Paper about Herbalife-related patient death removed after legal threats | Hacker News Reader

Paper about Herbalife-related patient death removed after legal threats

scienceintegritydigest.com

444 pointscomputer2376 comments

This happened back in the 1970s when some researchers tried to get follow-up information about whether alcoholics can moderately drink again.

Here’s what happened: The paper Sobell 1973 made the bold claim that alcoholics could moderately drink again if given the right kind of therapy. So, other researchers asked: What was the long-term outcome with those patients supposedly engaging on “controlled drinking”?

So this research team including one Pendery asked Sobell for the list of patients so they could perform a proper follow-up. The Sobell researchers refused to provide the list. The Pendery team managed to get the list anyway. The Sobells then sued Pendary and their team to stop them from using the list.

After some litigation, the Pendary team got the Sobell lawsuit dismissed and then followed up with those supposedly moderate drinking patients.

Out of the 20 patients the Sobell study claimed were moderately drinking, only one arguably was still moderately drinking a decade later. Four of those “moderate drinkers” died from drinking too much. Eight were engaging in out of control drinking. Six were completely abstinent from alcohol. And one could not be found, but appeared to be “gravely disabled”.

No wonder the Sobells tried to block the follow up study with lawsuits: Because it showed that those “moderate drinkers” were no longer moderately drinking.

After a lot of effort, the Sobell lawsuit(s) were dismissed and the truth won.

References: https://www.nytimes.com/1982/11/02/science/showdown-nears-in...

https://www.nytimes.com/1982/06/28/us/alcholism-study-under-...

> After a lot of effort, the Sobell lawsuit(s) were dismissed and the truth won.

The truth? Especially in anything related to alcohol/substance abuse, there most often is no objective "truth", as there are a lot of religious / moralizing standpoints involved. Alcoholics Anonymous, for example, has been shown to be more morals than science: https://www.theatlantic.com/magazine/archive/2015/04/the-irr...

What does help however is "housing first": https://www.samhsa.gov/homelessness-programs-resources/hpr-r...

Kinda makes sense, given that many people on the streets basically self-medicate either existing untreated mental health problems or the (enormous) stress that comes with living on the streets.

> The truth? Especially in anything related to alcohol/substance abuse, there most often is no objective "truth"

I dunno, sounds pretty objective to me:

> Out of the 20 patients the Sobell study claimed were moderately drinking, only one arguably was still moderately drinking a decade later.

Of course, to conclude from a 20 person study that moderate drinking can never work is another thing. But simply ignoring the study can't be the solution either.

> What does help however is "housing first"

This is probably true, but you didn't give any empirical evidence for that either. Also, not all alcoholics are homeless.

That article from The Atlantic uses outdated information. The figures Glaser used for AA efficacy do not come from peer reviewed science and are inaccurate. See https://www.thecut.com/2015/03/why-alcoholics-anonymous-work... for a direct rebuttal of that 2015 article.

The current science shows that Alcoholics Anonymous is an incredibly effective treatment for many alcoholics. Indeed, the 2020 Cochrane Review on AA shows that Alcoholics Anonymous is more effective then other treatments in getting alcoholics abstinent. See https://www.nytimes.com/2020/03/11/upshot/alcoholics-anonymo... or https://www.wbur.org/commonhealth/2020/03/11/medical-science...

Keep in mind that Cochrane reviews are the golden standard for high quality meta analysis of science, as pointed out at https://en.wikipedia.org/wiki/Wikipedia:MEDDATE

It's nice that the WBUR link includes a link to the review. Here it is for people who want it directly:

https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD...

The full 2020 Cochrane review on AA is paywalled until March, so I’m linking to media summaries of the review until it becomes open access.

What am I missing please?

https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD...

When clicking on the link multiple times, Cochrane no longer gives access to the full report.

It will become non-paywalled (or should I say, non-semi-paywalled) come March over at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7065341/

AA works, but it likely works more because of the community than anything else. Drug addiction is often a symptom of isolation. Drugs often mimic the brain chemicals we get through positive social interaction, serving as a placebo.

