Coffee With Scott Adams — Knowledge Archive October 9, 2026
Scott Adams Philosophy Archive
Search ideas

Context —

Now if you joined us last night I had a little technical difficulty. I was trying to bring on a special guest to talk about convalescent blood serum, and as soon as he's ready we're going to try that again this morning. So I'll wait till I see when Ian gets on. All right, let's tell him we're good now. In the news, my favorite story is that President Trump apparently wanted to go golfing this wee…

← Previous segment →

All right, let's see if we get Ian on the line. This will be a test of the technology. I think we got it. I think we got it. Oh, seriously? Because I know we had a Wi-Fi problem yesterday and it looks like we got a Wi-Fi problem today. Well whatever the problem is, it doesn't look like that's going to get solved. I'll try it one more time but then I think we should maybe give up on this if it doesn't work this time. Yeah, and we'll try you one more time. Yeah, hey, let's make this work. This is the first time the technology has worked more than five seconds, so I feel good about it. That's a good sign. All right, good strong signal.

All right, Ian, first of all let me see if I can pronounce your name correctly. Ian Hillgard Martius? Pretty close, yes. You nailed it. All right. Can you tell the audience what your background is and how that relates specifically to the convalescent blood plasma testing situation? Give them a little background on you.

Okay, so my background is that I worked in bioscience for about ten years and then I ended up kind of leaving academic bioscience and taught myself to program and became a data analyst and started working for a corporation as a data analyst. And when the COVID stuff started to hit I would figure out ways to harvest public data to do public data analysis that I could publish so to help people understand the situation a little better. And then through that process I started to think about different answers that I thought we needed that maybe we weren't getting from the media. And one of the things I thought about was, well what can we do now? What can we do before there's a vaccine? Because there's going to be people who can't wait that long and the development of new therapeutics takes so long that a lot of people would end up dying while they wait. And so we got to figure something out in the middle.

So tell us what you did, because I was sort of following along early on and you put together your own tests, right?

Well it wasn't necessarily my own test. I bought a commercially available test kit. I just happened to become aware of its existence pretty early. I suppose the way that I became aware of it is that I just got like a spam email because when you buy something from a vendor they put you on a mailing list and then they always blast out these kind of spam advertisements. And so I got one of those that said, oh hey we have these COVID-19 antibody tests. And so I thought, oh I want to buy one of those. And my friend and I were just going to test ourselves. So I bought one and then it had space for 40 tests and so I thought, well I might as well just fill this whole thing up and run 40 tests and kind of get a better sense of how many people are positive because at the time it was unclear how far spread the virus had been.

So you are really, I think you might have been the first person who did an amateur or professional test of how many people had antibodies. Do you know if anybody tested before you?

No, as far as I know I was the first to report or I did publish the first report on a community kind of serum survey for COVID-19 antibodies.

Now you've been sort of paying attention to this little serum antibody thing and I know there was an article in The Wall Street Journal that said they had some small study, 39 people or something. And how did that go?

So yeah, we posted on this convalescent serum. I mean the idea of taking blood from somebody who's been infected and recovered and putting their antibodies in somebody else, what's the state of that in terms of the science? I know you're surveying the field there.

Yeah, so on Friday I believe the first convalescent serum therapy clinical study was published and the authors claim it was the first in the world published, or at least the largest. And so they did treat 39 patients like you said with convalescent serum therapy and it was controlled. So there was a group that did not receive that treatment and there was a significant improvement in survival with the group that did receive the convalescent serum therapy. There was about a 12.8 percent death rate for the convalescent serum therapy group but it was 24.4 percent for those who didn't receive it. So that means that the serum therapy cut the death rate in half and that was for people who were pretty close to death to begin with.

And I think I saw in the article that they just assumed it would work better if you had it earlier but it hasn't been tested. I guess now is there any reason to think that any of this would ever be dangerous? It doesn't go through, it wouldn't have to go, I know this is sort of outside your area, but I don't think this would have to go through the whole FDA approval because it's their antibodies. It's something we know well enough that we don't need to go through the whole thing, right?

Well it is actually FDA approved now, at least with an emergency use authorization. So that was one of the first things that became approved as a treatment for COVID-19 because it is generally safe. There's a couple of different things people generally test for like blood typing. You want to match the donor and recipient and then Rh factors. But other than that it appears to be safe.

So what would happen if you didn't match? Would it hurt you or would it just not help you? Suppose you had the wrong blood type match.

