Hacker News Reader

Comment by i_am_proteus | Hacker News Reader

Author compiled a lot of studies on his web log: https://velvetgloveironfist.blogspot.com/2021/08/smoking-and...

The studies do appear to show that smoking cigarettes is correlated with a lower probability of getting the Covid, for a lot of different definitions of "getting the Covid."

They also appear to be, by eyeball aggregate, inconclusive as to whether the risk of dying from the Covid is higher or lower for smokers: it's certainly worse to be a smoker than not if you do catch the Covid.

Smoking also has some other bad health effects. I feel like that's pretty well-understood these days. I know these things; I was a smoker; I still am a smoker. I would not recommend taking up a cigarette habit just because of the Covid. Doesn't seem to cut down on the risk of dying from the Covid, still increases the risks of having bad health effects from everything else that's harmful about smoking.

View on HN

> The studies do appear to show that smoking cigarettes is correlated with a lower probability of getting the Covid

The problem is that non-smokers is not a single category of people, so the confounding factor for the correlation is "why X does not smoke".

For example, if you had a study which compared moderate drinkers to non-drinkers, but not carve out "cannot drink due to other medication" from the second group, you get a more discriminatory result from the study. The real problem of course is that confounding factors are almost fractal in nature.

On the other hand, if you had a comparison saying "Smokers replace their mucus linings more frequently than non-smokers, expectorating the virus with it", then that I would buy as a causation. Coughing up phlegm does offer a direct means of explaining a reduced exposure.

> I would not recommend taking up a cigarette habit just because of the Covid

So there's a weird thing that happens when news about a new improvement in life comes up.

A study gets published "people who eat dark chocolate live longer", but then a bunch of relatively unhealthy but optimistic people actively choose to eat dark chocolate or drink red wine or whatever new superfood, but fail to live longer as a result. Mostly because they just made that one change, plus there are now a thousand more manufacturers of the same product (say Manuka honey), with varying fades of quality as time goes on.

So a change in human behaviour is triggered by a study both for consumption and production, which ends up proving that the original study modified the circumstances in which it was originally true. And we are in the quest for a better causation of the original observation.

There is no such thing as a confounding factor for a correlation. It only matters for causality.

I will continue to maintain that, given this evidence amidst the existing evidence that smoking is not good for you, you should not start smoking for reasons related to health risks from the Covid.

> There is no such thing as a confounding factor for a correlation.

Sure there is.

As an example, let's say there's a positive correlation between the dose of some antidepressant and the probability of suicide. There could be a confounding variable with respect to the severity of depression in those samples such that people with more severe depression are more likely to take larger doses and more likely to commit suicide. The original correlation may not necessarily be between the dose of the antidepressant and the probability of suicide, but rather the severity of the depression and the probability of suicide. In other words, the correlation could be a correlation of the wrong thing because you didn't control for depression severity.

It does not matter only for causality, but also for correlation. In both cases, it's important to control for confounding variables so that you're measuring and analyzing the actual variable(s) that you're trying to, not some other variable(s) (i.e., confounding) indirectly. Otherwise, it may no longer be a relationship, whether causal or correlative, of the thing you think it is, but of some other thing.

In your example, the positive correlation between the dose of the antidepressant and the probability of suicide still exists in the data.

What you're describing is sampling bias: the correlation between the antidepressant and probability of suicide in the sample may not reflect a similar correlation in a larger population.

> In your example, the positive correlation between the dose of the antidepressant and the probability of suicide still exists in the data.

Of course it does because the confounding factor of depression severity wasn't controlled for. You can't always control for confounding variables after the fact.

> What you're describing is sampling bias: the correlation between the antidepressant and probability of suicide in the sample may not reflect a similar correlation in a larger population.

Yes and no. I am describing sampling bias, but that's because the sampling bias is the source of the confounding variable in my example. The point was that the correlation was supposed to represent the relationship between antidepressant dosage and probability of suicide, not the relationship between people who typically take antidepressants and their probability of suicide. In my example, the dosage ended up being a dependent variable along with the probability of suicide and the severity of depression ended up being the independent variable, when the dosage was supposed to be the independent variable.

