As an academic researcher, it's most frustrating in that it feels like IRB scrutiny doesn't seem to align with risk.
I do accounting research. My human-subjects research involves surveying practitioners about their jobs, interviewing them about their experiences, and conducting very simple experiments online that ask them to make decisions. There is virtually 0 real risk to any participants of my studies. IRB does always give my studies "exempt" status, but it still has to be reviewed. And they will pester me about different things like where will store our data (Onedrive is fine, but dropbox is not for some reason). This process will typically take a couple weeks of back and forth.
Yet I have a friend who participated in a Kinesiology dissertation study where they were asked to do extremely strenuous physical activity. He fainted(!) at one point from the activity. And it seemed to them that there were relatively few safeguards from that happening. Now, I'm sure that study did have IRB approval, but it really got me thinking... are we really scrutinizing studies optimally?
mmooss 28 days ago [-]
> IRB does always give my studies "exempt" status, but it still has to be reviewed.
How could they do it more efficiently?:
How could they know it's exempt without reviewing it? Should they take researchers' at face value? That seems to undermine the reason we have IRBs, which was unscrupulous researchers. Should we assume they are willing to torture people but not to mislead the IRB?
Why shouldn't data be private? How hard is it?
> a Kinesiology dissertation study where they were asked to do extremely strenuous physical activity. He fainted(!) at one point from the activity. And it seemed to them that there were relatively few safeguards from that happening.
Informed consent is essential to an IRB; your friend would have read and signed something detailing the activities and risks. Strenuous excercise to the point of exhaustion is a part of sports performance research, at least. As long as you inform people, they have the power to opt-out (not prisoners, 6 year olds, etc.), and no undue risk.
bpodgursky 28 days ago [-]
> How could they do it more efficiently?:
Think about how taxes work. The IRS doesn't check every person. You make rules, you perform random audits.
IRBs could easily do the same thing. Set some rules like (simplified) "No IRB necessary if your research doesn't physically touch a person". Researchers will read the rules and skip an IRB if they're in an exempt category. Then just audit at a high enough rate they won't lie.
dr_dshiv 28 days ago [-]
There is a huge opportunity in simplifying and automating IRB review.
Here in the Netherlands (we call it HREC, human research ethics committee) the process takes months and months to get permission to talk to people.
Like, “hey I’ve designed some cool new interaction design, I want to get permission to interview people and do human-centered design iterations.” Yes, you will need to spend dozens of hours on forms and wait months.
It cannot be ethical to put these many barriers on talking to humans. And, if you want to do something like an educational game for kids? Or support for the elderly? Or for your students? Well, those are vulnerable populations, so add a few more months.
Education research is extremely difficult here (in the USA it is exempt). How to conduct research on the use of AI in a class you are teaching? You can’t get retroactive consent.
The system is not reasonable.
mmooss 27 days ago [-]
Nobody dies if someone's taxes are wrong. Also, IRBs are not usually about lies, but about scientists not understanding the risks and solutions.
bpodgursky 27 days ago [-]
> Also, IRBs are not usually about lies, but about scientists not understanding the risks and solutions
You have never submitted an IRB.
mmooss 27 days ago [-]
Have I? Haven't I? Are you really saying that all these scientists - anyone doing human subject research at least in advanced democracies - are mostly liars trying to defraud people?
bpodgursky 27 days ago [-]
Have you submitted an IRB?
jampekka 29 days ago [-]
We are not. Here's a great essay about the subject:
> I do accounting research. My human-subjects research involves surveying practitioners about their jobs, interviewing them about their experiences, and conducting very simple experiments online that ask them to make decisions. There is virtually 0 real risk to any participants of my studies.
Cool, I guess. I wouldn't want to participate if my answers were visible to my employer because honest answers could put my employment at risk.
> As an academic researcher, it's most frustrating in that it feels like IRB scrutiny doesn't seem to align with risk.
As a worker^H^H^H^H^H^Hhuman I find it most frustrating when academics don't understand what things I consider to be risks.
If my answers to your research are leaked, can they be tracked back to me? Onedrive is terrible, just look at Microsoft's repeated flagrant disregard for security. Dropbox is awful too, just look at how easy it is to accidentally a whole folder. Why would you store your data in such places instead of on a research computer with locked down access? That screams to me of a researcher who doesn't understand the value (and risk) of the data they hold. Or perhaps of a researcher who doesn't value that risk appropriately. Perhaps also researcher who doesn't coordinate with their IT department, or an IT department who's equally or worse as competent.
p_j_w 28 days ago [-]
>Onedrive is fine, but dropbox is not for some reason
As far as I know, it's possible to get OneDrive even HIIPA compliant, but I don't think so for Dropbox.
eszed 28 days ago [-]
Dropbox claims that it can achieve HIPAA-compliance, and I personally know two (large and serious) organizations who use it as such.
In keeping with the theme of the OP, I don't know if that's been tested in court, soooo... Take that all for what you will.
derbOac 29 days ago [-]
The answer is no, you don't need permission, but you do incur a legal and ethical risk whenever you do research, and as such, the more you can do to strengthen the legal and ethical arguments in your defense, the better.
IRBs are simply a way to say "I had an independent group of ethical and legal authority review my research and they thought it was ok".
billjive 29 days ago [-]
That is right, and further - I think about IRBs like I do non-profit boards. One of the key tasks of a board is to determine what risk the organization is willing to bear. I think the IRBs at these institutions are 1) following federal laws but 2) making a judgement on the risk the institution is willing to sign up for. For example, for many clinical trials are distributed to multiple "sites" where the trial is actually administered. Many of these sites are universities. What the University of Iowa is willing to bear is different than the University of Michigan and the IRBs reflect that.
mionhe 28 days ago [-]
With medical device development, the IRB represents the institution taking liability for harm to patients. They're there to make sure they aren't about to kill someone or harm them and get the institution sued.
So they take it pretty seriously. More so in countries where it's easy to sue, though.
rights_reminder 29 days ago [-]
It's hard to take the article seriously since
1) It doesn't even reference the relevant law
2) The relevant law is so simple and clear
42 USC 289(a) states:
"The Secretary shall by regulation require that each entity which applies for a grant, contract, or cooperative agreement under this chapter for any project or program which involves the conduct of biomedical or behavioral research involving human subjects submit in or with its application for such grant, contract, or cooperative agreement assurances satisfactory to the Secretary that it has established (in accordance with regulations which the Secretary shall prescribe) a board (to be known as an “Institutional Review Board”) to review biomedical and behavioral research involving human subjects conducted at or supported by such entity in order to protect the rights of the human subjects of such research."[1]
To put it succinctly this only relates to any entity which applies for a grant, contract, coop. agreement under chapter 6A of 42 USC.
This in no way relates to private research. It's easy to get hung up on regulation spaghetti code, but regulations cannot exceed the scope of the statutes, which in turn cannot exceed the authority of Congress.
> To put it succinctly this only relates to any entity which applies for a grant, contract, coop. agreement under chapter 6A of 42 USC.
But the whole point of the article is that there are other cases when you need IRB approval besides this one. That's one of the myths he's busting.
rights_reminder 28 days ago [-]
The point that I was making was that there can be no federal law which requires an institutional review board for private research, which was something the author alluded to in the paragraph after the flow chart:
> The government can’t possibly claim to regulate what me and my roommates eat at home! That would be stupid!” Yes, it would be stupid. But who says the world makes sense? Now, would you actually be prosecuted for violating those rules?
Author says "those rules" without actually citing anything.
There is no scenario in which the Secretary of HHS can create rules which would do what the author was suggesting. It is not even something Congress could do, which is why they haven't created a statute which goes beyond their authority in requiring these boards for anything other than research receiving federal funding. If the Secretary were to create such a rule, it would obviously exceed his authority.
j16sdiz 28 days ago [-]
> Author says "those rules" without actually citing anything.
The author actually go into how universities make their own rule that potentially affects teachers and students doing research with their own time and resources.
karaterobot 27 days ago [-]
Then I apologize, I evidently misread your original comment.
not2b 28 days ago [-]
Yes, but a university or other institution will often also require IRB review in additional cases, because maybe they don't want to let researchers accidentally cause brain damage just because a billionaire rather than a government is paying.
mmooss 28 days ago [-]
Also pharma companies.
cal1co 29 days ago [-]
An article like this is interesting from a research (meta-research?) perspective, but I think the lack of relevant case law (nobody seems to have ever been prosecuted) really undermines the usefulness.
It's easy for people (especially the kinds of people who gravitate towards engineering) to get all worked up about the necessary implications of the regulations as-written. But it's also important to remember that being facings consequences would require
A) The intentional action of either a state prosecutor or an injured private party
B) The involvement of a human being as a judge to interpret the facts and applicable rules
When you look at it from that angle, it's obvious why organizations like META just do the research, imho.
Xcelerate 29 days ago [-]
> It's easy for people (especially the kinds of people who gravitate towards engineering) to get all worked up about the necessary implications of the regulations as-written
I was once trying to figure out taxes owed on some RSUs granted in California that vested in North Carolina (and some other edge case that I can’t remember). Hoh boy. What a rabbit hole. I didn’t think the CPA that I hired to do my taxes handled it correctly, so I started digging into the tax laws and regulations. Couldn’t find anything that covered my specific case. So I started digging into court cases involving disputes over RSU taxes owed in North Carolina and found conflicting outcomes.
I brought what I had found back to the CPA who basically said something along the lines of: there is no absolute “rule” for what to do in this case like you’re looking for. We make a good faith effort to mirror what was done in similar prior situations, write a number down, and then forget about it unless you hear from the IRS.
Did not make me feel very confident about the legal system.
floxy 29 days ago [-]
>...In the real world, people usually attempt to solve problems by forming hypotheses and then testing them against the facts as they know them. When the facts confirm the hypotheses, they are accepted as true, although subject to re-evaluation as new evidence is discovered. This is a successful method of reasoning about scientific and other empirical matters because the physical world has a definite, unique structure. It works because the laws of nature are consistent. In the real world, it is entirely appropriate to assume that once you have confirmed your hypothesis, all other hypotheses inconsistent with it are incorrect.
> In the legal world, however, this assumption does not hold. This is because unlike the laws of nature, political laws are not consistent. The law human beings create to regulate their conduct is made up of incompatible, contradictory rules and principles; and, as anyone who has studied a little logic can demonstrate, any conclusion can be validly derived from a set of contradictory premises. This means that a logically sound argument can be found for any legal conclusion.