You are kind of right. Being a part of a community all going through the same struggle definitely helps. Just not being around people getting wasted is a big help in itself. You can't just take a large chunk of your life away, it must be replaced with something.

The part about helping others is a big factor. It gives a reason other than self, which seems to be a big factor.

It's hard to say what makes it "work" for certain. The Oxford Group, the predecessor of AA, lacked the helping others aspect and did not see widespread success. This is what AA itself considers to be their unique magic.

Or maybe it's all the sayings! God, there are so many sayings.

Despite being very hostile initially, I got a lot of benefit from 12 step programs. It is not something I have stayed formally involved with, but I got a lot of helpful insight.

I don't think any one thing did it for me, it was a long and difficult process. The fact I'm alive and stable today is a small miracle. Most people in the state I was aren't so lucky.

I know what doesn't work though: Shame. I had to stop shaming myself before I could get well.

Maybe I'm misreading your comment, but I think you are mixing two different issues.

What about alcoholics who already have a home? "Housing first" won't help them, while AA arguably would.

Here's the English NICE advice about alcohol use disorders.

https://www.nice.org.uk/guidance/cg115

The evidence tab is useful to see what information they used to make the decision: https://www.nice.org.uk/guidance/cg115/evidence

Here’s the 2016 (i.e. done during the Obama administration) report from the US Surgeon General on some resources which help with recovery: https://www.ncbi.nlm.nih.gov/books/NBK424846/

Looking over at that Nice website, it looks like the Nice reports haven’t taken the information the 2020 Cochrane review on AA used in to account yet.

>No wonder the Sobells tried to block the follow up study with lawsuits

It seems like there would be a totally reasonable concern about privacy here as well. If I participated in a study with finite scope, I wouldn't want my identity and medical information transferred to another group of researchers.

The subjects were delighted (Edit: See below for actual quote from one of the Pendery team researchers) to be contacted by another research group, and some of them felt the treatment they got in the Sobell group was so shoddy, they ended up trying to sue the Sobell researchers.

Reference: https://www.nytimes.com/1984/09/11/science/panel-finds-no-fr...

>The subjects were delighted to be contacted by another research group

Where did you get that they were delighted from the article you linked? Were they all uniformly delighted?

It seems that either way, the original agreement/contract/TOS should govern what can be done with personal medical information.

The reference was for the fact some of the original patients tried to sue the Sobells.

The “delighted” bit comes from Alcoholism: A Review of its Characteristics, Etiology, Treatments, and Controversies by Irvine Maltzman, which goes in to the Sobell controversy in great detail.

Edit: Going back to the book, here’s the most relevant quote from the late Maltzman: “the patients contacted prior to the court injunction all expressed a willingness to cooperate and to be interviewed” Maybe “delighted” was too strong of a word.

Thanks for clarifying.

Yet in society as a whole, the desire to prevent more people going down this "moderate drinking" path and ruining their lives seems far more important than the need to prevent 20 people getting a phone call saying "Can I ask some followup questions?".

I feel like using someone's private data for the purpose of asking if they want to join a follow-up study should always be acceptable.

I feel like using someone's personal data for any purpose other than the one for which it was collected should never be allowable. Luckily I live in the EU, where this would already be unlawful.

Imagine it was discovered 5 years after the initial study that the medicine given during it was harmful and there are long term health consequences. The discovery was made by another team. Wouldn't you like to know?

Another example: a study for rare condition was conducted. No effective treatment was found. A few years later very promising treatment was discovered. Wouldn't you like to know, even if the original form didn't ask your permission?

The easy way to remedy this is that all forms should ask for that permission explicitly, and many do. When doing an MSc in neuroscience, even though our research was about the contribution of early visual areas to reading tasks, we always asked permission to get in contact if we found anything problematic on structural MRI scans. If someone explicitly says no to that (which I can't remember happening - as you suggest, the vast majority of people would like to know) that's their choice.

Edit: What no-one has said here is whether the initial study did say that, and the original researchers were going against the wishes of the participants. Given how the researchers behaved that doesn't seem entirely unlikely.