This is technically correct, but it hardly matters, because the author clearly wants you do draw casual conclusions from this correlation.

Even though looks a lot like a classic example of collider bias, where obesity looks like it protects you from dysglycemia.

> For example, if you had a study which compared moderate drinkers to non-drinkers, but not carve out "cannot drink due to other medication" from the second group, you get a more discriminatory result from the study.

Yeah. From what I remember about these studies is that people who used to smoke but don't anymore actually have higher risk of contracting severe Covid. This is easily explained by the fact that many people who quit smoking do so due to health reasons (which in turn are the cause for higher risk of Covid).

This is in fact what seems to have been underneath all those studies 10-25 years ago that seemed to show that non-drinkers have worse health outcomes than moderate drinkers. The problem is that "non-drinkers" includes a lot of people who can't drink because they're already sick or because they're alcoholics and drug addicts with long histories of abuse (and so on).

Once you control for that group non-drinkers do better than moderate drinkers.

Oh really? That sounds super plausible. Do you have a good source for that? I'd love to dig a bit deeper on this as the whole "drinking once a day is good" thing is something that comes up a lot.

From what I understand, it's hard to know exactly how non-drinkers are different (besides not drinking), but the nail in the coffin for the moderate drinking-endorsing studies which tried to correct for all the factors they could think of, was the rise of Mendelian randomization studies.

Some people have gene variants that interfere with alcohol digestion, effectively making hangovers come much faster and harder. Because these people drink less than other people for a well-understood reason, and are otherwise like everyone else (the gene variants are well mixed into the populations where they occur), it can be used as a causal wedge to pry apart the health contribution from alcohol from the contribution of basically anything else.

And when they did this, the supposed health benefits of alcohol, which were already smaller the more reasonable things you controlled for, vanished entirely.

You can look up meta studies of Mendelian randomization studies on alcohol use on Cochrane and so on of if you want to dig deeper into it.

There has been some pushback, trying to argue that if you don't assume the effect of alcohol to be linear (I.e, don't assume a normal dose-response relationship as you would by default on pretty much anything else), then there's too little data to tell. I don't buy that, as you can probably guess.

Equally we have many studies showing that vitamin D helps to reduce risk from covid (hint: it doesn't).

More studies show that ivermectin has a positive effect in covid patients. In fact, ivermectin has no effect or has only a very weak effect in certain patients.

In medicine observational studies suffer from certain problems that are hard to overcome. The biggest is that the groups are not equal. For example, smokers die younger. Covid risk increases exponentially with age. Consequently, covid will kill more people who have reached greater age as a result of not smoking. The researchers may think that they control for these factors but in reality it is very hard to do.

If you take a random article, it says smokers have greater risks from covid: https://www.renalandurologynews.com/home/departments/mens-he...

Not much evidence in either way, but before we study for reasons how smoking could protect from covid, we would like to know for sure that this is really true.

> Equally we have many studies showing that vitamin D helps to reduce risk from covid (hint: it doesn't).

But we know the reason for this; in order to get the circulatory and anti-viral benefits of vitamin D, you need to be exposed to a sufficient amount of UVA. But most folks who are getting a sufficient amount of UVA are also getting enough UVB to make enough vitamin D in the first place, which is what always causes the outcomes of the studies to look equivocal. It's really not that complicated, but researchers keep running fake studies to scam money out of the government and make pharma drugs look more effective by comparison.

I'm sure I'm just misunderstanding, but I don't follow the explanation. What do UVA and UVB have to do with this, and why does that lead to weird results?

The reason covid is dangerous is because it binds overly tightly to your ACE2 receptors on the cells in your endothelium and damages them, and when enough of these receptors are damaged the cells die. This can cause all sorts of issues like autoimmune cascades, blood clots, dysregulation of blood pressure and heart rate, etc.

The reason vitamin D is protective is because when you're exposed to UVA, the vitamin D acts as an enzyme that your body uses to produce nitric oxide in your skin and capillaries. Nitric oxide is directly antiviral, but even more importantly it upregulates your ACE2 receptors. With more ACE2 receptors there is less damage caused when some of them are destroyed, which means you are less at risk for blood clots, problems with your heart and blood pressure, deadly autoimmune cascades, etc.