Thanks for the link! I have been thinking a lot about topics related to this lately and this looks like an interesting read. Skimming through it I might not agree with all the ideas of the author I very much welcome new ideas about how to organize our society and our legal system. I have especially been thinking about how the legal system is mostly fiction and what makes it work is not so much what the laws say and what the punishments are for breaking them, but rather our shared belief in the system. And I think this is problematic because it takes away our agency to imagine other ways of doing things, we think our current legal system is the only possible system because it has to be. Thoughts of anything else would cause a collapse.
I have also been thinking about how basing morality on following laws is a bad idea. You should not be afraid of breaking laws or rules if they are stupid. All of modern workers rights, women's rights, etc are based on brave people of the past that were not afraid of breaking stupid laws of their time, why should today be any different.
noodlesUK 29 days ago [-]
I think this illustrates an issue in common law systems in general. Contrary to what we’re taught in school about how actions are generally legal by default, and only illegal if a law has been passed to prohibit them, in reality, actions exist in a superposition of legal and illegal until someone challenges that position and a court finds out (subject to any appeals). You can only evaluate in practice whether something is legal based on copying others in similar situations. The further you get from the known outcomes the fuzzier this all becomes.
This tends to be something people encounter a lot around tax, complex business structures, and various international dealings where you’re more likely to end up in an edge case that hasn’t been well explored.
miki123211 28 days ago [-]
And it's worth emphasizing that this is a good thing.
Civil law systems (which are used in the non-English-speaking parts of Europe for example) don't have this property. In such systems, court precedents don't really exist, so if something isn't clearly defined by law, and there's no extremely ubiquitous legal interpretation that everybody seems to follow, you don't know whether it's legal or not.
For example, the Polish tax authority lets you ask for a tax interpretation for cases that aren't legally obvious, something that I believe the US would have handled via the court precedent system. It's not unusual for two people in identical situations to receive opposite rulings, and Those rulings are not binding on the authority that issued them.
In other words, you don't know how to handle your tax situation, you ask your tax authority, they tell you "hey, we're the tax authority, we're telling you to do x", you do x, they audit you, they say "this isn't clearly defined in law, we now believe you should have done y instead, here's your orange jumpsuit."
garaetjjte 26 days ago [-]
Individual tax interpretations are not binding (in the way that they don't alter the law, and they might be rescinded at any time), but they grant immunity from prosecution (art. 14k of tax law).
And eventual orange jumpsuit would be decided by the court, not tax authority.
dekhn 29 days ago [-]
It took me a while to accept that the law is not a consistent system with hierarchical top-down rules that can be used to determine legality in an unambiguous way.
If you're a software engineer and you're knee-deep in analyzing court cases to find out how to file some tax correctly, you're definitely overthinking it.
The CPA was exactly correct.
Xcelerate 29 days ago [-]
Yeah but now I just switch my perspective from thinking of it as a formal deductive system to thinking of it as an optimization problem.
dekhn 29 days ago [-]
Yes, that's right. It's a Bayesian Game and you are attempting to reduce risk to an acceptable level (not under paying) while minimizing loss (not over paying tax) with only partial information (what the results of a lawsuit would determine).
rights_reminder 29 days ago [-]
Chances are your CPA was not correct. CPAs routinely violate the law as a regular course of duty and the IRS is fine with it, even encourages it, because it gives them more revenue. If you are looking to go down a multi-year rabbit hole on this, then try to find the law that imposes any tax on __your__ domestic income.
(Hint* You can't)
aqueueaqueue 29 days ago [-]
If you move from Germany to Brazil you might also hit edge cases.
Yes that is 2 different countries. But it is similar. States have their own laws.
You can't expect the law across 3 jurisdictions to work out every permutation and legislate for it. There are 52 (I guess more... lots of territories) * 51 permutations of moves so 2652 moves that are possible.
In addition you are talking RSUs and not some vanilla thing like claiming an expense.
appleorchard46 29 days ago [-]
This is addressed at a few different points in the article, I found this bit particularly insightful:
> So what happens in practice is politicians write a vague law. Bureaucrats turn that law into very detailed (but often still vague) specific rules. Those rules might or might not be “legal”, but nobody want to risk fighting them in court. If the regulations are particularly ridiculous or likely to be overturned if challenged, prosecutors may quietly stop bringing cases. But the regulations still sit there on the books. And people still usually pay attention to them, because why risk it?
billjive 29 days ago [-]
This specific quote from the article made me think of lawmaking - the more detailed the law the harder it is to pass (someone can always fine something to nitpick). Therefore, laws are best at establishing goals, principles, guide rails, etc. Nothing is perfect but the way in which laws are implemented SHOULD remain a separate part of the law passage.
Separate rant: I live on the west coast and am annoyed when I vote because of the number of ballot measures. None of them are sufficiently detailed and fall victim to vague language that's open to interpretation. I'd rather "hire" law makers to do the hard work of details law creation.
nradov 28 days ago [-]
Nah. There's no valid reason for lawmakers to delegate so much authority to bureaucrats. We should eviscerate the administrative state and if legislators want to prohibit something then they ought to specifically write it down. If that means it becomes harder and slower to pass new laws then that's fine. It should be a careful and deliberative process.
memhole 29 days ago [-]
Some of that vagueness is intentional and not always for nefarious reasons either. When getting my the FCRA cert, there definitely a few requirements that were obviously very open ended so the gov could either pursue a potential unforeseen circumstance or not stifle business. The vagaries in laws has always been interesting to me. Unfortunately, I don't remember any definitive examples since I took the test years ago now.
cal1co 29 days ago [-]
I think the article did sufficient to talk about 'what would happen if one were prosecuted,' given that it's really hard to predict.
What I thought was missing was an acknowledgement of the circumstances that would cause a _prosecutor_ or _other litigant_ to decide to _bring an action to court._
Prosecutors are political, have limited resources, and are ultimately more or less accountable to electoral forces. Their prerogative is (almost) never going after every single person who does an action that is technically illegal. Factors like harm, the reputation of one's actions in their community, and their alignment with the political establishment are going to be considerations.
As written, I could see the article maybe deterring a particularly rules-abiding individual from running a potato-eating weight loss experiment with their roommates. I've seen friends and relatives get all wound up worrying about whether something is technically illegal in some marginal way.
Understanding the human factors in the pipeline between a potentially unlawful action and a consequence for it is really important. All the institutions that get anything done are keenly aware of them.
appleorchard46 29 days ago [-]
It is briefly mentioned in the notes at the end that nobody has seemingly ever been prosecuted for this, and the laws are primarily aimed at institutions doing shady stuff rather than individuals wanting to feed their friends potatoes. But yes, the focus definitely isn't on that side of things.
I don't think it's fair to call that a deficiency, though; both the letter of the law and the practical applications are complex and interesting topics in their own right(s?), and either could fill a lengthy article as we see here. Both are welcome! One doesn't preclude the other. I would be interested in reading an article covering what you describe too.
nradov 28 days ago [-]
Violations of federal regulations (rules) are usually handled as administrative or civil matters. Prosecution is very rare and only possible for a limited set of rule violations. If you're contemplating doing something in a regulatory gray area then you can ask the competent agency for a formal opinion letter stating whether that thing would or wouldn't be a violation. They are required to issue those letters although it can be a slow process. Once you get an official approval letter that generally shields you against any sort of enforcement action.
mmooss 28 days ago [-]
> nobody seems to have ever been prosecuted
But they have been sued, sometimes for extraordinary amounts, as have their employers, corporate sponsors, etc. And they have lost funding for their research and the trust of everyone above (and of future IRBs). These are powerful disincentives.
nxobject 29 days ago [-]
Conversely, a social and “in practice” reason why government-funded institutions _do_ use IRBs is to allay fiscal hawks and criticism about state funding being misused.
ahi 29 days ago [-]
Also, a great deal of the research organizations like META "just do" wouldn't pass muster with any IRB worth the name.
tonetegeatinst 29 days ago [-]
This is just as confusing and explosives and energetics research.
Explosives are regulated by the ATF, but can also fall under regulations of the DOT and the EPA. And if your doing any pulse power z pinch or electronic detanator experiment or research, then that's also regulated.
Keep in mind this also has state regulations and ordinance to worry about and its almost like they are trying to get you to accidentally make some mistake that can technically be a felony. And they can visit and do inspections during your hours of operations.
Potato cannons, model rockets, and innovation and experiments in your backyard seem to be impossible these days.
Do I think safety is important, sure, but I'd argue that the paperwork of regulation has stifled the common man who want to follow the laws.
Also see: export control and gpu's
oneplane 28 days ago [-]
You'd think someone would notice how this process is suboptimal and either come up with a way to make the process better, or at least a way to have a single point of contact with knowledge and experience (and reputation) to deal with this for you, i.e. on a more local governmental level.
As a random citizen, you'd perhaps get away with an online self-service form, as a small company or startup you might need a form and an interview, but the entire "talk to 4 different agencies at different levels of abstraction" seems like a waste of time for everyone when it can be done in one go (provided the thing you wanted to do is common enough that it doesn't require some extreme documentation and safeguards etc).
If 100000 people were to want to try out a potato canon, it should be sufficient to do something at a local level where you select "I wanna play with a potato canon for a bit" in some drop-down, and it might ask you if you intend to shoot at people or not, and if you have enough room to experiment with this without hurting anyone (including yourself). If yes, problem solved, no need to fill out a form, if unsure or no, fill out a form since you probably are doing something novel, or something you might not want to do. If this then means some other agency at some other level needs to get notified, sure, automate that, who cares. If it turns out you filled out your form and specified radio active potatoes, perhaps one of those other agencies might want to have a chat first to see if everything is okay.
But the default of "talk to an arbitrary number of agencies in every case" is not great.
nxobject 29 days ago [-]
> I'd argue that the paperwork of regulation has stifled the common man who want to follow the laws.
