[appreciatorBus]

No one is asking to go back to Ancient Greece.

But given our track record, a little humility would go along way.

When a highly educated doctor tells you that something is safe, a person is going to assume that means that someone somewhere has proven that the substance is safe. If what they really mean is that no one really knows, but so far, no experiments have been able to prove danger, then we should say that instead.

[Aurornis]

> When a highly educated doctor tells you that something is safe, a person is going to assume that means that someone somewhere has proven that the substance is safe.

Contrast agent has been widely studied and determined to be reasonably safe. You’re not going to be administered any routine procedures or compounds that are known to be dangerous without an examination of the risks and benefits.

> If what they really mean is that no one really knows, but so far, no experiments have been able to prove danger, then we should say that instead.

“No experiments have been able to prove danger” is too generic to be usefully different than saying that it’s understood to be reasonably safe.

Even this paper isn’t saying that contrast agent is bad or dangerous in general. It’s exploring a potential effect that we can now detect and study.

[redwall_hp]

Exactly...it's also not reasonable to be asked to prove a negative. "Prove it's safe" (equivalent to "prove there isn't any danger") is "prove there isn't a teapot orbiting Venus" territory.

Every procedure has some negligible risk, and doctors are trained to mitigate major risks to peoples' health with screenings, medications and surgeries that are of lesser risk than the alternative of inaction. "Safe" is a reasonable explanation for the vast majority of laymen they have to communicate with.

[appreciatorBus]

My point is not that you must prove it safe. My point is that it is dangerous to communicate to people that something is safe, and simply assume that they understand that negatives can’t be proven, and you don’t literally mean that someone has proven it to be safe.

[aerox2]

This is pretty much how we get into the territory of "this product may contain peanuts" even if it has never even been near peanuts, but that warning is need because if in the offcase it has touched peanuts the company can't be sued. But this makes pretty much every other warning worthless.

We shouldn't have to clarify that everything is only 99.999% safe and assume that everything carries some form of risk even if small.

[Ekaros]

By that standards everything we do is unsafe. Every single activity we do carry some neglible risk. Explaining all of these would be lot more trouble than value in general.

[sarchertech]

By that definition there is literally no substance in existence that has been proven safe. Because the definition of safe is that no experiments have been able to demonstrate danger.

You can’t prove a negative.

[appreciatorBus]

You know that you can’t prove a negative. I know that you can’t prove a negative. Probably most people on HN know that you can’t prove a negative.

But when a person who doesn’t spend their time nerding out on science goes to the doctor and hears, “the substance is safe”, it is not a guarantee that they know that you can’t prove a negative. If you can’t be sure that your audience knows that it’s not possible to prove a negative, then you should be pretty cautious with your words.

[array_key_first]

Safe, in the context of living on Earth, means an acceptably low risk of a bad outcome.

Tylenol is safe. Tylenol can also permanently damage your kidneys.

Walking is safe. Walking can also permanently damage your cartilage.

Food is safe. Thousands of people die from choking.

We all know this, colloquially. When it comes to medicine, it is as if one's brain hops and skips right out of their ear. It's not magic, it works like everything else on Earth works.

[synecdoche]

I think there is a non-insignificant number of people who would understand the word safe as no risk, who if something bad happened to them after submitting themselves to such a safe procedure, would find themselves deceived. Technically, I think they would be correct. Therefore, it should be explained that there is a risk but that it is on some order that they can relate to, like the risk of walking down the street.

[sarchertech]

A doctor will 100% explain that a procedure has a risk. They will say something like this procedure is “generally safe” but there is a very small risk of complications. Then they will make them sign a consent form spelling out those risks.

[array_key_first]

> I think there is a non-insignificant number of people who would understand the word safe as no risk, who if something bad happened to them after submitting themselves to such a safe procedure, would find themselves deceived.

These people are then dishonest, because they know, deep down in their heart of hearts, this is absolutely not what safe means.

Again, everyone agrees eating an apple is safe. It's even good for you! But they also know every time they take a bite, there is a risk that they can choke and die. They know that. I know that. You know that. Everybody knows that.

Colloquially, even to the most naive, we know that zero risk does not exist, and that "safe" merely means "an acceptably small amount of risk". If we are changing our definitions based on the context, for example, everything on Earth and then medicine, that is dishonesty. If we are dishonest to ourselves, then we are delusional.

[jdietrich]

>no experiments have been able to prove danger

Which is strong evidence that the danger is very small, very rare, or takes a very long time to develop.

You don't need a large clinical trial to prove that being shot in the head is harmful; you do need a very large trial to detect that, say, a drug increases the risk of cardiovascular disease by 4% in a specific sub-population.

[bawolff]

By that definition, nobody knows anything is safe.

[appreciatorBus]

Yes, exactly, but that is the definition that people who are not doctors are going to use when doctors tell them that something is safe. So we shouldn’t do that.

[Spooky23]

Nobody has proven that not taking an action is safe.

[gf000]

There is nothing mathematically 100% safe, the human meaning of the word inherently involves some kind of uncertainty.

Going for a test itself via car has a quite significant risk itself, should the doctor say that you shouldn't move out of this room, it's not safe?

Like even regularly used medicine has some slight chance of an adverse reaction, that's how minuscule side effects multiplied to human population times the number it's taken results in.

Guess what often has many orders of magnitude greater risk? Continuing having the disease you went to the doctor with in the first place, or having it lie undiscovered.