Hacker News Reader

Comment by londons_explore | Hacker News Reader

I would like to think that in a global pandemic costing hundreds of thousands of lives, trillions of dollars of economic costs, and significantly lowering the quality of life of future generations, this research would have been done months ago.

How hard can it be to find 100 similar workplaces, and to 33 of each say "You guys must wear masks". "You guys must have all windows open all the time". "You guys must wear gloves all the time".

Then wait 5 days, and see how infections spread.

That activity might lead to the deaths, in the worst case, of a couple of workers. Yet the knowledge gained would save tens of thousands.

View on HN

You don't even have to force people to behave one way or another.

You can skirt around the ethical issues by giving Sample A an unlimited supply of free masks, and keeping Sample B as the control sample. Even with free masks, not everyone in A will wear a mask. And some in B might still bring their own masks. But A will have higher mask use on average, and with large randomly selected populations you can tease out the impact using IV regression[1].

[1] https://en.wikipedia.org/wiki/Instrumental_variables_estimat...

We need, all of us, to stop focusing on deaths as the measure of how bad this can be. You're asking people to not only place their lives on the line, but even if that doesn't happen they face a significant risk of weeks of ICU care, followed potentially by a lifetime of respiratory problems.

And medical bills!

With 30 seconds of thought you are failing to control for:

* Social habits of the company

* Number of infected people within the company

* Compliance with the protocol

* Age demographics

* Infection rates from places other than the company

* Weird selection bias resulting from the company needing to not lock down to begin with

Very, very difficult study

As long as the participants are randomly selected, all of those factors drop out into noise. The law of large numbers guarantees you that as sample sizes go to infinity, the magnitude in difference between the i.i.d. variables in the samples goes to zero.

* We don't have infinity, or sufficiently large numbers of companies to do this

* We don't have the resources to test sufficiently large numbers of companies like this

* The dependent variable in question is going to be entirely dependent on the number of people infected at a company at the beginning of the study, which is unknowable unless you can test everyone... which we can't.

We can accomplish many great things if we knowingly discard ethics for the sake of efficiency, yes.

Not really in this case though. Ethically there's not much of a concern with that study since we don't know what works so we're not doing any harm by prescribing known bad methods. The ethical harm would come from a poor study design that won't tell us anything useful. Getting good data from a study like this is really difficult, and we can't really accomplish anything by discarding good data for the sake of efficiency

I question if modern ethics are perhaps too focussed of the wellbeing of the individual, at the expense of the wellbeing of all of us...

Or, to put it more bluntly, a modern ethics committee would have rejected the idea of installing sewers in cities to prevent disease because one builder of the sewer might have died in the process. Without that, industrialisation probably would never have happened, and we'd all be subsistence farmers today.

This is especially true with studies.

Because it was thought unethical to conduct studies around masks and infectious disease, the evidence was sparse, so the WHO and CDC didn't recommend wearing them. This has directly led to tens of thousands of deaths.

> Or, to put it more bluntly, a modern ethics committee would have rejected the idea of installing sewers

You get to choose whether to be in the business of installing sewers. Saying that it's ok to force people into medical experiments against their will for the greater good is how you get Nazi holocaust atrocities. Medical ethics has a long history that it sounds like you might enjoy spending some time reading about.

We're all being forced into a medical experiment against our will. Might as well be a useful one.

> We're all being forced into a medical experiment against our will.

I'm disappointed by your false equivalence between being told to stay home and safe and being made into a guinea pig.

If being home and safe means you lose necessary income, then your ire should be directed at your governments at all levels for refusing to prepare critical safety nets and regulatory environments that would make it possible for you to stay alive and safe without laboring every day. But advocating a regression on everything that history has taught us about medical ethics is terrible and bad. We're talking about thousands of years of accumulated and refined history lessons here. Medical ethics is a perfect example of why learning history is important.

Yes, the natural experiment is being right now. You have the Western European countries with a social safety net that allows people to stay home for a few months to ride out the plague and the US, where such a thing does not exist.

This has economic repercussions. A good amount of meat processing plants are now idled because the workforce is off sick due to lack of protective equipment and sick leave. We are looking at meat shortages. I hope people make the right connections.

Might-have-beens have no relevance to the ethics of the choices we make now. We are in the middle of a vast medico-economic experiment. It is bizarre to consider the ethics of only the medical side of the experiment, not the experiment as a whole.

What "medico-economic" experiment?! This is much an experiment as "living in a war-torn country" or "being hit by a major natural disaster" is an experiment.

If people start to starve because we stopped being able to produce food or people start dying of other causes because our basic infrastructure start collapsing, then you can start talking about ethical dilemmas.

> If people start to starve because we stopped being able to produce food or people start dying of other causes because our basic infrastructure start collapsing, then you can start talking about ethical dilemmas.

I think its a safe bet that this is already happening on small scales. Noone is measuring the degree.

Food insecurity already happens in many places around the world, even rich countries. However, in those places it is an social-economical issue, not a lack of labor/resources to produce the food needed for the population.

There may be images of long lines of unemployed people that have to get from the government or food banks, but the food is still there. Farmers are still producing, distribution networks are still working, essential services operating. That part of the economy is still going.

> I'm disappointed by your false equivalence between being told to stay home and safe and being made into a guinea pig.

This is the false equivalence. How long does one have to stay home in order to be "safe"? If things calm down now, will they be worse in the winter? Will we still have to be "safe"? How do the consequences of curve flattening strategies that look more like half-hearted eradication attempts be compared to approaches that look literally like curve flattening?

Noone knows these answers.

OP should start with Mengele for some truly chilling experiments at the bottom of the slippery slope.

Very relevant to another article on HN today, "Why we can't build":

"The institutions through which Americans build have become biased against action rather than toward it. They’ve become, in political scientist Francis Fukuyama’s term, “vetocracies,” in which too many actors have veto rights over what gets built. That’s true in the federal government. It’s true in state and local governments. It’s even true in the private sector."

It's not hard to tell people to do that. It is hard to get people to really comply and avoid changing behavior just by enrolling them in a study. It's hard to select 100 workplaces that are similar enough that they can be compared to eachother and also representative of the myriad complex types of social areas that people congregate in - opening a window might work in one type of office but not at all in another situation. And also, you need a control. You can tell a group of people to not change their behavior in a global pandemic to see what the impact of each intervention was. But getting people to comply? You even have problems in clinical trials where groups of patients find ways to get together and discuss side effects they've been having to try and figure out who has the placebo and who has the real drug killing the trial. Running statistically rigorous trials on that scale to get good information is just really difficult because of the study subjects.

This experiment has kind of been done: some countries had everyone wear masks. Other countries said "whoever told you to wear a mask is dangerously wrong, and even if you have a mask you can't wear it for fear of public shaming because we've told everyone that you're cheating healthcare staff out of their protection by doing so".

And then look at the infection spread in each country.

I think part of the problem is that this doesn't fit neatly into any existing specialty. The "microscopic scale of nucleic acids or on the gargantuan scale of whole nations" both have very active research programs targeted at them, particularly the former. It's "the everyday face-to-face scale of inches and feet" that fails to have a scientific home.

Let me know when you're volunteering to be a rat in this experiment.

All of us have already volunteered to be a rat in a global experiment of sorts anyway. I'm not sure why volunteering for three plausible paths would be bad. With this kind of study, self selection might even be ok.

Reply on news.ycombinator.com