Basically being exposed to UVB causes your body to produce vitamin D, but once you have the vitamin D you need exposure to UVA for your body to produce nitric oxide, which is what causes the health benefits.

The reason smoking is beneficial for covid btw is because nicotine also upregulates the density of your ACE2 receptors. And the reason women are less likely to die is also, you guessed it, because they have more ACE2 receptors.

Aside from vitamin D, you can also obviously upregulate your ACE2 receptors by taking any other food that causes your body to produce more nitric oxide, and/or by taking any ACE inhibitor. OTC things in this category include garlic and hawthorne, and most of the other foods that are commonly used to lower blood pressure.

Do you know if you can get this beneficial UVA from an artificial lamp?

Is it as good as going outside?

Is it good enough to be a near-replacement or is it even worthwhile as a supplement?

> Do you know if you can get this beneficial UVA from an artificial lamp?

I mean artificial tanning uses UVA lamps. But since you can get UVA even in winter, it's probably safer and more effective just to go outside.

We have many studies showing that hypovitaminosis D is correlated with the risk of severe COVID-19 symptoms. What's not clear is whether there is a direct causative relationship, or if higher blood vitamin D levels are just a proxy for other generally healthy lifestyle factors.

Some studies have shown a protective effect from vitamin D supplementation and others have not. The general flaw in most of those studies has been that they gave all of the study patients the same dosage instead of titrating the dosage to hit a target blood level. There are huge variations between individuals based on sunlight exposure, diet, genetics, and other pathologies so the data is really noisy.

Before making any claims in this area I recommend reading through at least the abstracts of all studies indexed on this page. It's a complex area and we have no definitive conclusions yet.

https://vitamin-d-covid.shotwell.ca/

I wonder if people with higher vitamin D levels are more likely to touch grass. It's good for your immune system to go outside.

Probably. That's one of the confounding factors which makes this difficult to study in a really rigorous way.

People who smoke are probably more likely to smoke outside too with the recency of indoor smoking bans

There is very good evidence that vitamin D in general helps your immune response. It may be that there is a second factor here (another commenter suggested that UVA light was also involved in this process), but the evidence in favor of boosting your immune response by getting sunlight every day is pretty conclusive.

I would be surprised if this study was a vindication of vitamin D, though, rather than a false positive. I think smoking has been shown to generally have negative effects on your overall health (and immune system).

Vitamin D is obviously necessary, mostly for bone health, but any study of supplementing it for other purposes to people without any symptoms of deficiency have been disappointing.

> generally have negative effects on your overall health

smoking tobacco for years, every day, definitely changes your body. Most of those effects are not great in the long run, yet many people start and continue to smoke cigarettes for years on end. The change to lung tissue, mucus membranes, tongue, mouth ecosystem.. are real.. and for an otherwise healthy person, yes, they have to get "tougher".

It’s easy to be confidently wrong with statistics.

Ivermectin improves outcomes in a hospital setting… but only in developing nations.

There’s two confounding effects there! One, you have to be sick enough to be treated in hospital and two, living in a place with high levels of endemic parasite infections.

Ivermectin treats parasites. Parasites are helped by the steroids given for severe infections. Treating the parasite infections does -- demonstrably -- help outcomes, but not uniformly. It's not a prophylactic and it won't do much of anything in developed nations where less than 1% of the population has parasites.

I noticed just how bad these studies were when I took Tamiflu for the worst flu I had ever had. I went from crying in pain to "this feels like a mild sniffle" in less than 12 hours, the fastest I had ever experienced flu symptoms subsiding.

My doctor recommended against taking Tamiful because it was expensive (not really) and that studies showed that its effects were marginal, something like reducing the duration of symptoms by a mere 1/2 a day out of 2-3 weeks.

Reducing the duration... reducing the duration. You see, there's no objective method for measuring symptom severity, so during studies of new medications, they don't. They measure duration. But duration is very difficult to reduce, because of the way the learning process of the immune system works. There's a bunch of sequential processes that occur sequentially.

Tamiflu reduces symptom severity but not duration. So apparently it's "worthless", because it improves only outcomes that aren't measured.