I’d argue that it’s not the regulations themselves, but people who want to avoid liability, whether ethical or social. (The existence of institutional review is a good argument against, say, a fiscal hawk looking at a state university’s research output to identify the existence of fraud or abuse.)
tptacek 28 days ago [-]
The article suggests that simply agreeing with friends to experiment with diets would trigger the IRB requirement, and cites a NY state law requiring affiliation with an IRB-enabled institution to do human research. But the "definitions" section of that New York law says:
2. "Human research" means any medical experiments, research, or scientific or psychological investigation, which utilizes human subjects and which involves physical or psychological intervention by the researcher upon the body of the subject and which is not required for the purposes of obtaining information for the diagnosis, prevention, or treatment of disease or the assessment of medical condition for the direct benefit of the subject. Human research shall not, however, be construed to mean the conduct of biological studies exclusively utilizing tissue or fluids after their removal or withdrawal from a human subject in the course of standard medical practice, or to include epidemiological investigations.
lmm 28 days ago [-]
Are you claiming that agreeing with friends to experiment with diets does not fall under that definition? If so, which part do you think it doesn't meet? If you read "physical or psychological intervention by the researcher upon the body of the subject" in such a way that a friend telling another friend to follow a given diet doesn't qualify then most medical research would be exempt too - researchers generally give subjects medication and tell them to take it rather than physically pushing it down their throat.
awo495jwal34i5 29 days ago [-]
According to the Office of Compliance at the University of Texas at Austin, my graduate advisor didn't need to get IRB approval to perform experiments on human test subjects that risked severe bodily harm. I reported him, they ignored the complaint, and then gave him tenure.
They also didn't care about the part where he falsified his results or lied to his sponsors about how he was spending their money.
eesmith 29 days ago [-]
This topic came up 11 years ago after Facebook did human subjects research by tweaking a ranking algorithm.
This is basically why some researchers gravitate towards certain model systems in terms of abstracting the problem. It is just easier in terms of approval the simpler things get and you can basically grasp at the same sort of questions since basic cell biology is so conserved. You can do things like grow up a generation of transgenic yeast that glows green under UV light and then pour bleach over it and throw it in a carboy for disposal over the week without anyone needing approve anything but the form for the carboy to get incinerated. You try and write up a proposal for transgenic mice it might take months since the IRB needs planets to align for meetings to be scheduled. And you are paying orders of magnitude more for orders of magnitude fewer individuals in your sample that will take orders of magnitude longer to get data, than an organism that can measure reproduction rates on the log scale over hours and cost pennies on the billions.
trhway 29 days ago [-]
>"HEY! YOU CAN’T DO THAT! THAT’S HUMAN SUBJECTS RESEARCH! YOU NEED TO GET APPROVAL FROM AN INSTITUTIONAL REVIEW BOARD!”
reminded - The EcoHealth Alliance application successfully granted by NIH for those Wuhan coronavirus gain-of-function experiments did have the "Human Subjects Included" and "IRB" checkboxes checked :)
asimpleusecase 29 days ago [-]
I worked in environmental chemistry - commercial lab. We did not do tests on humans nor animals but we were constrained by federal law and commercial liability . If the test you were doing was covered by federal law appeared in the CFR ( code of federal regulations) you needed to make if you said you were doing a test that was detailed in law that you followed that procedure in a documented way. If you changed it in a significant way you were engaging in a felony. Then you had some
Tests that were not covered by federal regulations- so how do you protect yourself
from lawsuits ? ( the fact you could do a test that would cost $100 but the client was making a million dollar decision with the results meant there could be a lot of liability risk)so in those cases where there was no clear regulation you would opt for “industry best practice “ using processes, equipment, reagents that were commonly used for that test by others in the
Industry. If you did that then your insurance provider would defend yourself. If you had good documentation you could prove you did the right thing the right way and win the lawsuit.
Based on that background I suspect the biggest reason for review boards is liability protection.
caycep 29 days ago [-]
I feel like the OP is overthinking it. The simple answer is you submit an approval for an IRB waiver, and if your IRB isn't staffed with nitwits, they usually grant it. (i.e. most retrospective studies including w/ chart review, or "minimal harm" studies i.e. w/ asking questions etc, w/o any drugs/surgeries/blood draws etc, or any risk of disseminating private health information usually get the waiver)
scraptor 28 days ago [-]
This assumes that I already have "my" IRB. Most people who want publish an informal online survey do not.
nxobject 29 days ago [-]
And, conversely, when people _do_ weaponize institutional review requirements to police research, you have a much more pressing and bigger problem: the existence of people who would be happy to find any mechanism to police research.
thereisnospork 29 days ago [-]
I would argue that the weaponization is an intrinsic part of the nature of review boards. Not just for ethics, but accross pretty much all categories.
It's been pretty universal in my experience that people will use the power given to them to push their agenda, which is generally only somewhat aligned with the nameplate mission.
E.g. a vegan or vegetarian on an ERB has an obvious, if soft, conflict of interests that will come out in how they treat requests involving animals, or whose results might lead to future testing on animals.
caycep 28 days ago [-]
But...people and people-communications make it work, i.e "Technical problems are easy, people are hard." This is a truism of life. I think this comment in the marca linux kernel brouhaha thread is relevant.
(and most IRB squabbles that involve non-invasion/low risk research never really get as bad as that, at least in my experience, knock on wood)
Do you have experiences with IRBs? IME, most structures and people want to get along and at worst, gain favors and reduce enemies. People are short-sighted sociopaths, usually.
caycep 28 days ago [-]
a neurosurgeon I worked with could get his waiver approved in a day usually. But he made friends w/ all the IRB members and had a significant budget for donuts donations.
mmooss 27 days ago [-]
Oops, I meant to say, 'People are not short-sighted sociopaths, usually.' I need to double-check before submission!
caycep 29 days ago [-]
In which case conversations should be had about individual board member behavior. It is not a reason to not to have safety oversight.
cm2012 29 days ago [-]
IRBs are one of the most wasteful and hurtful institutions in the developing world today. In order to prevent a few possible ethical lapses, they smother hundreds of good potential research.
And yes, airport security is also a colossal waste
dekhn 28 days ago [-]
IRBs dont prevent ethical lapses; they reduce the liability of the contractee/grantee.
mannyv 29 days ago [-]
This is what good lawyers do: dig in to the sausage factory.
What's interesting is how agencies push the limits on what they can do. All agencies do this to one extent or another. And they obviously won't tell you if they're doing that to you or not.
Granted authority and actual authority are different. It's worth remembering that when dealing with any bureaucracy.
mannyv 29 days ago [-]
As a note, the FDA circulars in particular (like the labeling one) aren't technically law, but because they're (1) put out by the FDA and (2) a good summary people treat them as law. The thought is that you can always point to the circular if you need to.
You should read the actual regulations as well, but that's work so lots of people don't do it.
Here's an example of why you should read the regulations. Technically, supplements can't make structure/function claims. However, you can submit your structure/function claim to the FDA. Once you do that you can make that particular structure/function claim on your labels. Since the FDA has no process in place to actually review those claims your structure/function claim is suddenly valid.
It's amusing.
jp57 29 days ago [-]
I helped my daughter do a human study for her fifth grade science project. There were inevitably jokes among my friends on Facebook about getting approval from her school’s IRB.
froindt 29 days ago [-]
In my AP Statistics class, we had a few weeks after the test before school was out. We had to do a statistics project, and the teacher didn't particularly care what it was.
We were checking which candy was most popular from a giant bowl, and went classroom to classroom just asking a randomly selected student to pick a piece of candy.
We got 100% on the report, which included "We didn't inform the participants of and potential harm from participating in the study. We're heartless bastards, oh well."
Then I went to college and helped in quite a few study designs with a Professor who had worked with over 30 IRB's across his career. Ours was by far the most strict. We were primarily doing software usability studies, and every single question we would ask a participant needed to be verbatim. Your data anonymization, destruction, and analysis plans had to be fully defined, including storing the participant name to UID key separately from the data.
Deviation or asking follow-up questions based on a prior response was either not allowed or a whole additional huge layer of SCRUTINY (I'm not sure). I'm genuinely curious how any sort of therapy focused study could have occurred with that IRB.
mmooss 28 days ago [-]
Maybe the professor was training you and didn't care about the value of your undergraduate study, which was almost certainly not worth the professor's time otherwise.
froindt 21 days ago [-]
Late reply, but that experience was actually in grad school. I was involved with research as an undergrad as well, but it was definitely confusing from how I stated things.
I was funded through a DARPA grant. It was definitely not a training exercise.
CharlieDigital 28 days ago [-]
This is also my take; it's part of the process of practical transfer of knowledge and process.
He did extra work, but it was effectively practice.
nxobject 29 days ago [-]
Congratulations: you are seeing the effect of decades of philosophical wrangling about the ethics of research.
ikanreed 29 days ago [-]
And it's an interesting time to think about this problem.
Because we're about to see how all the structures that were put in place to enforce those ethics react to being told they must instead do keyword searches for forbidden phrases set by fiat.
If you're the kind of administrator who does boring work of making sure all professors in your biomedical campus submitted IRB paperwork, are you also going to be the person who makes sure their papers don't contain the word "Woman" or "Historically"?
nxobject 29 days ago [-]
It is, and I do think it’s as subtle as the conversation on any other checks-and-balances mechanism in government these days: we can weaponise both painstakingly following the rules, or weaponizing the fact that rule enforcement is a human process that people ignore by convention.
mmooss 28 days ago [-]
Has that happened to you?
ikanreed 28 days ago [-]
No, this is a very recent development in science policy.
mounceyboy 28 days ago [-]
This is a nice first pass at the common sense problems with interpreting FDA regulations. I imagine that those who have been dealing with this for some time aren't as confused as the author suggests we should be. On the first pass, for example, are human subjects involved? Do you interact with humans? No, then probably not. There are many subtleties after that, but it's not so bad as the poster suggests. I remember a comment one investigator told me about animal research, where he illustrated the problem by saying you can go into a grocery store and buy a leg of lamb and do anything you want with it, but don't call it research!
Isamu 29 days ago [-]
This gets my enthusiastic upvote. Reason: author bothered to do work to research his answer.
In contrast, a shockingly large amount of content is extremely lazy, gut-feeling reaction. Avoiding the work and disappointment of finding out that your hot take is wrong.
web007 29 days ago [-]
> As an analogy, driving a car is dangerous. Whenever I drive, I could easily kill someone. But the government doesn’t force me to submit a driving plan any time I want to go somewhere. Instead, if I misbehave, I am punished in retrospect. Why don’t we apply the same policy to research?
"We" decided that Tuskegee was bad enough that it should be stopped before harm is done, and that there is no appropriate or sufficient "punish[ment] in retrospect" for the fallout.
