Hacker News Reader

From reading the post it's not at all clear to me that 2 of the 3 goals you list were accomplished by the rules.

A) Instead of actually looking out of the patient a list of arbitrary rules were imposed.

B) The risk (minimal) was not properly evaluated but instead massively overstated.

Given this failure of the process to meet said goals it seems perfectly reasonable to question the process.

(A) The rules are not arbitrary. They are the product of much careful deliberation.

https://en.wikipedia.org/wiki/Belmont_Report

(B) Is experimentation on suicidal patients really minimal risk? Maybe, maybe not. But failure of the investigator to show they have thought about it is a huge red flag.

(A) They were arbitrarily imposed in this case. Just because a rule is a good idea doesn't mean it is a good idea all the time.

(B) We're not talking about injecting the patients with some experimental drug here. We're talking about asking them questions that they were already being asked anyways. The lack of risk is self evident to anyone who considers the question.

(A) This is kind of a red herring though. The rules are not uniform. Sick patients require more protections than healthy adults. Placing an implant in someone's leg requires a more comprehensive risk minimization plan than asking someone to look at the ground for 5 seconds and then read a flash card.

(B) Is the lack of risk really self-evident? To everyone? Or just to you? Thus the question who should decide? And by what process do they decide?

I think perhaps we are talking past each other. I certainly agree that there should be rules for conducting research on human subjects and that there should be a process for independent review of said research.

I'm just saying that in this case the people tasked with doing the reviewing did a really bad job. And it seems like, from other things I've read, this sort of bad job isn't exactly uncommon. It would be great if the people in charge of this stuff could reform the system so that IRBs did a bad job less often.

As for your B point, if we're going to play "but what if" I demand we play it consistently. What is the risk of this study not being performed? Is it self-evident that we are better off in a world where those concrete procedures killed this study?

This is one of the core flaws in bereuacracy, looking at the putative benefits of a policy without considering the costs. No one really denies that some oversight is needed, the question is, is this oversight needed?

"But we cannot be blamed for a different eventuality, like saving lives. It didn't happen, and its our job to make sure that it doesn't, unless policy is adhered to the letter."

> The lack of risk is self evident to anyone who considers the question.

Even the author disagrees with you (and, oddly, himself):

From the article:

> Also, psychiatric patients are sometimes…how can I put this nicely?…a little paranoid. Sometimes you can offer them breakfast and they’ll accuse you of trying to poison them. I had no illusions that I would get every single patient to consent to this study, but I felt like I could at least avoid handing them a paper saying “BY THE WAY, THIS STUDY IS FULL OF RISKS”.

If merely offering breakfast has a reasonable chance of getting re-interpreted as offering poison, offering a consent form for a test the purpose of which cannot be revealed before it is given certainly has a nonzero chance of triggering some kind of similar episode. So the risk of having such patients sign a consent form isn't zero, as the author somehow implies everywhere except this paragraph. Thus it should have been written into the risks section.

I'm concerned you conflate two things. A paranoid psychiatric patient may see a world full of overwhelming risk behind everything. It does not mean that risk exists. It does mean, however, that confirming their paranoia, or seeming to, would cause a panic. We're concerned about the panic because it is unjustified. A good procedure would avoid this unjustified panic. The IRB requires a procedure that doesn't avoid it. In this way, the IRB failed to protect the interest of the patient.

> The IRB requires a procedure that doesn't avoid it

The IRB requires getting informed consent from the patient. This is not unreasonable. In fact, IMO, experimenting on people and then publishing the results of those experiments without prior informed consent is grossly unreasonable and creepy.

Parent's point is that the author concedes that in a psych ward context, even asking for consent carries serious risks. I'm not sure if this is true because I have no personal experience, but the author seems to believe it is true.

Look, it sucks that doing any research in a psych ward carries substantial risk. But that doesn't mean that you can perform experiments on psych patients without their informed consent. Even if you think that the experiment is NBD. Guess what? You're not the patient and it's not your medical data, so it's not your call. Insisting otherwise is incredibly disrespectful and dehumanizing toward psych ward patients. They are humans, they have rights, and an inconvenienced researcher is not a reasonable justification for nullifying those rights.

Are there aspects of IRBs that are infuriating and unnecessary? Absolutely. But requiring informed consent before medical data is used in a non-treatment context is -- and should be -- a basic right afforded to every patient. Even psych patients.

I hate IRBs, especially for questionnaires, and especially the stupid training. But posts like yours convince me that both are sadly necessary.

> human experimentation is serious business.

Sometimes. When you just ask a few simple questions, it clearly isn't.

Your list of the benefits of IRBs looks valid. But without a corresponding set of costs to weigh them against, it's meaningless.

Well a few simple questions about deep seated emotional trauma (e.g. molestation), or embarrassing or taboo subjects can present risks to patients. So like you say. Sometimes.

The question is who decides and how they make the decision? Surely it should involve someone other than the investigator, and have clearly defined rules so people know what to expect. Hence the IRB.

Yeah, but that kind of system has real problems.

The incentive for the IRB is to be as thorough and put up as many obstacles as possible. Both from a blame deflection standpoint and a bureaucratic empire building standpoint.

I'm not aware of any force pushing in the other direction to make the scrutiny proportional to the risk.

