It looks like they have found a stastically significant correlation with a very small effect size, using an extremely simple regression model adjusting for only a handful of covariates, looking across, presumably, tens of thousands or maybe even hundreds of thousands of SNPs.<p>It seems highly irresponsible to title such a study in a way implying a mechanistic or causal link. I expected to find a whole lot of discussion in the post about all the other evidence supporting it, but nope - they shoved their data into a stats package, and put the result straight into a blog post with a misleading title. They don't even discuss if they adjusted for multiple testing.<p>There's peer reviewed science, then there are pre-prints pending review. This is <i>far</i> <i>far</i> below either of those.
Hot damn, I may be > 400 lbs but I'm taking Vitamin D, Zinc, AND have o-positive. I also wear masks everywhere, cause I'm not stupid. I'll let you know if I survive.<p>Before the lectures about 'have you tried losing weight', I was 690 in 2012, so yeah, it's just a SLOW burn, not a fast overnight thing, still going down, just slow and about 80 lbs is skin that needs removed and I just haven't been able to afford that surgery yet and was wanting to get lower still first anyways.<p>W/ covid here, seems not the best time for electives anyways.
A group of hospital researchers in Europe published a preprint with similar findings.[1] It has received recent coverage in the New York Times.[2]<p>[1] <a href="https://www.medrxiv.org/content/10.1101/2020.05.31.20114991v1" rel="nofollow">https://www.medrxiv.org/content/10.1101/2020.05.31.20114991v...</a><p>[2] <a href="https://www.nytimes.com/2020/06/03/health/coronavirus-blood-type-genetics.html" rel="nofollow">https://www.nytimes.com/2020/06/03/health/coronavirus-blood-...</a>
It is good to remember when seeing a study showing a link between blood-type and some outcome: blood-type itself means very little biologically (it is one largely irrelevant gene among tens of thousands), but blood-type is heavily correlated with ancestry/race, which is heavily correlated with income/class, which is heavily correlated with lotsa stuff. When you see a blood-type effect, you are usually seeing a race effect or a class effect. It looks like this study tries to correct for "race [and] ethnicity", but not income. Even when you try to correct for these things, you can often still end up with a residual effect.
That is very old news, this is from March 11:<p><a href="https://www.medrxiv.org/content/10.1101/2020.03.11.20031096v2.full.pdf" rel="nofollow">https://www.medrxiv.org/content/10.1101/2020.03.11.20031096v...</a><p>Type A is:
32% in the population
38% of those infected
41% of those that die.<p>Type O is:
34% in the population
26% of those infected
25% of those that die.
>Although one study found the blood group O only to be protective across rhesus positive blood types, differences in rhesus factor (blood type + or -) were not significant in 23andMe data. Nor was this a factor in susceptibility or severity in cases.<p>I'm O -ve and everyone else in my family are O +ve. Going by the rate of infections in my state, I'm of opinion that it's just matter of when and not if we'll get infected by COVID-19.<p>So, I'll update my data here when that happens.
This is a PR blog, of a private company, publishing "first blush" of "preliminary data".<p>While any contribution to knowledge is warranted, this should not be circulating.
What is the reason that nearly everybody refers to Covid-19 as the name of the virus, instead using SARS-Cov-2?<p>In the study they even refer to 'genetically studying covid' which does mean the disease caused by the virus, but I would assume one would study the virus genetically instead?
It seems that they're inferring the blood type from the ABO gene, although displaying this on the web site is a "sunsetted feature." Edit: the rhesus factor, on the other hand, seems to have been self reported.
Interesting.... But am I the only one who recalls seeing multiple articles stating that they suspect type A to be the most susceptible? It's curious because due to this study's findings it would appear that while type O fairs the the best, type A has slightly better chances compared to type B and type AB. So in other words type A is the second most resistant? Am I interpreting that right?
Not surprising. Serious COVID-19 cases frequently have blood clotting problems. Previous research on other diseases found that Type O blood is correlated with a lower risk of blood clots.<p><a href="https://www.ahajournals.org/doi/10.1161/ATVBAHA.119.313658" rel="nofollow">https://www.ahajournals.org/doi/10.1161/ATVBAHA.119.313658</a>
This was already known in April. But a Chinese study so nobody bothered. Two weeks later confirmed by a New Yorker hospital.<p><a href="https://www.pharmacytimes.com/news/study-suggests-blood-type-a-associated-with-higher-risk-of-covid-19" rel="nofollow">https://www.pharmacytimes.com/news/study-suggests-blood-type...</a><p>[old news]<p>This is now the confirmation study of both.