The government makes you get a license to drive at all, then "drive a Pinto" versus "drive a Trabant" are similar enough that they don't require more info. They require you to get different licensure to drive a bigger truck where you could potentially cause more harm, or to drive an airplane. In this analogy the IRB is the DMV/FAA/whatever, and you're asking for permission to drive a tank, a motorized unicycle, a helicopter, an 18-wheeler or a stealth fighter. You don't get a Science License rubber stamp because that's like getting a Vehicle License - the variation in "Vehicle" is big enough that each type needs review.
jhbadger 29 days ago [-]
>"We" decided that Tuskegee was bad enough that it should be stopped before harm is done, and that there is no appropriate or sufficient "punish[ment] in retrospect" for the fallout.
The thing is, although you and the linked article seem to be associating IRB approval just with human studies, these days you need it for mouse studies.
billjive 29 days ago [-]
There are different IRBs to review animal research[1]. I believe it created for an ethical framework around the use of animals in science. Same thing: what are "we" accepting of when it comes to research of this nature?
In this analogy, you're also asking permission from the IRB to ride a bike or skateboard.
nxobject 29 days ago [-]
In the context of universities, the equivalent of riding a bike or a skateboard here is having people fill out surveys after events, or piloting new services offered by a student health clinic.
(I guess the point of analogies like these are to force us to sweat the details and examples.)
nxobject 29 days ago [-]
Another equally good comparison is, say, the existence of flight plans for private pilots, flight logs, etc.
There are escape hatches, too: I doubt many rural Alaskan pilots worry (or need to worry) about these things.
xyzzyz 29 days ago [-]
Not sure what your analogy is here; private pilots typically don’t need to file any flight plans except in specific circumstances.
spondylosaurus 29 days ago [-]
Yeah, odd choice to use cars in that analogy when you very much need advance approval before you're allowed to drive a car.
A driver's license is more like a medical license than IRB approval.
xyzzyz 29 days ago [-]
Indeed, and a person with a medical license is able to do much, much more damage than the people who need to ask IRB for permission to do research.
If your point is that we could replace IRBs with some sort of a researcher license, that you need to obtain before being able to do studies that today require IRB approval, then I support it, because while not ideal, it improves over the status quo.
nxobject 29 days ago [-]
I do think that there’s a debate to be had on how reactive or proactive we should be in ensuring the ethical practice of… well, anything involving significant investment. As a wage case, reactive systems like malpractice suits or board actions against physicians aren’t easy to navigate if you don’t have many resources.
xyzzyz 29 days ago [-]
When HN was an entrepreneurship oriented community, the overwhelming attitude had been that it’s better to ask for forgiveness than permission. That’s because even if you’re doing something clearly good and morally unimpeachable, having to ask for permission slows you down and invited bikeshedding. Now that HN is a general industry forum, the attitude is more favorable towards preventing risk at a cost of reducing amount of value produced.
perihelions 29 days ago [-]
And yet, FAANG's somehow get way with employing credentialed psychologists and psychiatrists, to experiment on social media users, for maximizing their various engagement and click metrics.
For the scale and impact of non-IRB human psychology research, that one has to be at the very top.
nxobject 29 days ago [-]
I wouldn’t be surprised that, if Facebook was suddenly subject to extra scrutiny, they would just hire an independent IRB in the way that corporate clinical research does, e.g.:
I'm not sure how much this actually happens or if product teams and algorithms figure it out just by trying different things and chasing engagement metrics.
avalys 29 days ago [-]
Cite that this actually happens?
Note that “this person is a psychologist and is employed by Meta working on user research” is not evidence that supports the statement you made.
In what way does this article support the claim that psychologists “experiment on social media users for maximizing their various engagement and click metrics”?
jablongo 29 days ago [-]
My understanding is IRB is not a governmental regulation but an institutional regulation. For example if you want to design a study and then have acme medical center take part in conducting it, you would need to pass the Acme IRB. Governmental funding may also have restrictions that require you to have some kind of IRB review.
pc86 29 days ago [-]
It's quoted in another comment but 42 USC 289(a) requires the establishment of an IRB in any institution applying for federal grants, among other things. So it is a law that there exist some sort of IRB for any "biomedical or behavioral research" and my guess is most institutions see 99% of their research falling under behavioral.
skwb 29 days ago [-]
I mean tons of retrospective studies are literally "IRB exempt" (and highlighted as such in their methods) where no explicit consent is required or needed. Physicians doing case reviews don't need consent of the patient this work. Doing retrospective analyses for defining clinical phenotypes on patient data that has been aggregated isn't needed either. Collecting clinical MRIs to do deep learning doesn't require an IRB.
IRBs are only REALLY required when you are intervening in patient care or pose some theoretical risk to a human.
Some institutions still want you to submit approval for institutional data, but as a non-lawyer it seems that's much more of a CYA policy.
fluidcruft 29 days ago [-]
> Collecting clinical MRIs to do deep learning doesn't require an IRB
Yes it does, but typically the IRB will waive consent and waive notification for those sorts of studies if the images can be de-identified. There's also HIPAA involved which may or may not require establishing a BAA depending on what's being done if the images can't be de-identified. This is particularly an issue with brain MRI because it is usually trivial to generate an image that can be compared to full-face photographs (i.e. can be compared to drivers license/passport type photos to reestablish identify).
And longer term you're not getting anything even in the door at the FDA without an IRB and you're not selling anything without FDA approval.
Also please note that MRI is a Class II regulated device that deposits energy in subjects/patients so it doesn't qualify for a lot of the exemptions (early last year the FDA granted IRBs the ability to do things like wave full written consent for minimal risk research for FDA-regulated research).
skwb 29 days ago [-]
I've worked in MRI AI both in academic and non-academic centers, including technology that has received 510K clearance.
I admit it sounds pedantic, but I'm not discussing IRB *exemptions* that are sometimes required by an institution nor am I discussing BAAs. I was specifically talking about the specific IRB applications (which I've submitted and signed before) that the blog author was talking about. Yes, HIPAA and other state and local regulations also govern what you can and cannot do with the data, but that's not what I argued.
Sorta off topic but the FDA doesn't care so much about SAR unless you're directly programming the MR machine's pulse sequence. If you're just doing quantification of some brain structure for monitoring a biomarker, they're primarily concerned if your product 1) matches an existing prerequisite and 2) functionality that your product achieves performance that you say it does. That is why the marketing around most of the early DL / AI based radiology startups were focused on language for "study prioritization" rather than more specific claims.
fluidcruft 29 days ago [-]
MRI are Class II devices and until last year exemptions for IDE applications and for not obtaining written informed consent specifically excluded devices that by design or intent deposit energy in humans (regardless of overall study risk. Last year the FDA finally extended some power to IRBs to harmonize with HHS research regulations which have allowed IRBs to waive consent for minimal risk studies for a long time. But FDA directing you to an IRB rather than both an IRB and the FDA itself doesn't mean you're not dealing with an IRB.
FDA and human subject protections come from different laws with different legislative authority. The regulations are not the same except to the extent that the agencies themselves work to harmonize things. If you are doing anything covered by FDA you must follow FDA's regulations in addition to any other applicable human subjects research regulations. And because MRI scanners are Class II regulated devices it means that people are being scanned with a doctor's permission, an IRB's permission or the FDA's direct permission.
FDA "doesn't care" about SAR to the extent that they have published guidance that if you operate an approved MRI scanner within normal operating mode (which are settings defined by IEC that do not necessitate medical supervision), then the FDA will not automatically consider use of the scanner itself to elevate a study's risk (in the way that using something like a CT scanner with ionizing radiation would). Risk determination goes beyond whether or not the MRI itself is a risk though. For example a research study that diverts patients to MRI in a way that delays care in an emergent situation (say testing experimental sequences for stroke detection) is unlikely to overall qualify as minimal risk even if the scanner operates in normal mode because of other non-MRI risks associated with the study procedures.
Retrospective use of de-identified or anonymized medical records that already exist are of course a different thing because the risks to the patient are primarily privacy risks.
And you are correct the actual FDA labels of all the AI crap that's coming out are jokes compared to what a lot of sales bullshitters promise. But you better believe all the data submitted to the FDA by the MRI manufacturers support their accelerated acquisitions that use deep learning recons follow FDA's clinical trials regulations.
skwb 29 days ago [-]
I don't disagree with anything you're saying necessarily, but a lot of people are conflating my statement of IRB exemption with having an explicit IRB authorization. I guess more to my point is people seem to be failing to understand the role of retrospective research (and I will easily concede that different institutions have different legal interpretations) and how it's an important part of research. Don't get me wrong, you absolutely still have regulations about what you can and cannot do with that data, but saying you need an IRB for everything doesn't match the reality that I've seen first hand.
That said, there's plenty of buying and selling of radiological images for industry development on the second hand market. Now where the line of "research" vs. "industrial" work is, well that's something I would leave to legal council. But as you said any sort of "altering" of clinical outcomes is a clear IRB is required zone like DL based recon.
ska 29 days ago [-]
This is correct. IRB will be simple is not the same as not needing one.
skwb 29 days ago [-]
I have published multiple papers in peer review journals where we were not covered by IRBs. That is not to say other research ethics standards don't exist, but that IRBs are simply one portion of the healthcare research regulations.
ska 28 days ago [-]
Sure, there are also cases where it is not needed. But some called out in GP were not typically correct. And often it is opaque to the researchers only because someone did an IRB in an earlier step with usage that covers sharing it with you.
sxg 29 days ago [-]
I’m a radiologist, and I’ve worked on these sorts of research studies before. You absolutely do need an IRB to do deep learning on clinical MRIs. I’ve had to write IRBs just to collect retrospective MRIs. Even for studies that are potentially IRB exempt, going through the exemption request takes hours of work on my part and weeks to months before the IRB grants the exception.
As other commenters mention, the “theoretical risk to a human” encompasses nearly all research. For what I do, any imaging studies that aren’t de-identified pose a theoretical risk to patient privacy. If you try to de-identify images, you learn that this is nearly impossible. Sure, you can try your best to scrub DICOM headers, but these headers are mis-used by vendors, so identifying information can appear almost anywhere. You could delete the headers entirely, but then you lose a lot of metadata that you may need to properly display the images. Further, people contend that you can identify individual peoples’ faces if you 3D-reconstruct CT/MR images, so then you have to expend resources to delete faces from all head/neck/brain imaging. Confirming that this was done properly requires manual review and limits the size of your dataset.
Edit: I think the disagreements here are partly due to institutional differences in IRB requirements and partly due to conflating “IRB exemptions” with the idea of not having to interact with the IRB at all. You always have to interact with the IRB—even just to obtain an exemption. While obtaining IRB approval is a cumbersome process, obtaining IRB exemption is only slightly less cumbersome. I’m sure this varies across institutions, but I’ve been at three different large urban academic centers, and obtaining exemptions has been a multi-month process at all three.
potato3732842 29 days ago [-]
>IRBs are only REALLY required when you are intervening in patient care or pose some theoretical risk to a human.