If this is true - and I'm just some programmer reading stuff on the web who knows little of real life IRBs - the end result is a system that overproduces IRB red tape and underproduces science.

Of course, complaining about a system is easy. Figuring out a better system is not.

> I'm not aware of any force pushing in the other direction to make the scrutiny proportional to the risk.

Each research institution has its own IRB, which I expect is composed by faculty/researchers. The have an incentive not to block their own institution from producing any research output whatsoever.

The point reiterated several times is that they were already asking the patients all of those questions as part of their normal procedure, without any of the alleged protections provided by the IRB. Once they wanted to use the data that they were already collecting anyway it suddenly became a huge risk that required all of this bureaucracy to mitigate.

Yes, but there is a key distinction between the practice of medicine and clinical experimentation. The first is (in theory) is done for the benefit of the patient receiving treatment. The second is done to test a hypothesis.

Putting a patient at risk to treat them (e.g. chemotherapy) is very different than putting them at risk to get data for a publication or NDA. There's no reason what's tolerable in one context should be in the other.

The thing is, in this case, there was no one in the 2nd category who wasn't also in the 1st category. The question is already being asked to treat them. The risk is already being taken. Using the data from that action for research doesn't expose them to any additional risk.

Ultimately, the issue is this: it costs less overall for the author to just comply with the standard way studies are done, than for the committee to carve out and justify exceptions just for his study.

The author complains of the IRB not doing Methods 101. Perhaps the author should have done Methods 201, then he would understand the concerns (which many human experimenters in this thread have elaborated on). As stated elsewhere, so many of the author's wounds are self-inflicted.

The author's point is not that his time was wasted. It's that the system is broken.

Sure, he could have spent a lot of time and effort to learn and adapt to the broken system, and his study would have been done, and fewer people would perhaps be misdiagnosed.

But the point is that it shouldn't be this hard to make the world a better place.

The IRB is there to protect innocent subjects from malicious studies. It's not that the system is 'broken', it's just 'not perfect'.

It's actually been pretty disturbing to read this whole thread. So many HNers who complain endlessly about the relatively trivial metadata that corporations collect on us and sell, they would quite happily get rid of this committee that ensures any studies done on them meet minimum ethical requirements and ensure that subjects know that they're being tested on.

Read the responses in this thread from other people who have done studies on humans that requires passing through one of these committees. Actual people who know what they're for rather than armchair critics. All of them present the same opinion: IRB's are annoying and sometimes a little frustrating, but thank god they're there, because they protect the public. Every researcher is going to claim that their study will cure cancer, but not every researcher is being truthful (because they're humans, after all), and plenty of researchers don't care if they hurt their subjects in order to add a feather to their caps.

The nature of scientific research ensures that there can be no "standard way studies are done". When you are trying to discover entirely new things, you will undoubtedly come into new situations that were not accounted for when the standardization is created.

It may be inconvenient, but advancing human knowledge does require that humans use their capacity to reason and examine circumstances on an individual basis rather than mindlessly checking boxes on some one-size-fits-all form.

Well, since you're arguing against standards in scientific research, then we may as well disband NIST and get rid of SI units.

Yeah, the author was not quite as bureaucracy-savvy as you, it's clear. You do realize that the whole study was nothing but matching the results of a questionnaire that was already being used against interviews that were already being done?. Literally, all he's doing is observing. And maybe asking a few extra people to complete the questionnaire, the wording isn't clear.

Said this in another comment, but I'll reiterate here. The big difference is purpose/intent. Looking under the hood of someone's car can be totally fine or a violation depending on whether or not I'm being paid to fix it.

Same thing applies here. In one case the questions are being asked to provide treatment the patient is paying for. In the other it's being done to collect data for a publication.

Bullshit.

If my mechanic wants to publish a paper on wear-and-tear on 4 cylinder engines using just the information he gleaned from doing work on cars that I (and others) paid him for, he has caused zero damage to me versus just doing the work on my car.

The same applies here. Patients come in for treatment, and they provide exactly the same questions as if they weren't doing research.

Private information about my mental health is much more personal and sensitive than the wear of my car's engine. Handling that data to an untrained student intent on publishing a paper can cause plenty of harm.

I agree, though, that this study could have been done more easily by accessing the patients' records post-facto. But it's not the IRB's fault that the experiment could have been designed better.

Yes, mental health information is personal and sensitive, but it looks like the IRB didn't even address patient privacy, it was instead something that came out of a later audit.

[edit]

Also it seems that post-facto experiments are being encouraged over controlled trials, which seems the opposite of what you want from an experimenters point of view.

More privacy protections were enforced later according to the account, but the very first protection of them all is asking for consent to use the data. That's what I was getting to. Consent to use your responses to treat you is different from consent to use them for research.

The author seems to believe that he is incredibly bureaucracy-savvy though.

Hey did you know that there are pens made specifically not to allow stabbing for prisons? https://smile.amazon.com/Prison-Flexible-Non-Lethal-Ball-Poi...

Fill out this 40 page form requesting for an exemption to the no-pen rule. In 3-5 years we will deny the request due to increased risks of patients choking on the more flexible pens.

Did you fill out the 100 page exemption to do the 40 page form? Did you also remember to complete the form completion class and ethics sub-committee on completion?

(Yeah, pretty sure is the the deepest pit of hell)

There wasn't a no pen rule by the irb, there was a mandatory pen rule.