Roche just announced they are sub-licensing Tamiflu broadly. WSJ picked up the story (subscription required) and noted that some countries aren't waiting and have allowed generic production infringing Roche's patents.

I was able to obtain some more Tamiflu today here in Austin at my local People's Pharmacy. While there is apparently tremendous pressure on Roche at the international level, it looks like the rest of the supply chain has not yet picked up on the coming shortage and telegraphed the price rise. Gas prices these are not.

With no human to human transmission yet, it is hard to produce a vaccine because we do not know what the final pathogen will look like as it has not mutated yet. The risk is that it will spread quickly, but another risk is that it will not spread at all unfairly delegitimizing everyone who raised the warning.

It's a lot to ask people to st0ckpile their own Tamiflu (40 doses is about $300 enough for two acute courses if you show symptoms or 40 days worth of deterrence). But it lasts for three flu seasons. Spending $100/person per year would be $30 billion/year. Roche might part with a license to sell at a few cents a pill in those volumes and the post office distribute it getting the price down to a few bucks a person a year.

But who will st0ckpile it for you if you don't do it yourself? All it takes is 1% of families to buy to make personal st0ckpiling bigger than Roche's US sales in a single flu season. There were only 13,000 prescriptions last year. So if 1% of families bought demand would be increased by a factor of a hundred. Then maybe the wheels of government would move to build some more "push packs" for flu and not just bioterror.

While we are talking good public health policy, maybe we could use Tamiflu prophylactically every year in hard hit regions and not wait for bird flu. This and other measures like more widespread vaccination outreach may even cut the tens of thousands of deaths from regular flu seasons down to the 160 of a typical hurricane season. It would also give the Center for Disease Control good practice.

Wouldn't widely using Tamiflu every year just produce resistant strains?

Wouldn't widely using Tamiflu every year just produce resistant strains?

I'd have to agree that using Tamiflu for mundane flu outbreaks probably will develope resistance strains. It doesn't sounds quite as bad as a regimen of anti-biotic treatment (which can last a couple of weeks), but you'll still have compliance problems with people not completing their treatment. Second, as I understand it, it mitigates not cures. So something will survive which could pass on any potential resistance to Tamiflu.

Rather, I think we need to take a more intelligent approach here. There are certain categories of people who are not only at high risk of catching the flu, but also of passing it on to many, many others. For example, medical care professionals, police offices, teachers, airline ticketing agents, etc. Targeting the people with high numbers of infectable contacts (or at least getting these people to stay away from work while they are sick), would probably do a lot to restrain the spread of any flu.

"Wouldn't widely using Tamiflu every year just produce resistant strains?"

If 1/6 of American's get flu shots and 10% take tamiflu if they are in an identifiable cluster, sure, flu would become more resistant.

I knew someone who put down rugs on top of car rugs so that the rug below wouldn't get damaged.

Why not use the tamiflu anyway? The harm of using it is dominated by the benefit from using it. That is, what good is a treatment that is effective but few can use? Would you rather have a 50% effective treatment helping 1,000,000 people or a 100% effective treatment helping 10,000?

From instapundit, a request to calm down a bit on all of this:

http://instapundit.com/archives/026224.php

The point of this note is that most deaths due to flu are from post-flu infections, not the flu itself. Our ability to fight post-infections have gotten much better since 1918.

That Instapundit entry is most unfortunate. Instapundit later linked to Grand Rounds, which had a link to this page on Medipundit, which points out that for avian flu, the deaths have been due to direct viral lung damage, not secondary infections.

Think about it: don't you think they would have tried antibiotic treatment on these desperately ill patients (50 percent of which are dying!) if it would help, or that they tried it and it failed?

Tamiflu is not an antibiotic. It works by blocking an enzyme that the virus needs to destroy the host cell and release the new virii invade other cells. So a resistant flu from use of this drug isn't very liable to happen.

Bryan: viral resistance to antiviral drugs is quite common. All the virus needs to do is mutate its targeted protein in such a way that the compound no longer binds tightly to it. Viral replication is quite error prone, so this happens quickly.

For HIV, this resistance mechanism means that treating a patient with a single antiviral drug is useless -- the virus population in the patient rapidly mutates and adapts. The breakthrough idea was treating the infection with multiple antivirals at the same time, so that the chance of a virus mutating so it could bypass all of them was sufficiently low.

Thanks for the links, Paul! Good info on direct dam

(continued)age to the lungs.

