I don't understand vaccines please help

[Captain.Sassy]

I'm really confused by this paper:

http://www.plosone.org/article/info:doi/10.1371/journal.pone.0004915

"In fact, if the vaccine-resistant strain has higher virulence than the vaccine-sensitive strain, the vaccination program is always effective, even though the program engenders the emergence of a vaccine-resistant strain. On the other hand, if the vaccine-resistant strain has lower virulence, we must carefully manage vaccination to prevent the spread of a vaccine-resistant strain."

Here's what I understand [after skimming two papers on strain replacement and part of the wikipedia article on the subject] and please please correct me where I’m wrong cause I’m trying to understand this:

Viruses compete, like any other organism (yeah they are barely life forms but they kind of are). Different strains of the same virus are in the same 'niche' and so compete with each other. If you are sick with one strain of the virus, it makes it harder for another strain of the virus to infect you (except I just learned today that this isn’t the case for dengue fever, and possibly H1N1 though the results aren’t conclusive on the latter).

Virulence is a measure of how quickly a virus (or strain of a virus) can reproduce itself and transmit itself. A virus that reproduces itself in your body more will make you sicker than a strain of the same virus that doesn't reproduce as much in your body. So usually (from what I gather) a more virulent virus will

- reproduce faster
- transmit from host to host faster
- make the hosts sicker

So generally, the more virulent strain of the virus is the stronger one, and will be the dominant one, since it infects more hosts and weaker strains can't take these hosts over. So when you vaccinate against a virus, vaccines are usually developed for the stronger strain. Vaccinated people get antibodies against that strain, and the strong strain is prevented from spreading. This gives a weaker strain an opportunity to spread.

This all seems to makes sense to me, but I must have missed something in the reasoning, which is what I want someone to sort out for me...

Cause I think that article I posted uses a mathematical model of an epidemic to model what happens when you vaccinate against a virus, and that model says that if you vaccinate against the more virulent virus, the less virulent strain of the virus can in fact cause a larger epidemic than the more virulent virus would have, without a vaccination program [though maybe I misread the paper].

If you are developing a vaccine against the dominant strain, wouldn't the dominant strain be the more virulent strain always? And if the strain you vaccinate against is more virulent than the strain that emerges after vaccination, don't you reduce the size of the epidemic? How can a less virulent strain infect more people than a more virulent strain?

Is there a way to intuitively understand the model in the paper in the kinds of terms I'm using (if I'm using terms wrong please correct me) or is the result they talk about an artefact of some very specific parameters in their model?

I'd be really thankful if anyone could answer this stuff for me. Sorry for spamming about vaccines today.

[Dancing David]

Because the people who are more resistant to one strain will be infected for a shorter time and therefore infect fewer other people? So the non-vaccinated virus does infect more people for longer?

[blutoski]

There are different kinds of viral strategies for 'success'.

Generally, a virus that is too lethal will not infect many people, as the host dies too quickly. The less fatal strains tend to dominate over time, as they are the ones with living hosts.

[Dymanic]

Your question might really be better expressed as something like: "Is there a way to intuitively understand mathematical modelling?".

I don't think there's any simple answer to that. It's going to vary from one person to another and from one model to another. Some aspects of some models may be more intuitively obvious than others, etc. The waters you're sailing on may be deeper than you realize. If it's something you find interesting enough to deserve some spending some time on, I'd recommend starting here:

http://scienceblogs.com/effectmeasure/2007/03/modeling_antiviral_resistance_2.php

(The specifics deal with antiviral resistance rather than vaccines, but the modelling itself is really the central focus).

[Captain.Sassy]

Thanks for the suggestion Dynamic. I'll check this out. I have some exposure to modelling (mostly economics) but maybe I need to learn more about it in the context of epidemiology.

I think this is my problem though:
if you vaccinate against the more virulent virus, the less virulent strain of the virus can in fact cause a larger epidemic than the more virulent virus would have, without a vaccination program

If I need to fully understand the model they use to understand this part of the paper, so be it. I'll give it a shot.

[Captain.Sassy]

There are different kinds of viral strategies for 'success'.

Generally, a virus that is too lethal will not infect many people, as the host dies too quickly. The less fatal strains tend to dominate over time, as they are the ones with living hosts.