Probably the most extensive critical account of that 1973 Sobell study, and how the Pendery 1982 follow-up came to be published, is the book Alcoholism: A Review of its Characteristics, Etiology, Treatments, and Controversies by Irvine Maltzman. All quotes in this post come from that book.

One patient’s “attorney had complained to the Hospital about the treatment that he had received.” As I describe elsewhere in this thread, some of the patients ended up trying to sue the Sobells. So, no, the patients were not happy with the treatment they received.

The subjects had no problem being contacted for follow-up: “the patients contacted prior to the court injunction all expressed a willingness to cooperate and to be interviewed”

Indeed, a judge familiar with the medical privacy laws of that era (we’re talking the mid-to-late 1970s here) made a court decision that contacting the patients for follow-up was perfectly OK: “Judge Hauk concluded that on balance it was more important to determine how the patients were faring following treatment than any possible breach of confidentiality and invasion of privacy which were protected by their right to refuse to participate in the study.”

The takeaway is that it's not always possible to think about all the cases it's worth it to contact the person. At the end of the day it's one phone call or email. Mild inconvenience for potentially life changing chance or information. As long as it's a persona in a position like a doctor o researcher making a call I am happy to waste a minute of my life trusting their judgement.

I don't really like where the EU privacy laws are going. I want "try to be reasonable" option. If they are not I can always tell them to delete my contacts later.

EU data protection laws are reasonable. They allow for public health authorities to take all the steps you're suggesting, and they require companies which hold such information to make them available to those authorities. They do not allow researchers to transfer data to a third party to perform further research. Simple, clear, reasonable, and protective of the basic human rights of the study participants.

There are clear processes to go through for these situations. Usually, individuals can specify in the original contract if they are open to follow up.

>Another example: a study for rare condition was conducted. No effective treatment was found. A few years later very promising treatment was discovered. Wouldn't you like to know, even if the original form didn't ask your permission?

You are literally describing healthcare providers giving confidential medical information to third parties without consent. In no way should anyone but the patient be able to void or reinterpret the contract, their presumption of the patients wishes should never override those stated by the patient.

It's interesting how you are so convinced of it even though it's easy to come up with scenarios assuming about any ethical system where privacy concerns are less important.

I gave them something dangerous during the trial and they didn't agree to a follow up. I discovered they need treatment now or they will die (may happen with a poisoning for example). I am still contacting them and I surely hope you would too.

I think we might be making different assumptions. There are 3 options as I see it

1) If the contract signed by the patient is silent on the topic, and you have their information, you are free to contact them.

2) If the contract signed by the patient explicitly says you can not contact them for this reason, then I think it clearly unethical to do so.

3) If the contract is ambiguous, or the intent is unclear, then it is a judgement call.

I am talking about #2 and I think you are talking about #1 & 3.

There are valid reasons why people would want to select #2, and if they do, their wishes should be respected.

In the EU, those cases would be covered by the lawful basis of being in the vital interest of data subject. However, that would only cover that particular reason for processing, and in the scenarios above would be handled by a public health authority. It would not allow personal data to transfer to a third party for the purpose of continued research.

There are a lot of people who "feel" that data should be able to be used in situations that are much more sketchy than you're suggesting. If you don't abide by the initial permissions that people give, how do you decide whose feelings win out? The best you can hope for is to ask people explicitly in the original study whether they're happy to be contacted for a follow-up - it's not clear to me whether that was the case here.

This is all quite true, the loophole seems to be that you can take these rules seriously but have much less obligation to keep what you publish within the limits your choices within the rules create.

The original publisher could only claim to know it had an unconfirmable suppressive affect on drinking for the short period they were allowed to contact the subjects. The limits were clearly used to imply more and alternative researchers that responsibly asked for sufficient permissions at the start would have been at a double disadvantage. (Harder to recruit subjects and results that were less likely to be remarkable.)

'always'... maybe ask first during the first study? Why do you get to decide inconditionnally that I'd ever want to hear about anything to do with this after the study? Did they agree for 50 years of phone calls about 'magic new treatment' for their disease?

