4 ms·

This is going to be disagreeable, but as someone who was blocked from starting a COVID surveillance program at my uni due to the IRB constraints on "human subje

by mjfl 5y ago

This is going to be disagreeable, but as someone who was blocked from starting a COVID surveillance program at my uni due to the IRB constraints on "human subjects" - would need to buy special, expensive software in order to prevent test results from being 'deanonymized' - I think this space is too regulated already. In some cases, people with apparently legitimate ethical issues should have been told to shove it, in my opinion. Innovation is being destroyed to protect people's feelings.

et2o 5y ago
Could you post your study proposal and the IRB response? This does not seem realistic.
shadowgovt 5y ago
> Innovation is being destroyed to protect people's feelings. Yes, it is. This is not a bad thing. A lot of the calculuses that conclude it's a bad thing are fundamentally dehumanizing.
- mjfl 5y ago
  Calculus: Having campus-wide COVID surveillance running (before any alternative testing options had sprung up) is more important than preventing ANY RISK of someone experiencing stigma for having had COVID. I think this calculus holds up and is not that dehumanizing. I'm not saying we should get rid of all protections, but that pendulum is too far on the regulatory side right now.
  - shadowgovt 5y ago
    In that case, just cut out the middle man and start reading their medical records. Who needs privacy? (... And then people rationally respond to such an invasion by no longer going to the doctors and lying about their medical history...) We can't interpret people's desire for privacy as damage and route around it. There's decades of history of scientists interacting with the public to show why that has negative consequences for society.
DavidPeiffer 5y ago
I'm with the other reply, genuinely curious what was submitted for review. I'm glad you were prevented from starting a study without the fundamentals in place. I've been on the IRB submission side for Iowa State, and I can tell you it was a rigurous process, though the training emphasized why things are the way they are. If participants are concerned about their results becoming deanonymized, the basic trust we have that there will not be negative ramifications for the participant and that participants will be genuine in their responses and participation go out the window. That's a fundamental aspect of conducting good research that is valid and as broadly applicable as possible. Deanonymized results may not seem like a big deal, however we have to also consider future impacts on the participant. What if discrimination based on that test result was permissible and they couldn't find a job? What if later analysis shows conclusions we couldn't imagine right now that would hurt the participant? What if your boss found out you were participating in a drug trial for PrEP, outing you as likely gay? What if social norms change and new standards are applied to old data? I'm also curious about the expensive software they were wanting. The standards shift over time, but in my experience we might have a paper key for participants to participant number until _x_ months after the study is completed, at which time it would be destroyed. The key would probably only exist on paper and would be stored separately from the data, behind a separate differently keyed locks (or encrypted separately if electronic). All of this would be explicitly defined in the documents provided to the IRB. For surveys, Qualtrics was an approved vendor, so we would often try to do all data collection in Qualtrics.

by mjfl 5y ago

et2o 5y ago

Could you post your study proposal and the IRB response? This does not seem realistic.

shadowgovt 5y ago

> Innovation is being destroyed to protect people's feelings. Yes, it is. This is not a bad thing. A lot of the calculuses that conclude it's a bad thing are fundamentally dehumanizing.

mjfl 5y ago
Calculus: Having campus-wide COVID surveillance running (before any alternative testing options had sprung up) is more important than preventing ANY RISK of someone experiencing stigma for having had COVID. I think this calculus holds up and is not that dehumanizing. I'm not saying we should get rid of all protections, but that pendulum is too far on the regulatory side right now.
- shadowgovt 5y ago
  In that case, just cut out the middle man and start reading their medical records. Who needs privacy? (... And then people rationally respond to such an invasion by no longer going to the doctors and lying about their medical history...) We can't interpret people's desire for privacy as damage and route around it. There's decades of history of scientists interacting with the public to show why that has negative consequences for society.

DavidPeiffer 5y ago

I'm with the other reply, genuinely curious what was submitted for review. I'm glad you were prevented from starting a study without the fundamentals in place. I've been on the IRB submission side for Iowa State, and I can tell you it was a rigurous process, though the training emphasized why things are the way they are. If participants are concerned about their results becoming deanonymized, the basic trust we have that there will not be negative ramifications for the participant and that participants will be genuine in their responses and participation go out the window. That's a fundamental aspect of conducting good research that is valid and as broadly applicable as possible. Deanonymized results may not seem like a big deal, however we have to also consider future impacts on the participant. What if discrimination based on that test result was permissible and they couldn't find a job? What if later analysis shows conclusions we couldn't imagine right now that would hurt the participant? What if your boss found out you were participating in a drug trial for PrEP, outing you as likely gay? What if social norms change and new standards are applied to old data? I'm also curious about the expensive software they were wanting. The standards shift over time, but in my experience we might have a paper key for participants to participant number until _x_ months after the study is completed, at which time it would be destroyed. The key would probably only exist on paper and would be stored separately from the data, behind a separate differently keyed locks (or encrypted separately if electronic). All of this would be explicitly defined in the documents provided to the IRB. For surveys, Qualtrics was an approved vendor, so we would often try to do all data collection in Qualtrics.