In practice, sure, but that's not what the spaghetti of rules that one is compelled by law to follow state. This disconnect is the whole point of the article.
skwb 29 days ago [-]
I just pointed out the numerous exemptions to IRBs for research. People can literally make a research career and never touch IRBs.
The point I'm trying to make is that the blog is so narrowly defined that it doesn't match the reality of clinical research.
fastaguy88 29 days ago [-]
>IRBs are only REALLY required when you are intervening in patient care or pose some theoretical risk to a human.
Theoretical risk is a very broad category. IRB approval is required to download de-identified human genome sequences, in part because the genome sequence can be used to identify the individual. Drawing blood from an individual, or using blood that has been drawn for other purposes, requires IRB approval. And IRB approval is required for many sociology studies (or surveys), in part because publishing an individuals answers to a survey might put the subject at risk.
So the "theoretical risk" threshold can be very low. It's theoretical after all, so the harm need only be imaginable.
levocardia 29 days ago [-]
No, even IRB exempt studies need to be approved as IRB-exempt by the IRB. Funny isn't it? A lot of other things in your comment are wrong too btw.
lukeinator42 28 days ago [-]
I am in a lab that does EEG research and have recently been doing research using self-supervised learning on EEG. We had to get approval for "secondary use of data" to use open datasets from https://openneuro.org/ and we had to jump through a bunch of hoops with our IRB. It was harder to get it approved than the EEG experiments we run ourselves for some reason.
kazinator 29 days ago [-]
Do you need permission from the government to invite people to your house and let them try food and drinks?
renewiltord 28 days ago [-]
IRBs are like run of the mill CSOs. Ideally, they'd prefer you do zero research since that involves zero risk. Waste of everyone's time. Just like ethicists. Once we clean up funding and reward research we can get rid of all these makework scolds.
photochemsyn 29 days ago [-]
The reason nobody is interested in the independent research efforts of little online community groups is that it's almost certain that they didn't put enough thought into the statistical design of their study, and thus the results are meaningless. Generally professional academics have to submit their research proposal to a funding agency and the statistical design is examined at every stage of the process to ensure that results will have some validity - and even then, it's not unusual for preliminary studies that look promising to be overturned by later more comprehensive studies.
Even with all this professional double-checking, a non-trivial amount of fraud and data manipulation takes place as academics rush to climb the bureaucratic ladder or get large corporations to buy their government-subsidized startups, often making it difficult even for experts in the field to determine what's reliable and what isn't.
29 days ago [-]
motohagiography 29 days ago [-]
the challenge with IRB's is that they are para-bureaucratic orgs without the transparency or accountability of an actual public or government review.
my limited experience with them was to do with governing access to repositories of personal health information and really, just bet that unethical research that abuses powers is going to happen, and that research product gets fed back to policymakers to increase the unaccountable powers of policymakers. any public benefit is for show.
arnonejoe 29 days ago [-]
When I was an undergrad I volunteered for a few different studies conducted by Phd candidates. How much independent research is conducted at universities w/o permission?
asdff 29 days ago [-]
If they took on human subjects it all has to have written approval.
29 days ago [-]
mannyv 28 days ago [-]
Can someone set up an independent IRB that you can use that is more of a UL-style IRB?
(Just so you know: on HN, the convention is to link to other threads only if there are meaningful comments there.)
(Reposts aren't considered dupes unless the previous submission got significant attention within the last year or so - see https://news.ycombinator.com/newsfaq.html)
(It's on my list to implement karma sharing so in the future, multiple submitters can share in the 'win' when a story makes the frontpage)
bell-cot 29 days ago [-]
Is that version of the article meaningfully different?
pinoy420 28 days ago [-]
Medical device research to reduce risk (liability)
63737389 28 days ago [-]
Rico
michaelt 29 days ago [-]
As a foreigner something I've always found hilarious is that on one hand, Americans in academia act like mailing a survey to adults is Performing Research On Humans and it's wildly unethical to do so without IRB approval, informed consent, data control, the right to withdraw at any time and so on.
But at the same time American corporations will gleefully A/B test to determine the most clicked ads, the most profitable search result ordering, and the algorithmic feed parameters that maximise time spent scrolling - all with nary a thought about any ethics whatsoever.
nickff 29 days ago [-]
There's an unfortunate history of scientists harming study subjects, which caused most institutions to institute these safeguards, so as to avoid reputational harm which would cause the government to cease funding them. Corporations are not subject to political whims, though they often do have lawyers review their study techniques, as they are subject to lawsuits.
voxic11 29 days ago [-]
That is because most academics work at institutions which receive federal funding and all human research at such institutions requires IRB approval. American corporations in general do not receive federal funding and so IRB approval is only required in limited circumstances such as research on unapproved medical drugs and procedures.
cess11 29 days ago [-]
Are you sure it's not common for US corporations to have access to public funds, either through subsidies or public investment of some sort?
voxic11 29 days ago [-]
Tax subsidies are received by many corporations but they do not fall under the types of funds covered by the "Common Rule".
> Federal funds administered by a Federal department or agency may not be expended for research involving human subjects unless the requirements of this policy have been satisfied.
Does that somehow make it something else than public funding?
sfpotter 29 days ago [-]
But sometimes, when everyone is having fun, some very mean very bad people show up and say, “HEY! YOU CAN’T DO THAT! THAT’S HUMAN SUBJECTS RESEARCH! YOU NEED TO GET APPROVAL FROM AN INSTITUTIONAL REVIEW BOARD!”
I can't stand this precious style of writing. I wish it were less ubiquitous. I think what upsets me about it is that it's disingenuous. When it registers as "cute and charming", it flies over the reader's head and doesn't encourage them to evaluate for themselves whether something like an IRB (or whatever) should exist. If it comes across as disingenuous---or if the reader doesn't feel like the "very mean very bad people" are "very mean" or "very bad"---then the reader is put in the position of feeling like they're going to rain on someone's parade if they disagree. Basically, it's a rhetorical tactic used to forestall argument or discussion about something the writer finds distasteful. I find myself unwilling to read an article if it opens this way because I don't feel comfortable trusting anything the writer says since they've already opened their piece with a piece of rhetoric which is essentially dishonest.
appleorchard46 29 days ago [-]
It's intended to be an amusingly simplistic strawman in order to contrast with and highlight the actual (deeply) complex reality described thoroughly through the rest of the article.
They don't actually think anyone who suggests IRB approval is very mean and very bad, that line really isn't the thesis statement. If you read a bit further I think you'd be less annoyed by it.
sfpotter 29 days ago [-]
Assuming the author isn't being disingenuous, what does this line actually bring to the table? If the article is thoughtful and nuanced, why not write with the maturity that a thoughtful and nuanced topic deserves? You can do it in a way that invites thought and deliberation and isn't stuffy or overly serious. I just don't see what value interjecting this kind of preciousness is supposed to bring. If it's supposed to be funny... I mean... it isn't.
appleorchard46 28 days ago [-]
If you're really asking what the point of humor is, I don't think there's any answer you'll find satisfying. I'll take a stab at it though.
In this case, it functions as a hook; it's there to give you a slight chuckle before diving into a long and relatively dry informative article. It's some clear sarcasm makes you evaluate the supposed assertion made - that people who push for IRB evaluations are meanie-heads - and challenges any preconceived notions you might have of it being a simple "this is good" or "this is bad" thing either way because of course it's not that simple. This invites the question to the reader: if it's not as simple as that, then how is it actually?
Of course, if the tone doesn't land and you read it as a factual statement - and, importantly, it's so distasteful that you immediately bail and don't read the rest of the article to see what it's actually about, and just assume it's a petulant tirade against IRBs - none of that works. If you're at all interested, I would recommend reading some Adam Mastroianni / Experimental History posts, as it has a very similar feel but plays to the strengths of the tone a bit better (IMO). This one [0] actually touches on the use of humor in scientific papers, and I think makes a pretty good case for why clinical language isn't necessarily a good thing.
It seems you'd answer your own questions here and maybe some of your frustration if you actually proceeded to read the rest of the article, which explains quite clearly the point being made. You being exasperated by the tone or style isn't really relevant to the substance of what was written about, or to anyone else really.
28 days ago [-]
machine_ghost 29 days ago [-]
The fact that the very foundation of our legal system is fundamentally undemocratic (in the sense that no regular citizen can be reasonably expected to understand it) is starting to bubble up into societal awareness.
With everything Trump is doing to destroy the rule of law, and all the potential for technology to help us simplify it, I can only hope that when Trump burns it all down, the people who rebuild will be wise enough to do better.
kubb 29 days ago [-]
I think this might be a missed analogy, rule of law isn't like a house that you can burn down and rebuild stronger.
The default state without rule of law is rule of the strongest. The law is like pillars that support a tunnel. Once these are weakened, the tunnel goes crashing down, and it has to be dug again.
The society backslides to authoritarianism. And the price for digging a new tunnel is often a lot of blood.
c048 29 days ago [-]
I have little faith that, should it all be burned down, recreating it all from the ground up that it would be better.
People need to accept that what Trump is doing is only possible because of what his predecessors put down as law. That means democratic presidents too.
I didn't see Trump's detractors call all of this out when the winds were blowing their way. So no, I have no faith that rebuilding it from the ground up would make it one iota better. Not without deep societal changes.
mmooss 29 days ago [-]
> the people who rebuild
That's you and I.
bell-cot 29 days ago [-]
The article is about medical research on human subjects. And quickly concedes that a great many technically-need-permission things which you might do are of no actual interest to the gov't.
EDIT: <sigh/> The second sentence of the article is -
> Like, let’s eat potatoes and see if we lose weight.
- which, in the minds of 95% of likely jury members, is not "medical research". Nor, at small scale, of plausible interest to the gov't.
If the US Gov't is looking for excuses to crush me, or I've got enough money to be worth a lawsuit, then nothing resembling medical research is needed for them to go ahead with those. Doubtless, I evilly conspired to not pay $0.06 of Use Tax in 2003. And some poor innocent was traumatized for life by my cat looking at them cross-eyed, in a photo I put on the web.
And, notice my phrasing - "of no actual interest to the gov't". Not "IAAL, and you're 100.00% okay to run with this".
PaulHoule 29 days ago [-]
People might still sue you.