Yeah it could definitely hurt you down the road. It has been done without blood type matching. In fact when it was first used, the reports I've seen when it was first used are from the early 1900s and they were not doing blood type matching at that time but all those reports showed significant improvement.

Well if you had to guess based on, you know, isn't there two other cases where they've used this convalescent blood serum approach? Right, was it HIV was one?

Well I don't think it was used, or I'm not aware of it being used for HIV, but the treatment for HIV is monoclonal antibodies and same with Ebola. And so it's similar in a way because convalescent serum therapy is basically using antibodies. It's more of a kind of gross product of antibodies where you're just taking everything under the sun that's in somebody's serum and giving it to somebody else and you're not selecting for one specific antibody. But it has been used in SARS. There is actually a guy who got the 1901 Nobel Prize for the development of serum therapy. He solved a diphtheria epidemic in Germany in the 1890s by using serum therapies and that's kind of where it started. And so he's the pioneer in the field. His name's Emil von Behring.

Do you know what it would take to scale this up? Like let's say that in a few weeks we said yes this is the thing, we need more of this. Given that you have to actually take it from people unless you're using machines to clone it I guess, how quickly can, I know you did some calculations on how quickly you could scale up. What would you guess? I know this is an unfair question off the top of your head but let's say if we were to start today and say all right we want to get everybody convalescent blood serum therapy, like how close are we to being able to do that? Is that something that scales up very well?

I think that is generally reserved for the people that are in poor condition, in the hospital. So not everybody is going to need it but for those people it would be possible. It's obviously going to be dependent on your area. New York got hit so hard that I don't know if they would have been able to keep up with demand. But generally speaking you have about 5% of cases who end up in the ICU and those are the people that would be candidates to receive the therapy.

Would they really? Because why wouldn't you give it to people even before they have it so that if they did get it they could handle it more easily? Just because of scalability it is tough. You need to do plasmapheresis to get the plasma from donors. You also have to be checking people ahead of time to make sure that they actually had an antibody response against the virus.

Let me ask you some sort of detailed questions. If you're going to draw blood from one person and turn that one person's blood into antibodies, how long would the whole process take from putting the needle in to draw the blood to having it in your hand and ready to give somebody else?

I think that you can do plasmapheresis in a couple of hours. It depends obviously how much you're trying to harvest. The amount that the FDA allows I believe for a donation is a liter per week or two liters every two weeks. And then in the study that you were talking about they used about 500 milliliters which is a half a liter per patient. So it looks like if one donor giving one donation, over a two-week donation period, could handle two people.

One donation. If I donate blood once, how many people would that handle? So that's good. That's kind of hard to scale, isn't it?

Yes exactly. And so that's why I say it should be reserved for only the people that are in poor condition or in the hospital.

Now if you looked into the, I don't understand the differences or the nuances of monoclonal where you're just taking the best antibodies you can get from one real person and then you're cloning it with machines to scale it up. Is there a limit to how quickly you can do that if you just keep building machines? Couldn't we do a ventilator type thing where we just say all right everybody's building monoclonal devices? If you looked into those, do you know?

So they're not. Those are the first therapies that will come online for COVID-19 is monoclonal antibody therapies. It takes a little bit of time to develop because of the process. You build what are called hybridomas and that's a fusion between a cancer cell and an antibody creating cell and the cancer cell makes it immortal and then the antibody secreting cell is fused and it starts just producing antibodies en masse. And then you can create a ton of those cells. They're all identical and make a bunch of antibodies in a bioreactor and that's being done now. It just takes a little bit of time to identify the specific antibody secreting cell you want to use that's making good antibodies.

Now once you've identified the antibodies isn't it just a question of making more of those machines and more of those facilities? Couldn't you just keep scaling it up?

Definitely, correct. Yeah you could, especially if somebody identified a good antibody secreting cell and then made the hybridomas and if they were to share those with other people and kind of decentralize the production and make it scale faster that would maybe be a good way to do it. There's also, you should be probably combining a lot of different hybridomas to get good coverage for your antibodies, make sure that there's good protection there.

Have you heard, I haven't seen any news coverage of the monoclonal. Do you know if anybody's getting any traction actually doing that?

I haven't seen anything that's become available yet with those. I've seen some anecdotal reports about strong protection there. I think last week there was a report from a company that said they had a good candidate but it's still a little early. And the other thing is that there's a difference between having an antibody that you can see recognizes COVID-19 and then also whether that antibody neutralizes the virus. So I've actually seen some evidence in my hands where somebody can test positive for antibodies that recognize the virus but those antibodies don't necessarily neutralize the virus. And then conversely somebody could have a low level of antibodies that recognize the virus but those antibodies are very strong at neutralizing. So there is nuance there.