Whether you're crying in pain or not is apparently not objective enough...

> Ivermectin improves outcomes in a hospital setting… but only in developing nations.

Precisely. The only studies I could find that were well done and did show correlation were in developing nations. Turns out that having parasites + a severe viral infection can be bad for you.

However, most intestinal parasites are cleared with a much lighter regimen that many ivermectin advocates were suggesting, some with megadoses that are harmful to the liver. There are other dewormers that are much better - some are even single dose.

Have we seen studies from India, Peru and a few other countries, japan as well I believe, on ivermectin?

> Tamiflu reduces symptom severity but not duration. So apparently it's "worthless", because it improves only outcomes that aren't measured.

To be fair, that's not really "being confidently wrong with statistics" so much as "being confidently wrong by ignoring all of the statistics that are inconvenient to collect".

> But duration is very difficult to reduce, because of the way the learning process of the immune system works. There's a bunch of sequential processes that occur sequentially.

I'm genuinely very curious for more detail about this. Can you describe what some of the processes are (or better, where I can go to learn about it)?

I read about this process in the Immune book by Dettmer.

Your dendritic cells are specialized to ingest pathogens, rip them apart with peroxide, then push their broken parts into cellular windows all over the dendritic cell's surface.

The dendritic cell is also traveling in the lymph system while this is going on, and upon arrival at a lymph node it searches for a t-cell that matches some part that it is presenting. Once a t-cell is found that matches, that cell activates and clones thousands of copies.

Half of the clones go towards the problem sites (where they trigger "ultra killer mode" on macrophages, among other things), and half stay behind to find and activate b-cells, which will then produce antibodies.

This entire process does take a few days.

https://www.washingtonpost.com/books/2022/02/17/dettmer-immu...

Here are some videos that this group made:

Bacterial infection: https://m.youtube.com/watch?v=lXfEK8G8CUI

Compliment system: https://m.youtube.com/watch?v=BSypUV6QUNw

Measles: https://m.youtube.com/watch?v=y0opgc1WoS4

Rabies: https://m.youtube.com/watch?v=4u5I8GYB79Y

There are plenty of studies that have participants rate the severity of their symptoms based on some questionnaire inputs. That they didn’t do it on Tamiflu isn’t reflective of an inability to gather that data, it’s just reflective of the fact that it wasn’t gathered.

And, of course the results aren’t totally objective, but absolute objectivity isn’t needed in all circumstances, particularly not when we have no objective way to measure a thing.

> Equally we have many studies showing that vitamin D helps to reduce risk from covid (hint: it doesn't).

How so?

Because people did a bunch of small, poor-quality studies that seeded to show improvement. But when we did larger scale, properly controlled, randomized studies, where we gave people vitamin D, it didn't show much of an effect.

And yet people persist in different theories why it is so. I don't know the mechanism, but randomized, placebo-controlled studies currently are gold standard in medicine. Their results are much more trustable than many observations studies. Sometimes we cannot do randomized studies and observational studies is all we can do but if if we can and research report only observational studies, I take with a big dose of salt.

I dunno.

The only thing convincing that I could find was that a single high doses of Vitamin D had no effect on outcome. No other studies seemed conclusive.

> Vitamin D2 and D3 fills were associated with reductions in COVID-19 infection of 28% and 20%, respectively

- https://www.nature.com/articles/s41598-022-24053-4

- https://news.ycombinator.com/item?id=33648827

People just stand farther away from smokers, even outdoors. Because smokers stink. And if your having trouble breathing cause of the Covid, you will go to even greater lengths to not get close to smokers.

But also, people with chronic respiratory problems, ie pretty much all smokers, you won’t notice you have mild Covid. Everyday of being a smoker is like having mild Covid, so less likely to get tested etc.

Lastly smokers are on average lower IQ, lower SES, etc. very hard to control for that health research. I think we do know that lower SES communities were less likely to seek health care including vaccines and testing.