Sorry, I don't see it. If you don't have the consent of the original subjects, do another study with 50 other people...

Do you think it was their 'reasonable concern about privacy' that drove their zeal to suppress the report or do you think it was because the was immediate proof that their study was nonsense?

Does it really matter? The cost of throwing out the privacy baby along with the bad actor bathwater is much higher than that of this one bad study, even if it was mitigated slightly by the good intent of the follow-up.

It would have been better to run a new study and try to reproduce the results but as we’ve seen with the crisis in reproducibility throughout the sciences, that’s problematic in itself.

That’s not what the judge looking at this very issue felt:

“Judge Hauk concluded that on balance it was more important to determine how the patients were faring following treatment than any possible breach of confidentiality and invasion of privacy which were protected by their right to refuse to participate in the study”

Source: Alcoholism: A Review of its Characteristics, Etiology, Treatments, and Controversies by Irvine Maltzman

(Keep in mind this was the mid-to-late 1970s in California)

So in the same era as the Tuskegee Syphilis Study? I couldn't really care less what the judge concluded, I hope we've since established that medical ethics weren't what they should be.

The better of the quotes you've dropped in this discussion is

> the patients contacted prior to the court injunction all expressed a willingness to cooperate and to be interviewed

That is to say, the participants' expressed their consent to follow-up action. I'd still prefer not transferring personal information to another group of researchers, though - the same thing could have been achieved with some kind of advertisement in the media ("participated in a Sobell study? Call this number").

Tuskagee is about the worst example you could pick. The subjects in the trial would have benefitted substantially from having other researchers contact them.

Not really; by that point the damage was done.

The subjects would’ve benefitted equally by the doctors being put in prison for the rest of their lives (or given the death penalty, per Nuremberg) if only to set an example to the rest of the field.

What? No. Punishing the researchers benefits society through deterrence. What the patients needed was treatment. The earlier the better.

Massive damage was already done to their bodies, but you can still get rid of late stage syphillis with penicillin and prevent further damage.

Treatment's a matter for a public health service rather than other researchers though, right? There's not much science in another group of researchers taking over to test how penicillin treats syphilis; that's already settled. Access to the affected patients would then be a legal matter - the records would be seized by the police and/or prosecutors and that's a very different kettle of fish to the original example.

For what it’s worth, Tuskegee just happened to be the ethical failure that came to mind because it's been in the media recently (in the context of "why do PoC have reticence to take vaccination", i.e. for COVID-19). There's plenty of other examples where the damage was irreparably done [1, 2] where it would be questionable to hand the participants' personal information over to another group of researchers for further study.

1: https://en.wikipedia.org/wiki/Unethical_human_experimentatio...

2: https://en.wikipedia.org/wiki/Study_329

Comparing Pendery 1982 to the Tuskegee Syphilis Study?? That’s a really emotionally charged and completely invalid reach.

I’m not sure how a reasonable conversation can continue. We’ve moved to a Twilight Zone alternate reality at this point. The only way we can go further in to la la land is to make a completely invalid comparison to Nazism so that Godwin’s Law can be invoked.

Another thing: Discarding what a judge says about a what is legal matter is not a particularly compelling argument. It’s the kind of thinking done by people who believe in conspiracy theories. As a contemporary example, people who are convinced that the 2020 United States presidential election was rigged, when pointed out that judges have concluded there is no reasonable evidence supporting that assertion, will say something like “I couldn't really care less what the judge concluded”.

I mean, when your entire post is a rant it's hard to converse reasonably ¯\_(ツ)_/¯

If your problem's with Tuskegee specifically (and at no point did I compare your alcoholism study with Tuskegee, that’s a straw man), there's plenty of other ethical failures I've referenced in sibling replies where it would also be problematic to hand patient data to a separate group of researchers, no matter the good intentions.

My original point, since it seems to have been missed, was that no matter the failings of the original study it would be problematic to pass patient data to a third party without a legal requirement to do so.