And do we have any more visibility on how long antibodies would last? Let's say you got them either naturally by having it and recovering or you got it from the plasma from somebody else. Do you know? Is there a difference in how long they would last?

That's a good question. I don't. You might expect there to be a shorter window for donor antibodies because they might start to get marked as non-self by your immune system and then kind of removed. But it's a tough thing to study because you start to get this mix of potentially antibodies made by yourself and antibodies from the donor and how do you differentiate which one came from which? But I think generally like two years is what a lot of people say. If you look at data from the original SARS patients, two years of protection is what people are getting at least. Some got protection out to seventeen years.

Now would you say that it comes down to literally just how fast we can make this stuff? I mean that's really the gating factor, right? Because at this point we kind of know it's safe and we kind of know it works or all common sense says it works. It would be hard to imagine it doesn't work.

Yeah it definitely works and one of the limitations is just getting access to the people who are producing those antibodies so that then you can kind of identify those antibody secreting cells and scale this. So I would encourage anybody that's ever had it and recovered to look into getting tested for antibodies and look into donating their serum for convalescent serum therapy because you could donate every week or every couple weeks and that serum can be banked for the future where maybe in the fall we have a second wave or something. So I think we should be stockpiling serum now in anticipation for the fall and winter.

Kind of interesting. So there's, I'd never heard before, which is we have all these strategic stockpiles of everything from oil to PPE and ventilators. It would make perfect sense to have a national stockpile of convalescent serum. That's actually a great idea. I hope somebody's watching this. We can do something about that.

Yeah, and that requires people who have had it to get tested for the antibodies and be kind of active participants. I think that a lot of people are looking for something that they can do to help nowadays and that's one pretty easy way to do it. And I think that those people should get paid to donate their serum. A lot of them have probably accrued pretty significant hospital bills along the way and this would be a way to offset those bills for them.

Now you're talking about how this might be a money-making thing. Are you aware of anybody who's ever charged for their antibodies for any kind of situation? Does that ever happen?

I'm not aware of a specific instance where somebody's donated their antibodies because they just had the greatest antibodies out there. But there's plasma donation centers all across the country. I don't know the name of one offhand but I know there's national chains of plasma donation centers and they pay people.

This is very interesting because the patriotic mood in the country at the moment is so high that you could get all the blood you wanted basically if the President of the United States said here's the deal, if you've tested positive I need you to go in and offer your blood. We're going to try to see how much we get. I believe that patriotism alone would give you all the blood you wanted. But if you really wanted to take it to the next level, people are people, and if you could offer more money to let's say go in a second or third or fourth time as you're suggesting, I would imagine that a monetary incentive would probably have a gigantic effect.

Yeah, and I've talked to some people about the monetary incentive and people get a little bit uneasy sometimes about that but my perspective is that it's going to be administered in the hospital and is the hospital making money? Yeah. So why shouldn't the donor?

Exactly. Now so we've got safety. It's really just a production problem it seems to me. And it's almost like gosh it feels like this is such a big deal. The biggest deal of course is wearing masks but I would think that the second most promising thing, if we have to wait for vaccines for months and months, the things that we can do now, the second most promising thing is probably this. What do you say?

In terms of things that we know that work and not having to wait around for something that may never actually happen. There's no guarantee that we'll have a vaccine or anything like that but we know this works and so why not embrace it while we also pursue other avenues to solve the problem.

I was just going to point out, so I started a company that we're going to be doing antibody testing as well and I'm specifically doing a neutralization test which means we're not just testing for the presence of antibody, we're testing to see whether your antibodies neutralize the virus. And we plan to take that data and publish the de-identified data on open source dashboards for anybody to check out just so everybody can get a general sense of where their community is in terms of relative protection. And then I plan on creating an avenue for people who test positive for neutralizing antibodies to be sent information on how to donate their plasma. So we'll be identifying those candidates and so then they should be told here's what you can do if you're interested in helping other people with your unique situation or unique set of antibodies.

Context —

So I've been watching the comments and I just want to call out something that I always talked about: winning attitudes and losing attitudes. And I'm watching a lot of losing attitudes in the comments, specifically people who don't quite appreciate, and I'm going to make them appreciate it now, the power of your talent stack. Because people are sort of saying, oh data analyst, what's that got to do…

Next segment → →