I had a stressful event and had a psychogenic fever and lost 15 pounds in two weeks and kept it off for more than six months. There is some research on this [1] [2] in particularly that author wrote a case study about a Japanese girl who was bullied at school who developed a stress fever.
Given there is an obesity crisis and everything this might be an effective diet plan, but to do it, you have to induce severe stress. A protocol might be "get 30 people to show up for a weekend encounter group, pick out the 1 person who seems the most susceptible, have the other 29 bully that person." Cal Lightman might get away with something like that in Lie to Me but no way you can get away with that in real life.
You can only get away with weird social bullying experiments if you're doing it for Youtube.
Don't try crashing a plane for Youtube though, that went down badly with the FAA.
PaulHoule 29 days ago [-]
There was a driver in California who would provoke people on the road and then capture their unhinged reactions on YouTube, editing out the provocation.
Eventually he got in a car crash and got in a lot of trouble.
eadmund 29 days ago [-]
Of no actual interest to the government, unless you are someone that someone in the government wishes to crush. ‘Gosh, Opponent of the Regime, when you wrote your blog entry in 2019 about cutting eggs out of your kids’ diets, that was un-reviewed human research. To jail with you!’
nxobject 29 days ago [-]
Perhaps the issue there is in electing a government that wishes to crush critics by creatively interpreting the law, and not the existence of IRBs.
selimthegrim 29 days ago [-]
This is inconsistent with the general line of the Party, comrade. Your train to Vorkuta awaits.
ls612 29 days ago [-]
Yeah, this post reads like “socially awkward guy discovered the mechanism by which the rule of law is a sham and always has been” to me.
bilbo0s 29 days ago [-]
That's playing with fire.
You could find yourself standing in front of a courtroom if something goes wrong, or God Forbid, if someone dies, trying to convince a judge and jury with nothing but, "I didn't need government permission".
That's not a good place to be.
temp0826 29 days ago [-]
I used to facilitate the use of an alternative medicine (kambo), which could be dangerous if used incorrectly or given to someone with a contraindicated condition. (It has been linked to a few deaths and has recently been banned in Australia). It occurred to me that even a great signed waiver wouldn't save me if something went sideways and decided to stop. I honestly don't think a waiver wouldn't matter in cases like this too.
selimthegrim 29 days ago [-]
see also certain sweat lodge practitioners
handoflixue 29 days ago [-]
> if something goes wrong, or God Forbid, if someone dies
Does emacs improve your typing speed? STOP! That's human experimentation!! That's the point of the article. Why would you worry someone is doing to die testing their typing speed, or filling out an online survey?
You've probably conducts numerous Experimental Human Trials over your lifetime - anyone who is even slightly curious and empirical has!
Xmd5a 29 days ago [-]
He's right you know. I have no opinion on this subject but there has been a crackdown against these kind of activities, especially after covid.
Not accurate. As the article points out, there's a whole category for "benign behavioral interventions" which are exempt from IRB review.
nxobject 29 days ago [-]
> Why would you worry someone is doing to die testing their typing speed, or filling out an online survey?
No.
Because you don’t have to. This seems like a glib answer, but I assure you that, as a university employee in a research unit, I have both tested my typing speed and filled out an online survey. Whataboutism doesn’t help the conversation here become more informed.
I do accounting research. My human-subjects research involves surveying practitioners about their jobs, interviewing them about their experiences, and conducting very simple experiments online that ask them to make decisions. There is virtually 0 real risk to any participants of my studies. IRB does always give my studies "exempt" status, but it still has to be reviewed. And they will pester me about different things like where will store our data (Onedrive is fine, but dropbox is not for some reason). This process will typically take a couple weeks of back and forth.
Yet I have a friend who participated in a Kinesiology dissertation study where they were asked to do extremely strenuous physical activity. He fainted(!) at one point from the activity. And it seemed to them that there were relatively few safeguards from that happening. Now, I'm sure that study did have IRB approval, but it really got me thinking... are we really scrutinizing studies optimally?
How could they do it more efficiently?:
How could they know it's exempt without reviewing it? Should they take researchers' at face value? That seems to undermine the reason we have IRBs, which was unscrupulous researchers. Should we assume they are willing to torture people but not to mislead the IRB?
Why shouldn't data be private? How hard is it?
> a Kinesiology dissertation study where they were asked to do extremely strenuous physical activity. He fainted(!) at one point from the activity. And it seemed to them that there were relatively few safeguards from that happening.
Informed consent is essential to an IRB; your friend would have read and signed something detailing the activities and risks. Strenuous excercise to the point of exhaustion is a part of sports performance research, at least. As long as you inform people, they have the power to opt-out (not prisoners, 6 year olds, etc.), and no undue risk.
Think about how taxes work. The IRS doesn't check every person. You make rules, you perform random audits.
IRBs could easily do the same thing. Set some rules like (simplified) "No IRB necessary if your research doesn't physically touch a person". Researchers will read the rules and skip an IRB if they're in an exempt category. Then just audit at a high enough rate they won't lie.
Here in the Netherlands (we call it HREC, human research ethics committee) the process takes months and months to get permission to talk to people.
Like, “hey I’ve designed some cool new interaction design, I want to get permission to interview people and do human-centered design iterations.” Yes, you will need to spend dozens of hours on forms and wait months.
It cannot be ethical to put these many barriers on talking to humans. And, if you want to do something like an educational game for kids? Or support for the elderly? Or for your students? Well, those are vulnerable populations, so add a few more months.
Education research is extremely difficult here (in the USA it is exempt). How to conduct research on the use of AI in a class you are teaching? You can’t get retroactive consent.
The system is not reasonable.
You have never submitted an IRB.
https://journals.sagepub.com/doi/abs/10.1177/174701611100700...
The author also has a blog about IRBs: http://www.institutionalreviewblog.com/?m=1
Cool, I guess. I wouldn't want to participate if my answers were visible to my employer because honest answers could put my employment at risk.
> As an academic researcher, it's most frustrating in that it feels like IRB scrutiny doesn't seem to align with risk.
As a worker^H^H^H^H^H^Hhuman I find it most frustrating when academics don't understand what things I consider to be risks.
If my answers to your research are leaked, can they be tracked back to me? Onedrive is terrible, just look at Microsoft's repeated flagrant disregard for security. Dropbox is awful too, just look at how easy it is to accidentally a whole folder. Why would you store your data in such places instead of on a research computer with locked down access? That screams to me of a researcher who doesn't understand the value (and risk) of the data they hold. Or perhaps of a researcher who doesn't value that risk appropriately. Perhaps also researcher who doesn't coordinate with their IT department, or an IT department who's equally or worse as competent.
As far as I know, it's possible to get OneDrive even HIIPA compliant, but I don't think so for Dropbox.
https://help.dropbox.com/security/hipaa-hitech-overview
In keeping with the theme of the OP, I don't know if that's been tested in court, soooo... Take that all for what you will.
IRBs are simply a way to say "I had an independent group of ethical and legal authority review my research and they thought it was ok".
So they take it pretty seriously. More so in countries where it's easy to sue, though.
"The Secretary shall by regulation require that each entity which applies for a grant, contract, or cooperative agreement under this chapter for any project or program which involves the conduct of biomedical or behavioral research involving human subjects submit in or with its application for such grant, contract, or cooperative agreement assurances satisfactory to the Secretary that it has established (in accordance with regulations which the Secretary shall prescribe) a board (to be known as an “Institutional Review Board”) to review biomedical and behavioral research involving human subjects conducted at or supported by such entity in order to protect the rights of the human subjects of such research."[1]
To put it succinctly this only relates to any entity which applies for a grant, contract, coop. agreement under chapter 6A of 42 USC.
This in no way relates to private research. It's easy to get hung up on regulation spaghetti code, but regulations cannot exceed the scope of the statutes, which in turn cannot exceed the authority of Congress.
Edit: add source [1] https://www.law.cornell.edu/uscode/text/42/289
But the whole point of the article is that there are other cases when you need IRB approval besides this one. That's one of the myths he's busting.
> The government can’t possibly claim to regulate what me and my roommates eat at home! That would be stupid!” Yes, it would be stupid. But who says the world makes sense? Now, would you actually be prosecuted for violating those rules?
Author says "those rules" without actually citing anything.
There is no scenario in which the Secretary of HHS can create rules which would do what the author was suggesting. It is not even something Congress could do, which is why they haven't created a statute which goes beyond their authority in requiring these boards for anything other than research receiving federal funding. If the Secretary were to create such a rule, it would obviously exceed his authority.
The author actually go into how universities make their own rule that potentially affects teachers and students doing research with their own time and resources.
It's easy for people (especially the kinds of people who gravitate towards engineering) to get all worked up about the necessary implications of the regulations as-written. But it's also important to remember that being facings consequences would require
A) The intentional action of either a state prosecutor or an injured private party B) The involvement of a human being as a judge to interpret the facts and applicable rules
When you look at it from that angle, it's obvious why organizations like META just do the research, imho.
I was once trying to figure out taxes owed on some RSUs granted in California that vested in North Carolina (and some other edge case that I can’t remember). Hoh boy. What a rabbit hole. I didn’t think the CPA that I hired to do my taxes handled it correctly, so I started digging into the tax laws and regulations. Couldn’t find anything that covered my specific case. So I started digging into court cases involving disputes over RSU taxes owed in North Carolina and found conflicting outcomes.
I brought what I had found back to the CPA who basically said something along the lines of: there is no absolute “rule” for what to do in this case like you’re looking for. We make a good faith effort to mirror what was done in similar prior situations, write a number down, and then forget about it unless you hear from the IRS.
Did not make me feel very confident about the legal system.
> In the legal world, however, this assumption does not hold. This is because unlike the laws of nature, political laws are not consistent. The law human beings create to regulate their conduct is made up of incompatible, contradictory rules and principles; and, as anyone who has studied a little logic can demonstrate, any conclusion can be validly derived from a set of contradictory premises. This means that a logically sound argument can be found for any legal conclusion.
The Myth of the Rule of Law:
https://drive.google.com/file/d/1I-JhqpU3_0r_HL06hP-5DABhEtG...
I have also been thinking about how basing morality on following laws is a bad idea. You should not be afraid of breaking laws or rules if they are stupid. All of modern workers rights, women's rights, etc are based on brave people of the past that were not afraid of breaking stupid laws of their time, why should today be any different.
This tends to be something people encounter a lot around tax, complex business structures, and various international dealings where you’re more likely to end up in an edge case that hasn’t been well explored.
