HR1313 - Workplace Collection of Genetic Information

[Trebla]

HR1313

Relevant section (emphasis mine):

(b) Collection Of Information.—Notwithstanding any other provision of law, the collection of information about the manifested disease or disorder of a family member shall not be considered an unlawful acquisition of genetic information with respect to another family member as part of a workplace wellness program described in paragraph (1) or (2) offered by an employer (or in conjunction with an employer-sponsored health plan described in section 2705(j) of the Public Health Service Act (42 U.S.C. 300gg–4(j))) and shall not violate title I or title II of the Genetic Information Nondiscrimination Act of 2008 (Public Law 110–233). For purposes of the preceding sentence, the term “family member” has the meaning given such term in section 201 of the Genetic Information Nondiscrimination Act (Public Law 110–233).

This doesn't seem to be making massive headlines in any of the major outlets, I'm not sure why. Or maybe I'm just searching wrong. Least biased article I could find...? Snopes

TL;DR - Workplaces can "require" employees to submit to DNA/Genetic testing. It is not mandatory, but workers can be penalized (or, "fail to receive incentives" if you want to quibble over syntax) if they refuse to participate.

[strike]This passed committee along party lines (22 R for, 17 D against).[/strike] Edit: Not sure where I saw this, when I look at the bill, all I see is "Referred to committee", I apologize (a) if this is incorrect and (b) that I can't find strikethrough tags...

I don't think this could possibly pass, but it's pretty appalling that it's a thing at all. I won't pretend to understand exactly how this interacts with all the other Acts and Laws cited, maybe it's much ado about nothing.

[Dauric]

In a way it's worse than that. THe part you bolded says that if you have a relative that comes down with some kind of genetically related disease, then either their wellness program, - or the employer provided health insurance- can demand the genetic screening information of that relative.

IE: your sibling works for Corporation A, you are diagnosed with a genetic malady of any severity (from full blown cancer to edge-case diabetes), and your sibling's company and/or insurance provider gets to know about it. You don't even work for that company, you don't have insurance with that insurer, but they get to demand access to -your- medical records and it's legal.

[Opus_723]

Well, shit. I just had to apologize to my wife for teasing her years ago when we watched Gattaca and she was concerned that the future would actually work like that.

I would suggest that Democrats nickname this the Gattaca Bill to demonize it, but that's actually a really cool name and would totally backfire.

[Chen]

This Vox article seems to go into a bit more detail on it: http://www.vox.com/policy-and-politics/ ... nformation

I'm not sure how the family part would work though. Take the situation Dauric mentioned with two siblings. If the sibling working for the company simply filled out a questionnaire and answered questions like "does anyone in your family have X disease" it's probably legal. There's no way the company could actually get the non-workers medical records since the worker in question has no way to obtain said records in the first place to provide to the company (barring having power of attorney and the like I guess). If the law as written somehow allows that it will get thrown out ridiculously fast.

[Dauric]

It's subtle but after reading the Vox article I see how it works. It allows health insurance or workplace health programs to provide 'discounts' (pretty steep ones at %30) if you provide the information. Technically you're not 'required' to provide that data, you provide it 'voluntarily' and in return you get a 'discount'.

Normally these requests would be illegal to even ask for much less receive, but this bill removes that restriction.

Again the hack is that providing the data is technically "voluntary" even though the economic incentives to do so are significant enough that they could border on duress.

[Chen]

30% is the existing discount value, since this program is already in place with the ACA, though it restricts them from asking for the genetic stuff. The new bill allows the genetic stuff AND increases the discount to 50% if what I read elsewhere was correct.

I'm torn on this whole thing much in the same way as those GPS incentives for car insurance. If you can legitimately somehow show that the price you get is in fact discounted, rather than them just jacking up the price and then giving you the "discount" down to the old normal price, I'm ok with it. I'm not sure I'd necessarily do it myself, just due to the possible downsides, but choosing to get money up front in exchange for possible future downsides I'm ok with in principle. The risk to me seems to be the other option I mentioned. All the prices increase and the "discount" is really just the old price. Car insurance is at least easy to shop around for. Health insurance is much more difficult when its through your employer and its much less transparent in why the costs change for it.

[The Great Hippo]

I'm torn on this whole thing much in the same way as those GPS incentives for car insurance. If you can legitimately somehow show that the price you get is in fact discounted, rather than them just jacking up the price and then giving you the "discount" down to the old normal price, I'm ok with it.