Sam,

Why not use the tamiflu anyway? The harm of using it is dominated by the benefit from using it. That is, what good is a treatment that is effective but few can use? Would you rather have a 50% effective treatment helping 1,000,000 people or a 100% effective treatment helping 10,000?

Your assertion is unfounded. I think there is good reason to believe that the cost of using tamiflu indiscriminately outweighs the benefits, just as is the case with antibiotics. As I mention earlier, it seems to me that you can be a lot more effective by targetting people who are most likely to pass on flu infections.

Remember several things. First, a tamiflu treatment runs around $300, right? Second, that flus with higher resistance would diminish the value of that investment. And that even for the worst flus, the majority of people don't come down with it. So using tamiflu as a prophylactic is expensive. Ie, tamiflu is really useful only when you first catch the flu.

There are studies which indicate that even the $10-20 dose of flu vaccine isn't economically justified. I'm not sure if I buy that, but if it's so difficult to demonstrate the worth of universal vaccination against the flu, then I don't see widespread tamiflu use justified since it is considerably more expensive and somewhat less reliable (right?).

Paul,

Think about it: don't you think they would have tried antibiotic treatment on these desperately ill patients (50 percent of which are dying!) if it would help, or that they tried it and it failed?

I agree here. Further, I imagine that antibiotic treatment is standard for really bad flu cases. After all, secondary infections are quite common even with the normal varieties of flu. Seems a shame to survive the flu only to fall to an obvious danger like a secondary infection.

70,000 fatal cases of flu * 7,000,000/life (the average people place on their own life for the purpose of these sorts of calculations) = $490 billion

divided by 300 million doses = 1633/flu season. Old folks might not place as high a value on their lives, but even at 1/7 the typical person they get a 10x return on their investment in a flu shot based on lower risk of death. Lost work and productivity is not counted here.

Suppose prophylactic use of tamiflu reduces risk of death 0.3% during a fatal flu pandemic. Then the typical American would pay $21,000 for it if they had the money. $300 is two orders of magnitude lower in price and depending on where you are effectiveness percentage might be two orders of magnitude higher.

Can you point out any studies that say vaccination or prophylactic treatment is not economically justified?

Ok, I guess I never picked the right google search term before. I'm getting several studies that strongly support (in at least the at risk groups like children and the elderly) flu vaccinations. This study claims that universal vaccination of children would have a net benefit. Apparently, this study indicates that pre-schoolers are primary drivers of flu epidemics.

Elsewhere, we have calls (from Mayo Clinic doctors) for universal vaccination of the elder. Finally, there are calls for universal vaccination of the entire population.

More specifically, I looked at workplace studies. One study indicated that there was mild economic benefit to an employer (not just the employee!) to have workplace vaccinations.

OTOH, we have a study that indicates a substantial increase in the resistance of human flus to certain cheap antiviral agents in the past few years.

On your post, Sam:

70,000 fatal cases of flu * 7,000,000/life (the average people place on their own life for the purpose of these sorts of calculations) = $490 billion

Society doesn't place the same value on people's lives that people place on their own lives. Still involuntary risk of death (eg, as from the flu or third party harm) is probably within a factor of two (IIRC) and hence not significant.

Flu vaccination is probably justified. It apparently reduces the severity of the flu and its complications. That means you are less likely to infect others and reduces the harm you personally suffer. Further, the flu vaccine is tailored to the current versions of the flu and more likely (IMHO) to remain effective.

Widespread Tamiflu use for normal flus appears unjustified though. You have to balance the cost of the treatment, and the reduced effectiveness of it in the future against its benefits. I think the current scheme of using Tamiflu for high risk cases (eg, when people are already exhibiting early flu symptoms or are caring for someone infected with the flu) is justified.

I have never argued that widespread use of Tamiflu during a particularly lethal flu pandemic was unjustified.

Would you rather have a 50% effective treatment helping 1,000,000 people or a 100% effective treatment helping 10,000?

I'd feel better knowing that 10,000 people who are actually sick have a 100% chance of surviving and not re-transmitting a disease, rather than have 500,000 people walking around thinking that they're immune and therefore get sick and/or transfer a disease to other humans. In the case of a major outbreak, a treatment that is 100% effective has a much better chance than a 50% effective treatment at actually stemming the spread of a disease.

There's too much guessing about whether the avian flu virus will mutate into a human virus. I have neither heard nor seen any any statistical calculations as to what the probability is.

In my consideration, it's equally likely that the virus will mutate into a form that dies. But nobody talks about that.