Civil law systems (which are used in the non-English-speaking parts of Europe for example) don't have this property. In such systems, court precedents don't really exist, so if something isn't clearly defined by law, and there's no extremely ubiquitous legal interpretation that everybody seems to follow, you don't know whether it's legal or not.
For example, the Polish tax authority lets you ask for a tax interpretation for cases that aren't legally obvious, something that I believe the US would have handled via the court precedent system. It's not unusual for two people in identical situations to receive opposite rulings, and Those rulings are not binding on the authority that issued them.
In other words, you don't know how to handle your tax situation, you ask your tax authority, they tell you "hey, we're the tax authority, we're telling you to do x", you do x, they audit you, they say "this isn't clearly defined in law, we now believe you should have done y instead, here's your orange jumpsuit."
And eventual orange jumpsuit would be decided by the court, not tax authority.
If you're a software engineer and you're knee-deep in analyzing court cases to find out how to file some tax correctly, you're definitely overthinking it. The CPA was exactly correct.
(Hint* You can't)
Yes that is 2 different countries. But it is similar. States have their own laws.
You can't expect the law across 3 jurisdictions to work out every permutation and legislate for it. There are 52 (I guess more... lots of territories) * 51 permutations of moves so 2652 moves that are possible.
In addition you are talking RSUs and not some vanilla thing like claiming an expense.
> So what happens in practice is politicians write a vague law. Bureaucrats turn that law into very detailed (but often still vague) specific rules. Those rules might or might not be “legal”, but nobody want to risk fighting them in court. If the regulations are particularly ridiculous or likely to be overturned if challenged, prosecutors may quietly stop bringing cases. But the regulations still sit there on the books. And people still usually pay attention to them, because why risk it?
Separate rant: I live on the west coast and am annoyed when I vote because of the number of ballot measures. None of them are sufficiently detailed and fall victim to vague language that's open to interpretation. I'd rather "hire" law makers to do the hard work of details law creation.
What I thought was missing was an acknowledgement of the circumstances that would cause a _prosecutor_ or _other litigant_ to decide to _bring an action to court._
Prosecutors are political, have limited resources, and are ultimately more or less accountable to electoral forces. Their prerogative is (almost) never going after every single person who does an action that is technically illegal. Factors like harm, the reputation of one's actions in their community, and their alignment with the political establishment are going to be considerations.
As written, I could see the article maybe deterring a particularly rules-abiding individual from running a potato-eating weight loss experiment with their roommates. I've seen friends and relatives get all wound up worrying about whether something is technically illegal in some marginal way.
Understanding the human factors in the pipeline between a potentially unlawful action and a consequence for it is really important. All the institutions that get anything done are keenly aware of them.
I don't think it's fair to call that a deficiency, though; both the letter of the law and the practical applications are complex and interesting topics in their own right(s?), and either could fill a lengthy article as we see here. Both are welcome! One doesn't preclude the other. I would be interested in reading an article covering what you describe too.
But they have been sued, sometimes for extraordinary amounts, as have their employers, corporate sponsors, etc. And they have lost funding for their research and the trust of everyone above (and of future IRBs). These are powerful disincentives.
Explosives are regulated by the ATF, but can also fall under regulations of the DOT and the EPA. And if your doing any pulse power z pinch or electronic detanator experiment or research, then that's also regulated.
Keep in mind this also has state regulations and ordinance to worry about and its almost like they are trying to get you to accidentally make some mistake that can technically be a felony. And they can visit and do inspections during your hours of operations.
Potato cannons, model rockets, and innovation and experiments in your backyard seem to be impossible these days.
Do I think safety is important, sure, but I'd argue that the paperwork of regulation has stifled the common man who want to follow the laws.
Also see: export control and gpu's
As a random citizen, you'd perhaps get away with an online self-service form, as a small company or startup you might need a form and an interview, but the entire "talk to 4 different agencies at different levels of abstraction" seems like a waste of time for everyone when it can be done in one go (provided the thing you wanted to do is common enough that it doesn't require some extreme documentation and safeguards etc).
If 100000 people were to want to try out a potato canon, it should be sufficient to do something at a local level where you select "I wanna play with a potato canon for a bit" in some drop-down, and it might ask you if you intend to shoot at people or not, and if you have enough room to experiment with this without hurting anyone (including yourself). If yes, problem solved, no need to fill out a form, if unsure or no, fill out a form since you probably are doing something novel, or something you might not want to do. If this then means some other agency at some other level needs to get notified, sure, automate that, who cares. If it turns out you filled out your form and specified radio active potatoes, perhaps one of those other agencies might want to have a chat first to see if everything is okay.
But the default of "talk to an arbitrary number of agencies in every case" is not great.
I’d argue that it’s not the regulations themselves, but people who want to avoid liability, whether ethical or social. (The existence of institutional review is a good argument against, say, a fiscal hawk looking at a state university’s research output to identify the existence of fraud or abuse.)
2. "Human research" means any medical experiments, research, or scientific or psychological investigation, which utilizes human subjects and which involves physical or psychological intervention by the researcher upon the body of the subject and which is not required for the purposes of obtaining information for the diagnosis, prevention, or treatment of disease or the assessment of medical condition for the direct benefit of the subject. Human research shall not, however, be construed to mean the conduct of biological studies exclusively utilizing tissue or fluids after their removal or withdrawal from a human subject in the course of standard medical practice, or to include epidemiological investigations.
They also didn't care about the part where he falsified his results or lied to his sponsors about how he was spending their money.
https://petrieflom.law.harvard.edu/2014/06/29/how-an-irb-cou...
reminded - The EcoHealth Alliance application successfully granted by NIH for those Wuhan coronavirus gain-of-function experiments did have the "Human Subjects Included" and "IRB" checkboxes checked :)
It's been pretty universal in my experience that people will use the power given to them to push their agenda, which is generally only somewhat aligned with the nameplate mission.
E.g. a vegan or vegetarian on an ERB has an obvious, if soft, conflict of interests that will come out in how they treat requests involving animals, or whose results might lead to future testing on animals.
(and most IRB squabbles that involve non-invasion/low risk research never really get as bad as that, at least in my experience, knock on wood)
https://news.ycombinator.com/item?id=43037699
See Scott Alexander's works on IRB review.
Links:
https://slatestarcodex.com/2017/08/29/my-irb-nightmare/
https://slatestarcodex.com/2017/08/31/highlights-from-the-co...
Doesn't that describe all safety and security measures? Airport security wastes millions of hours in order to prevent a few possible problems.
> ethical lapses
That's quite a euphemism for death and bodily injury. I'm sure you wouldn't mind it yourself.
There is plenty of evidence of what happens without IRBs - just look what happened before IRBs.
And yes, airport security is also a colossal waste
What's interesting is how agencies push the limits on what they can do. All agencies do this to one extent or another. And they obviously won't tell you if they're doing that to you or not.
Granted authority and actual authority are different. It's worth remembering that when dealing with any bureaucracy.
You should read the actual regulations as well, but that's work so lots of people don't do it.
Here's an example of why you should read the regulations. Technically, supplements can't make structure/function claims. However, you can submit your structure/function claim to the FDA. Once you do that you can make that particular structure/function claim on your labels. Since the FDA has no process in place to actually review those claims your structure/function claim is suddenly valid.
It's amusing.
We were checking which candy was most popular from a giant bowl, and went classroom to classroom just asking a randomly selected student to pick a piece of candy.
We got 100% on the report, which included "We didn't inform the participants of and potential harm from participating in the study. We're heartless bastards, oh well."
Then I went to college and helped in quite a few study designs with a Professor who had worked with over 30 IRB's across his career. Ours was by far the most strict. We were primarily doing software usability studies, and every single question we would ask a participant needed to be verbatim. Your data anonymization, destruction, and analysis plans had to be fully defined, including storing the participant name to UID key separately from the data.
Deviation or asking follow-up questions based on a prior response was either not allowed or a whole additional huge layer of SCRUTINY (I'm not sure). I'm genuinely curious how any sort of therapy focused study could have occurred with that IRB.
I was funded through a DARPA grant. It was definitely not a training exercise.
He did extra work, but it was effectively practice.
Because we're about to see how all the structures that were put in place to enforce those ethics react to being told they must instead do keyword searches for forbidden phrases set by fiat.
If you're the kind of administrator who does boring work of making sure all professors in your biomedical campus submitted IRB paperwork, are you also going to be the person who makes sure their papers don't contain the word "Woman" or "Historically"?
In contrast, a shockingly large amount of content is extremely lazy, gut-feeling reaction. Avoiding the work and disappointment of finding out that your hot take is wrong.
"We" decided that Tuskegee was bad enough that it should be stopped before harm is done, and that there is no appropriate or sufficient "punish[ment] in retrospect" for the fallout.
The government makes you get a license to drive at all, then "drive a Pinto" versus "drive a Trabant" are similar enough that they don't require more info. They require you to get different licensure to drive a bigger truck where you could potentially cause more harm, or to drive an airplane. In this analogy the IRB is the DMV/FAA/whatever, and you're asking for permission to drive a tank, a motorized unicycle, a helicopter, an 18-wheeler or a stealth fighter. You don't get a Science License rubber stamp because that's like getting a Vehicle License - the variation in "Vehicle" is big enough that each type needs review.
The thing is, although you and the linked article seem to be associating IRB approval just with human studies, these days you need it for mouse studies.
[1] example: https://animalcare.umich.edu/institutional-animal-care-use-c...
(I guess the point of analogies like these are to force us to sweat the details and examples.)
There are escape hatches, too: I doubt many rural Alaskan pilots worry (or need to worry) about these things.
A driver's license is more like a medical license than IRB approval.
If your point is that we could replace IRBs with some sort of a researcher license, that you need to obtain before being able to do studies that today require IRB approval, then I support it, because while not ideal, it improves over the status quo.
For the scale and impact of non-IRB human psychology research, that one has to be at the very top.
https://www.sterlingirb.com/ https://www.pearlirb.com/
Note that “this person is a psychologist and is employed by Meta working on user research” is not evidence that supports the statement you made.
IRBs are only REALLY required when you are intervening in patient care or pose some theoretical risk to a human.
Some institutions still want you to submit approval for institutional data, but as a non-lawyer it seems that's much more of a CYA policy.