I'm not sure I see the distinction between a discount and a penalty; regardless of whether or not they stick with the 'old price', you're ultimately asking people who don't want to share genetic information to pay more for the same thing.

Besides, the old price is always going to change in response to the market, isn't it? And that discount is part of the market: If I offer a $1 coupon for my product, it seems natural to expect that -- as more people use my coupon -- I will increase the price of my product. Maybe it isn't that simple on account of this being such a complex situation, though? Health insurance is weird.

[Chen]

I'm not sure I see the distinction between a discount and a penalty; regardless of whether or not they stick with the 'old price', you're ultimately asking people who don't want to share genetic information to pay more for the same thing.

We presumably the "discount" is given by the insurance company to your employer because they better know how to assess the risk each person poses if they have their genetic history. It's not clear what someone with a bad genetic history would get here though. Presumably the employer would give them the discount, but their rate should be significantly higher due to their genetic history.

Much in the way the car insurance companies want those GPS devices to monitor your car and reduce rates accordingly. I presume if after a year they see you're constantly speeding or otherwise driving recklessly, you're probably going to lose that savings. A sick person giving their genetic history would be largely counter-productive it would seem to me.

[The Great Hippo]

Isn't there a big difference between penalizing someone for their behavior ("stop speeding!") versus penalizing someone for their genetics ("stop being at-risk for heart-attacks!"), though? I know that in a lot of ways, we've moved past the whole nature-vs-nurture debate -- but there's still something deeply unsettling about being penalized for something you have literally no control over.

A sick person giving their genetic history would be largely counter-productive it would seem to me.

Right, and it's even creepier because you might have had no idea you were genetically predisposed toward a certain illness. Imagine finding it all out at once:

"Oh, hey, we got your genetic tests back; turns out you're probably going to die of this terrible disease in ten years. Also, on a completely unrelated note, we've decided you don't qualify for our employee life insurance program and we're increasing your medical coverage rates."

EDIT: Or even worse: "We got your genetic test back; thanks for agreeing to it! Here's your discount. On a completely unrelated note, we've decided you don't qualify for our employee life insurance program and we're increasing your medical coverage rates."

[LaserGuy]

I'm amused that I just got back from a conference where this topic came up at a few of the talks, and all of the doctors and geneticists were basically saying "Genetic tests for health screening have fairly poor predictive strength of future illness except in very specific cases. Family history is almost always more reliable." What genetic testing is good for is, one you've been diagnosed with something, seeing if there is a genetic component to the disease, or a pathway that suggests a particular treatment course.

[idonno]

I'm amused that I just got back from a conference where this topic came up at a few of the talks, and all of the doctors and geneticists were basically saying "Genetic tests for health screening have fairly poor predictive strength of future illness except in very specific cases. Family history is almost always more reliable." What genetic testing is good for is, one you've been diagnosed with something, seeing if there is a genetic component to the disease, or a pathway that suggests a particular treatment course.

I could see that changing pretty rapidly with the ongoing success of machine learning and a data set containing medical records and genetic information for the majority of the nations work force and their families.

[Liri]

I'm amused that I just got back from a conference where this topic came up at a few of the talks, and all of the doctors and geneticists were basically saying "Genetic tests for health screening have fairly poor predictive strength of future illness except in very specific cases. Family history is almost always more reliable." What genetic testing is good for is, one you've been diagnosed with something, seeing if there is a genetic component to the disease, or a pathway that suggests a particular treatment course.

I could see that changing pretty rapidly with the ongoing success of machine learning and a data set containing medical records and genetic information for the majority of the nations work force and their families.

There's more to disease risk than simple genetics. Running genome-wide, or even targeted, methylation arrays is hugely impractical - for the time being. I went to a talk just a couple months ago about the potential use of methylation profiles as a biomarker for remission and relapse in a specific kidney disease. The research has been going on for years and they're only vaguely closer. Teasing out risk from circular interactions between SNPs (single nucleotide polymorphisms) (aka genetic differences), methylation, and chromatin is quite difficult, and isn't close to being able to be used as a diagnostic or predictive tool. The correlations people get for methylation/chromatin state and what actually gets transcribed are usually just on the verge of significance, if significant at all. There's a lot we don't know.