With respect to my disregard for what a judge says, if your basis for ethics is "if it's legal then it's fine" then you're definitely in cloud cuckoo land.

Probably time to relax the laws on defamation a little bit, to prevent cases like these. The laws date from a time when access to media was hard, and just getting your thing published was an impossible amount of work. Now that we have social media, it's not clear that we need the law to "protect" us here. If someone writes a hit piece about you, can probably get more people to read your defense. That's new, and the laws haven't been updated to account for that.

(Unlike more pressing issues facing society, this one doesn't require spending money, or passing a Constitutional amendment, so it feels within reach!)

First time I've encountered this opinion, and this makes so much sense. Thanks :)

Seems like exactly the wrong approach. New technology makes false things you write about someone easier to discover and more likely to be read in the future by, e.g. a company considering hiring you.

Defamation is more damaging to the average nobody than it ever could have been 50 years ago. If anything the laws should be strengthened.

I disagree, for the same reason that forums typically ban trolls rather than just let them be. It takes an order of magnitude more effort to defend against a claim than it does to make the claim.

I could hypothetically accuse you of being a rapist right now. I can just pick a random day I say it happened, let's say last February 23rd. Cool, it took me about 30 seconds to come up with that. Let's say I then blast that all over Twitter and Facebook. Are you okay with having no legal recourse? Are you okay with "jrockway is a rapist" being one of the top Google results for your name? Maybe it's better, maybe you did really well and your rebuttal is the first Google result. I would still be really uncomfortable that my first Google result is me attempting to say that I'm not a rapist.

That's assuming people will even read your defense. There are some classes of accusations where you will be tried in the court of public opinion before you are even aware, and your attempts at defense will only be interpreted as further evidence that you are guilty. The rise of Fake News accusations should tell you exactly how much telling the truth is going to help your case.

Rape is a particularly heinous thing to accuse someone of, but I chose it specifically because of the "believe the victim's narrative" messaging that surrounds it. Accusations of a crime fall under defamation laws, and so long as they don't file a police report about it, defamation is the only law I'm aware of that it would violate.

Is it worth suppressing life-saving science to clean up my Google profile, though? Everything can be abused, but some abuse is more harmful than others. As the laws stand right now, legitimate science journals are afraid to publish legitimate science. That's super bad.

Any link to a copy? I think Streisand Effect needs to happen to this one.

(Removing content usually just makes people more interested in it.)

It's still on sci-hub for now:

https://sci-hub.se/https://doi.org/10.1016/j.jceh.2018.08.00...

Some of that stuff is pretty thin - you'd expect to find 16s rRNA (microbial genetic material) in just about anything that contains biological material like herbs. The isolation and sequencing process for this type of material amplifies what is found in the sample by many orders of magnitude using techniques like PCR, so presence alone doesnt mean much - it certainly doesnt mean there were live, pathogenic bacteria in the sample (there are other, better tests for clinically relevant pathogens). I spent quite a few years doing metagenomic research, and let me tell you, bacteria and bacterial genetic material is everywhere.

Most of the rest of the article I couldnt comment on, though I found it a little funny they list Humulus lupulus as a "suspected toxic component" - thats hops, the same hops used to make beer (also used in cosmetics and other products). Matricaria chamomilla is also listed as a suspected toxin. Its chamomile, the flower used to make the relatively common tea.

Bacterial and fungal contamination is quite common with plant derived dietary supplements. Ideally a microbial count should be performed. We did not expect to see various bacterial profiles in a vacuum packed food supplement. Also, Bacillus subtilis contamination in Herbalife has been reported before (Stickel et al, Journal of Hepatology). Please see this paper on plant origin dietary supplements and contamination potential. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6362155/ This can actually promote spoilage and can lead to damaging consequences. Hops use other than in Beers - such as direct consumption has been shown to be toxic to animals and also adverse events have been noted in humans.Chamomile is usually safe, but its interaction with other herbs and herbal metabolites could lead to systemic toxicity. This is always a concern with polyherbal formulations.

The amount of barium and chromium in the samples they analysed is surprising.

