all the other fake drugs too.
Step 4. Hope the company selling a non-working drug doesn't have the clever idea of publishing countervailing "studies" backed by much larger budgets and with more existential motivation. They won't be assessed by FDA so these studies can just say whatever they want, of course.
Step 5. Lobby for outlawing of a drug that, in fact, didn't break any law whatsoever. Obviously it can't be illegal to sell a non-working in a regime that has no way to assess whether a drug works. So you really just gotta hope to post enough times that a drug is bad that people stop taking it.
But cigarettes' negative effects are significantly stronger than most drugs' positive (or negative) effects. So every single step is going to be dramatically harder, require more people taking the drug, more data, more time, more sophisticated analysis, more direct scrutiny from "the market."
Give it a few decades and, maybe a couple hundred or couple thousand deaths, and with a bit of luck "the market" will have eliminated one of the thousands of fake drugs proliferating across the market.
Thankfully for the company behind it, they can always just start up another totally legal fraud by just picking a different chemical (inert or not) and claiming a different medical benefit.
Oh yeah, meanwhile, doctors are just flooded with absolute bullshit claims 24/7 with no repercussions for anyone just lying to them. It's good stuff.
Literally one of the dumbest ideas I've ever heard.
>Literally one of the dumbest ideas I've ever heard.
It is capitalism 101. Not sure if that is the dumbest idea, because the whole world seem to run on it.
I am clearly talking about both positive and negative effects of drugs. Both would be generally extremely hard to detect, and literally impossible to detect without both 1) massive consumption of that drug and 2) a civilizational-scale rewriting of the data infrastructure that actually exists.
Perhaps you should explore Capitalism 201. You'd learn that markets only have all their most lovely features in an environment of complete information, which this is not. The more the information asymmetry (or complete ignorance) there is, the less-well markets work.
The proposal being put forth here is: what if we just had virtually no actual information about these products?
Well, the markets don't work. Capitalism 201.
Then you are answering to a point that was not raised.
> most lovely features in an environment of complete information, which this is not. The more the information asymmetry (or complete ignorance) there is, the less-well markets work...
I am well aware. But for what I as well as the parent is saying, this only require information regarding if the thing works or not for the people who use it as long as that information is not censored.
That's answering the question.
> this only require information regarding if the thing works or not for the people who use it as long as that information is not censored.
Can "the market" currently detect whether there is complex life in the Messier 54 star cluster?
No.
Is it because that information is censored?
No.
It's because that information doesn't exist.
Or more specifically, because that information is indistinguishable from noise. Even in the scenario of cigarettes' extremely strong effect (stronger than nearly every drug) deployed at massive scale (higher scale than nearly every drug), it required decades to establish causality.
Especially in the regime of drugs being released to market solely on safety data, the information about whether they work would be generally indistinguishable from noise. For way, way longer than cigarettes given the smaller population, smaller effect sizes, and the (presumed) proliferation of noisy concomitant "medications."
Do you think companies spend billions of dollars running RCTs for fun? The FDA doesn't require RCTs, they require only strong evidence of efficacy and safety. They do ultra-expensive RCTs because without them, it's virtually impossible for anyone to know – including the patients themselves – whether the drugs work.
Not if the signal is strong. Whether a thing works or not, is a pretty strong signal. Your equating it to bad effects of smoking is flawed and is not comparable.
You other arugments are also pretty bizzare, but not going into that for sake of focus.
> if the signal is strong, it will be visible
> your specific example of it taking decades of massive deployment and close scrutiny to detect an effect significantly stronger than most drugs is ummm... flawed and not comparable for no reason I can articulate
okay buddy
That's okay, you're someone who thinks the FDA is in cahoots with insurers and drug companies to approve ineffective or unsafe drugs at scale lmao. And here you are arguing that the market would actually naturally detect exactly that.
What the heck are you smokin? Did paracetamol have a weaker affect on temperature than smoking causing cancer?
Thank you for demonstrating my point.
Even if that is true, is it an intrinsic problem with trials or just bad regulation? If it is the latter then you need to change the regulations? Is the problem global - is every regulator everywhere demanding byzantine trails?
I wish people would stop saying this. First, controls aren't necessarily "fake treatment", they are often compared to other standard treatments.
Second, the treatment being tested can actually harm the patient more, therefore the people receiving your alleged "fake treatment" can actually come out better off. Which is the "fake treatment" now?
I don't disagree with your final point, but mainly with this increasingly pervasive and wrong framing of RCTs.
This is just nonsense. First, everyone in a trial is informed of the situation. It's not "unethical" unless you lie about it. If you participate in a trial, you do so knowing that you might not get the experimental drug. It's a selfless, honorable thing to do, and we shouldn't be framing it as some kind of scam.
Second, we don't give terminally ill people "fake treatment" (placebo trials). We give them current standard of care. Giving someone a placebo trial doesn't prove anything that would change clinical practice, because you want to know if the drug works better than what is out there today. Rarely is that standard of care "nothing", and this (bad controls) is actually a primary reason that a lot of drug company trials are rejected by the FDA.
If I didn't see the Wall Street Journal editorial board repeating the same garbage in defense of patent medicines, I'd write you off as simply having a sophomoric understanding of how trials work. I'm convinced that someone is driving this absurd narrative.
I think it’s pointless to quibble about very specific things like this, but that gets thrown out the window when someone claims “Definitions are the very basics in writing, check them before you publish” and then use their own nonstandard definition!
Seems like they wanted to nitpick about term usage when: a) the usage they’re taking issue with is actually fairly commonplace (e.g. https://www.who.int/news/item/27-10-2017-who-to-review-its-i...) b) their own suggested replacement is very heterodox (I’ve certainly never heard anyone call it that).
This is not happening.
> I was surprised to find that they usually discard papers based on observational evidence wholesale.
He he...welcome to the real world!
I noticed that as well. My eyes feverishly scanning the previous paragraph for the definition.