Yes it does, but typically the IRB will waive consent and waive notification for those sorts of studies if the images can be de-identified. There's also HIPAA involved which may or may not require establishing a BAA depending on what's being done if the images can't be de-identified. This is particularly an issue with brain MRI because it is usually trivial to generate an image that can be compared to full-face photographs (i.e. can be compared to drivers license/passport type photos to reestablish identify).
And longer term you're not getting anything even in the door at the FDA without an IRB and you're not selling anything without FDA approval.
Also please note that MRI is a Class II regulated device that deposits energy in subjects/patients so it doesn't qualify for a lot of the exemptions (early last year the FDA granted IRBs the ability to do things like wave full written consent for minimal risk research for FDA-regulated research).
I admit it sounds pedantic, but I'm not discussing IRB *exemptions* that are sometimes required by an institution nor am I discussing BAAs. I was specifically talking about the specific IRB applications (which I've submitted and signed before) that the blog author was talking about. Yes, HIPAA and other state and local regulations also govern what you can and cannot do with the data, but that's not what I argued.
Sorta off topic but the FDA doesn't care so much about SAR unless you're directly programming the MR machine's pulse sequence. If you're just doing quantification of some brain structure for monitoring a biomarker, they're primarily concerned if your product 1) matches an existing prerequisite and 2) functionality that your product achieves performance that you say it does. That is why the marketing around most of the early DL / AI based radiology startups were focused on language for "study prioritization" rather than more specific claims.
FDA and human subject protections come from different laws with different legislative authority. The regulations are not the same except to the extent that the agencies themselves work to harmonize things. If you are doing anything covered by FDA you must follow FDA's regulations in addition to any other applicable human subjects research regulations. And because MRI scanners are Class II regulated devices it means that people are being scanned with a doctor's permission, an IRB's permission or the FDA's direct permission.
FDA "doesn't care" about SAR to the extent that they have published guidance that if you operate an approved MRI scanner within normal operating mode (which are settings defined by IEC that do not necessitate medical supervision), then the FDA will not automatically consider use of the scanner itself to elevate a study's risk (in the way that using something like a CT scanner with ionizing radiation would). Risk determination goes beyond whether or not the MRI itself is a risk though. For example a research study that diverts patients to MRI in a way that delays care in an emergent situation (say testing experimental sequences for stroke detection) is unlikely to overall qualify as minimal risk even if the scanner operates in normal mode because of other non-MRI risks associated with the study procedures.
Retrospective use of de-identified or anonymized medical records that already exist are of course a different thing because the risks to the patient are primarily privacy risks.
And you are correct the actual FDA labels of all the AI crap that's coming out are jokes compared to what a lot of sales bullshitters promise. But you better believe all the data submitted to the FDA by the MRI manufacturers support their accelerated acquisitions that use deep learning recons follow FDA's clinical trials regulations.
That said, there's plenty of buying and selling of radiological images for industry development on the second hand market. Now where the line of "research" vs. "industrial" work is, well that's something I would leave to legal council. But as you said any sort of "altering" of clinical outcomes is a clear IRB is required zone like DL based recon.
As other commenters mention, the “theoretical risk to a human” encompasses nearly all research. For what I do, any imaging studies that aren’t de-identified pose a theoretical risk to patient privacy. If you try to de-identify images, you learn that this is nearly impossible. Sure, you can try your best to scrub DICOM headers, but these headers are mis-used by vendors, so identifying information can appear almost anywhere. You could delete the headers entirely, but then you lose a lot of metadata that you may need to properly display the images. Further, people contend that you can identify individual peoples’ faces if you 3D-reconstruct CT/MR images, so then you have to expend resources to delete faces from all head/neck/brain imaging. Confirming that this was done properly requires manual review and limits the size of your dataset.
Edit: I think the disagreements here are partly due to institutional differences in IRB requirements and partly due to conflating “IRB exemptions” with the idea of not having to interact with the IRB at all. You always have to interact with the IRB—even just to obtain an exemption. While obtaining IRB approval is a cumbersome process, obtaining IRB exemption is only slightly less cumbersome. I’m sure this varies across institutions, but I’ve been at three different large urban academic centers, and obtaining exemptions has been a multi-month process at all three.
In practice, sure, but that's not what the spaghetti of rules that one is compelled by law to follow state. This disconnect is the whole point of the article.
The point I'm trying to make is that the blog is so narrowly defined that it doesn't match the reality of clinical research.
Theoretical risk is a very broad category. IRB approval is required to download de-identified human genome sequences, in part because the genome sequence can be used to identify the individual. Drawing blood from an individual, or using blood that has been drawn for other purposes, requires IRB approval. And IRB approval is required for many sociology studies (or surveys), in part because publishing an individuals answers to a survey might put the subject at risk.
So the "theoretical risk" threshold can be very low. It's theoretical after all, so the harm need only be imaginable.
Even with all this professional double-checking, a non-trivial amount of fraud and data manipulation takes place as academics rush to climb the bureaucratic ladder or get large corporations to buy their government-subsidized startups, often making it difficult even for experts in the field to determine what's reliable and what isn't.
my limited experience with them was to do with governing access to repositories of personal health information and really, just bet that unethical research that abuses powers is going to happen, and that research product gets fed back to policymakers to increase the unaccountable powers of policymakers. any public benefit is for show.
(Reposts aren't considered dupes unless the previous submission got significant attention within the last year or so - see https://news.ycombinator.com/newsfaq.html)
(It's on my list to implement karma sharing so in the future, multiple submitters can share in the 'win' when a story makes the frontpage)
But at the same time American corporations will gleefully A/B test to determine the most clicked ads, the most profitable search result ordering, and the algorithmic feed parameters that maximise time spent scrolling - all with nary a thought about any ethics whatsoever.
> Federal funds administered by a Federal department or agency may not be expended for research involving human subjects unless the requirements of this policy have been satisfied.
https://www.ecfr.gov/current/title-45/subtitle-A/subchapter-...
I can't stand this precious style of writing. I wish it were less ubiquitous. I think what upsets me about it is that it's disingenuous. When it registers as "cute and charming", it flies over the reader's head and doesn't encourage them to evaluate for themselves whether something like an IRB (or whatever) should exist. If it comes across as disingenuous---or if the reader doesn't feel like the "very mean very bad people" are "very mean" or "very bad"---then the reader is put in the position of feeling like they're going to rain on someone's parade if they disagree. Basically, it's a rhetorical tactic used to forestall argument or discussion about something the writer finds distasteful. I find myself unwilling to read an article if it opens this way because I don't feel comfortable trusting anything the writer says since they've already opened their piece with a piece of rhetoric which is essentially dishonest.
They don't actually think anyone who suggests IRB approval is very mean and very bad, that line really isn't the thesis statement. If you read a bit further I think you'd be less annoyed by it.
In this case, it functions as a hook; it's there to give you a slight chuckle before diving into a long and relatively dry informative article. It's some clear sarcasm makes you evaluate the supposed assertion made - that people who push for IRB evaluations are meanie-heads - and challenges any preconceived notions you might have of it being a simple "this is good" or "this is bad" thing either way because of course it's not that simple. This invites the question to the reader: if it's not as simple as that, then how is it actually?
Of course, if the tone doesn't land and you read it as a factual statement - and, importantly, it's so distasteful that you immediately bail and don't read the rest of the article to see what it's actually about, and just assume it's a petulant tirade against IRBs - none of that works. If you're at all interested, I would recommend reading some Adam Mastroianni / Experimental History posts, as it has a very similar feel but plays to the strengths of the tone a bit better (IMO). This one [0] actually touches on the use of humor in scientific papers, and I think makes a pretty good case for why clinical language isn't necessarily a good thing.
[0] https://www.experimental-history.com/p/the-rise-and-fall-of-...
With everything Trump is doing to destroy the rule of law, and all the potential for technology to help us simplify it, I can only hope that when Trump burns it all down, the people who rebuild will be wise enough to do better.
The default state without rule of law is rule of the strongest. The law is like pillars that support a tunnel. Once these are weakened, the tunnel goes crashing down, and it has to be dug again.
The society backslides to authoritarianism. And the price for digging a new tunnel is often a lot of blood.
People need to accept that what Trump is doing is only possible because of what his predecessors put down as law. That means democratic presidents too.
I didn't see Trump's detractors call all of this out when the winds were blowing their way. So no, I have no faith that rebuilding it from the ground up would make it one iota better. Not without deep societal changes.
That's you and I.
EDIT: <sigh/> The second sentence of the article is -
> Like, let’s eat potatoes and see if we lose weight.
- which, in the minds of 95% of likely jury members, is not "medical research". Nor, at small scale, of plausible interest to the gov't.
If the US Gov't is looking for excuses to crush me, or I've got enough money to be worth a lawsuit, then nothing resembling medical research is needed for them to go ahead with those. Doubtless, I evilly conspired to not pay $0.06 of Use Tax in 2003. And some poor innocent was traumatized for life by my cat looking at them cross-eyed, in a photo I put on the web.
And, notice my phrasing - "of no actual interest to the gov't". Not "IAAL, and you're 100.00% okay to run with this".
I had a stressful event and had a psychogenic fever and lost 15 pounds in two weeks and kept it off for more than six months. There is some research on this [1] [2] in particularly that author wrote a case study about a Japanese girl who was bullied at school who developed a stress fever.
Given there is an obesity crisis and everything this might be an effective diet plan, but to do it, you have to induce severe stress. A protocol might be "get 30 people to show up for a weekend encounter group, pick out the 1 person who seems the most susceptible, have the other 29 bully that person." Cal Lightman might get away with something like that in Lie to Me but no way you can get away with that in real life.
[1] https://pubmed.ncbi.nlm.nih.gov/27227051/
[2] https://pubmed.ncbi.nlm.nih.gov/30459027/
Don't try crashing a plane for Youtube though, that went down badly with the FAA.
Eventually he got in a car crash and got in a lot of trouble.
You could find yourself standing in front of a courtroom if something goes wrong, or God Forbid, if someone dies, trying to convince a judge and jury with nothing but, "I didn't need government permission".
That's not a good place to be.
Does emacs improve your typing speed? STOP! That's human experimentation!! That's the point of the article. Why would you worry someone is doing to die testing their typing speed, or filling out an online survey?
You've probably conducts numerous Experimental Human Trials over your lifetime - anyone who is even slightly curious and empirical has!
https://en.wikipedia.org/wiki/Thierry_Casasnovas
No.
Because you don’t have to. This seems like a glib answer, but I assure you that, as a university employee in a research unit, I have both tested my typing speed and filled out an online survey. Whataboutism doesn’t help the conversation here become more informed.