Hello. I am the first and corresponding author of the Herbalife study that was removed due to legal threats. Please find this folder to access all pertinent documents related to this incident including the legal ones.

https://drive.google.com/drive/folders/1WukaGm8FfVWXyNulpc60...

Regards Dr Abby Philips.

Ooof, if mere Herbalife can get Elsevier to do this, one wonders what science about the oil industry, tobacco, sugar, etc has been concealed by Elsevier.

Or any pharmaceutical company...

There is nothing “mere” about Herbalife.

Herbalife has had revenues of almost $4.9B in 2019 and $311M of net profit. They aren't really a small player that I would call "mere".

They'd be a Fortune 600 company - around #550 - if they weren't officially incorporated on the Cayman Islands and therefore technically not a US company.

So there are 550 better resourced companies in the US alone, with the leading ones making $270B to Herbalife’s 5?

Sure, and there is also roughly 10.75 million companies in the US alone with worse "resources" by that metric, although we're talking revenue here not resources.

Having a cool $300M of essential "fuck it" money (net profits) per year, to mess with your enemies if you want, is substantial too.

I did not know that one can settle a pyramid scheme in US. Good to know.

I really love how they settle up pyramid scheme with government especially.

Herbalife is a cult masquerading as a business. The members are indoctrinated to the point that the term brainwashing probably applies and the group is extremely litigious and vengeful against those that leave, much like any other cult. I've seen the parents of a good friend be sucked in to this, their whole life revolves around that trash and they are up to their necks in debt because of it. It's as if Scientology, multi-level marketing and homeopathic treatments had an evil child.

I think Herbalife may get a bit of a Streisand effect here. I never would have heard about them having toxic product issues in India without the legal threats.

Here is a Google drive link to the paper and related correspondence:

https://drive.google.com/drive/folders/1WukaGm8FfVWXyNulpc60...

And John Oliver did an exposé on Herbalife's business practices:

https://youtu.be/s6MwGeOm8iI

I can really feel the tide turning in the health industry right now. Up until this point, we often looked to add things to our diet to increase our health (whole grains, fiber, antioxidants, minerals, supplements, etc, etc).

The evidence from things such as intermittent fasting is suggesting the path to better health is to remove from our diet and abstain from food, not the other way around. Removing all grains from your diet is probably vastly more healthy than adding whole grains.

Coincidentally the group of people who would fall for Herbalife is similar to the group of people who believe insane anti vaccine theories.

Is there any way that lawyers can be sent to jail for making egregiously bad legal threats? Does it ever happen?

Idk about jail but I assume there is a risk of being disbarred.

But if a cop breaks the law, he goes to jail right? Know anything about that one? XD

Edit: Sorry. This is a crazy comment. Cops go on the Brady list and are forced to work in low income neighborhoods. I knew that. Wait, but that’s crazy?

The rules of federal procedure has a rule for this, rule 11. It’s designed to punish lawyers or litigants who either bring unsupported cases, or otherwise misuse various aspects of the federal court system. Rule 11 allows for federal judges to directly sanction bad-faith actors, up to paying their opponents fees.

Like many rules designed to reign in those with power who misuse it, it seems that it isn’t applied very often.

Yes, it appears to be very rare.

Otherwise Guliani and the rest of the "Elite strike force" legal team would be fully sanctioned by now.

US law does not matter to law firms in Delhi...

In the past, the US has indemnified all Americans for being sued in jurisdictions with bad defamation laws. So assuming you don’t have any business in India, a fair response is “sue me in America, or <bleep> off”.

"up to"? That should be the minimum.

I wish.

Until that becomes common (or even possible), the best alternative we have available is to make sure that the companies and legal teams do not succeed in their ultimate goal.

That objective, of course, is to create a "chilling effect" or, to put it bluntly, get people to shut the fuck up.

If they don't want $thing known, the best way to fight back against that is to make sure that as many people as possible know about $thing.

In other words, introduce them to the Streisand effect.

Some states have anti-SLAPP laws.

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

Reply on news.ycombinator.com