OK, so the model they use suggests that the death rate of the more virulent strain is high enough to stop the virus spreading to as many birds as the less virulent strain? I guess that kind of makes sense, but then how would the less virulent strain's prevalence be affected? The more virulent strain kills off its hosts, and can't spread, and so can't 'crowd out' the less virulent strain. I don't see how vaccination against the more virulent strain would affect this, unless it's by keeping the birds that would have died from the more virulent strain alive so they can transmit the less virulent strain. Is that right?

[Dymanic]

One of the fun things about this is that, if you really look at it, when you attempt to approach it intuitively, you ARE attempting to model it mathematically; it's just that you aren't going about it in a very systematic way. And even when you are going about it systematically, it's hard to avoid the necessity to make some intuitive judgement calls (how much weight to attribute to each variable, how to initialize those variables, etc). Certainly there are some places where you can simply defer to empirical data, but there are also some places where you can't. The tools used to collect data have improved considerably since the last time there was a major flu pandemic, and future models will benefit from that, but it doesn't do us all that much good right now.

The reveres have also written some excellent (and much shorter) pieces clarifying some of the basic terminology:
http://scienceblogs.com/effectmeasure/2009/07/transmission_pathogenicity_vir.php

A lot of people (including quite a few medical professionals, unfortunately) use "virulence" and "pathogenicity" more or less interchangeably, and sometimes just flat-out incorrectly. It can help to get clear on all that before you get too deep into this quagmire.

[Captain.Sassy]

Thanks Dynamic that was interesting. I think I now understand the difference between pathogenicity, virulence, and transmission.

pathogenicity- you show symptoms
virulence- how sick you get
transmission- how easily the disease spreads

Unfortunately I still don't really understand how these three variables are related to the 'dominant' and 'crowded out' strains of a virus.

I haven't had a chance to check those other links you posted, I'll take a look as soon as I get a chance.

[Dymanic]

Unfortunately I still don't really understand how these three variables are related to the 'dominant' and 'crowded out' strains of a virus.
No, that's good. Claiming to understand all of this is a bit like claiming to understand quantum mechanics, I think.

It gets much worse, because there are a lot more variables than just those three. And some of them are essentially composities of other variables. The same virus that is easily transmitted between birds of a given species may also be easily transmitted to or between members of another species; and it may not. Or, it may cause serious illness in most members of one species and produce no readily apparent symptoms in another, etc. Nuances of species physiology (or even individual physiology) play a role, as does the immunological history and profile of the host population. What we perceive as "illness" is not merely something a virus does to a host; it is something a virus and a host immune system do together -- and in the same sense, what we perceive as an epidemic is not something a virus does to a host population but something a virus (a viral swarm, really) and a host popuation do together.

I have to say I'm pretty confused by the article you linked as well. It's not hard to grasp the basic idea that once enough members of a population have been infected with the same virus and recovered (or not), the virus will have a tougher time finding new susceptibles to infect, and that the way influenza handles this is by shifting its antigens. They seem to be making a special case for immunity aquired through vaccination rather than infection, and I'm not getting that (among a bunch of other things; they don't even have a "recovereds" variable? WTF?). It might become clearer on a more thorough reading, but I'm not quite ready to make that investment in time right now.

[kellyb]

This is a really complicated topic, and no one really understands for sure how it works yet.
And it's very importent to remember that every pathogen is different, so you can't extrapolate from one disease to another.

Influenza is probably a particularly bad pathogen to start off with if you're curious about this stuff, too. (it's epidemiology is is probably the most mysterious out of all the well studied diseases.) So the best anyone can do is make a well informed, but essentially wild guess.

[Captain.Sassy]

Gotta say thanks for all this, I really appreciate you guys all taking the time to help!

[Skeptic Ginger]

pending review of what I posted

[UnrepentantSinner]

There are different kinds of viral strategies for 'success'.

Generally, a virus that is too lethal will not infect many people, as the host dies too quickly. The less fatal strains tend to dominate over time, as they are the ones with living hosts.

A good apples and oranges comparison for this is Ebola Ziare and HIV. Both are fatal, but Ebola Ziare is more easily transmissible, exhibits symptoms faster and kills faster so it has not spread as pervasively as HIV has.