> smokers are on average lower IQ

Source?

https://pubmed.ncbi.nlm.nih.gov/18606723/

> Results: Lower IQ measured at ages 18 to 20 was weakly associated with increased prevalence of smoking, independently of indicators of mental illness and social misbehavior measured in late adolescence. By contrast, smoking cessation later in life among those who smoked at ages 18 to 20 was not associated with IQ. Among smokers, lower IQ was significantly associated with a lower level of smoking after adjusting for other factors.

> Conclusion: Low IQ was associated with an increased prevalence of smoking in adolescence. However, the main part of this association disappeared after adjustment for measures of mental health and social function in early life. IQ was not associated with likelihood of quitting smoking.

After being given this information, a reasonable person would conclude that while smoking might decrease their risk of getting COVID, starting a smoking habit (possibly leading to addiction) would mean likely worse health outcomes than just accepting the risk of getting COVID (which can be lowered significantly through much safer means).

Unfortunately, there are lot of unreasonable people out there. I would not be surprised if a ton of people, if they were to hear the WHO or CDC agree that smoking reduces the risk of COVID infection, would start smoking, and get addicted. Some of those who were already addicted and were trying to quit would decide to stop trying.

The issue is that this information seems to have been suppressed, censored, and outright lied about by authorities. Now, on one hand, I get it: organizations like the WHO and CDC are not only interested in the raw outcomes of trials and research, but are also responsible for influencing public health policy. I think it's pretty safe to say that telling people smoking helps protect from COVID would be against public health goals.

I honestly don't know where I stand on this. I think I'm fine with orgs like the WHO and CDC deliberately not making recommendations on, or being silent about, things that they don't believe are healthy to encourage. But outright lying, denying the results of trials that do prove a correlation, crosses the line for me.

Agree, with the nuance that I think the psychology at WHO / CDC / etc is probably focused on existing smokers who may choose not to (try to) quit because of data like this ("WHO says I should keep smoking!")

But there are going to be misinterpretations and bad outcomes no matter what. So I'd err on the side of honesty and integrity, and perhaps trying to mitigate harm by producing data on, say, life expectancies for smokers who quit today versus those who don't, and (presumably) the data that says it's still way better to quit. And if the data doesn't support that, well, data is data.

I think the most likely outcome is for a certain subset of the population to think “I don’t need to get vaccinated, I smoke cigarettes.”

And they would have been right.

No they wouldn't. Smoking gives a little protection against infection, but once you have it it's likely much worse. Vaccines give you a little protection against infection and also reduce the probability of severe courses of illness.

50-68% more protection which is better than the vaccine. And it did provide better outcomes when patients were treated with patches in the hospital.

It seems controversial even saying this, and a bit absurd… But I am a heavy smoker and have somehow managed to dodge Covid so far. It is probably coincidence, but I would love it to be studied into more no matter how inconvenient it may be…

I'm a non-smoker but use nicotine daily. My mother (nicotine free) died of COVID, but I have never had it. But we are mere anecdotes, and prove nothing.

Sorry to hear about your loss. I haven’t experienced loss on that scale yet.

It would be interesting to see what impact nicotine tablets had on infection outcomes.

A longer-term study on dementia would also be helpful.

Nicotine is expressed by plants as a pesticide, but the effect on humans is obviously more nuanced. There might be beneficial effects, especially if the delivery mechanism does not involve setting anything on fire.

Also, I'd imagine the people who were smoking when COVID started being a thing were smokers well before then.

So whatever factors make smokers less likely to contract COVID may be because of some of the long term effects of smoking. So picking up smoking isn't necessarily going to make you immediately safer.

One doesn't actually need to smoke cigarettes -- mixing my own vape juice (no flavours, tuned pg/vg and nicotine levels no one dictates) works just fine.

Edit: in any case rolling tobacco is way better than any cigarette one can buy (not to mention 36mg Indonesian killsticks)

The editorializing on “smoking is bad” is moot—we all know it and it’s not the point.

> on his web log

Also called blog, since ~20 years.

https://news.ycombinator.com/newsguidelines.html

> Be kind. Don't be snarky. Have curious conversation; don't cross-examine. Please don't fulminate. Please don't sneer, including at the rest of the community. Edit out swipes.

Thanks for reminding me. Point taken.

blog? That word is so last year. That's an extented tweet or non-visual tiktok movie.

Reply on news.ycombinator.com