Pharmaceutical innovation is very, very good

From Frank Lichtenberg:

We examine the impact of pharmaceutical innovation, as measured by the vintage of prescription drugs used, on longevity, using longitudinal, country-level data on 30 developing and high-income countries during the period 2000-2009. We control for fixed country and year effects, real per capita income, the unemployment rate, mean years of schooling, the urbanization rate, real per capita health expenditure (public and private), the DPT immunization rate, HIV prevalence and tuberculosis incidence. Life expectancy at all ages and survival rates above age 25 increased faster in countries with larger increases in drug vintage. The increase in drug vintage was the only variable that was significantly related to all of these measures of longevity growth. Controlling for all of the other potential determinants of longevity did not reduce the vintage coefficient by more than 20%. Pharmaceutical innovation is estimated to have accounted for almost three-fourths of the 1.74-year increase in life expectancy at birth in the 30 countries in our sample between 2000 and 2009, and for about one third of the 9.1-year difference in life expectancy at birth in 2009 between the top 5 countries (ranked by drug vintage in 2009) and the bottom 5 countries (ranked by the same criterion).


People play some pretty intense games by claiming things like the "quality of life" is better within the modern houses and urban lands than it was in the 1950s so we should be grateful that our potential wives and daughters are embittered corporate concubines throwing the fetuses of our unborn into the dumpster, (assuming the sperm and egg can ever get together to begin with) and our brothers and sons are shooting drugs if they aren't getting HIV anally injected. It's all "choice" divorced from the economics of vital statistics and basic biological territory.


There is some sort of truth in there somewhere. I often wonder whether the purpose of choice is choice or the purpose of choice is to find the right choice.

Roger Sterling wants his schtick back.

Now let's talk about how little innovation in drugs there has been in the past few years, despite the fact that Americans pay much more for brand drugs than any country in the world. I'd say we're not getting our money's worth, at least not relative to other countries. This lack of innovation is largely the reason that Medicare Part D did not cost what it was estimated it would. Few new blockbuster drugs have hit the market since it was passed. Stagnation, great.

Alex would like to project this problem as being an FDA issue, but it is not. If it were, we would see drugmakers developing innovative products for introduction in other countries, going where the regulation allows.

Maybe we pay more BECAUSE the other countries pay less?

That's certainly the major factor and has been well documented. However, Jan is correct about the slow down in innovation (and total absence in some areas such as antibiotic R&D). It is getting progressively more difficult to identify targets for drug development and the "discover the next blockbuster drug" mentality is still quite prevalent in the pharma industry. Additionally, many of the biotechnology drugs once touted as inherently safer because they are human proteins are proving to be anything but.

There is certainly a real stagnation due to the diminishing low-hanging fruit of small-molecule development and application. All that means is new development will be more expensive and require some paradigm and regulation evolutions. So, rather than us paying less it is time for other countries to step up and help out with the development costs.

It's my opinion that we (the U.S.) are the fools in this matter, not the countries that are paying less. And good luck making demands for increased payments to brand-name pharmaceutical manufacturers a foreign policy priority, no matter who's president next year.

Collaboration on a prize system for development of drugs that treat priority conditions (e.g. a new fourth line antibiotic) would make sense, but I don't have much hope that would happen anytime soon.

There's no money in antibiotic R&D. Why develop a drug for Africian River Blindess when the only people who would use it can't afford to pay the development costs. There's more money in giving a white guy a harder boner or a marginally better version of Lipitor than there is in curing infectious disease.

In any case, gaining life expectancy in developed countries isn't as simple as prescribing a drug. The major killers are, in this order, tobacco, obesity, and alcohol. Changing those requires changing ingrained habits, not popping a pill. I don't necessarily think that it's were innovating less, it's that there's trouble getting from lab results to public health results.

We do need new antibiotics. Resistance is a real problem. MRSA is increasingly common.


There’s more money in giving a white guy a harder boner or a marginally better version of Lipitor than there is in curing infectious disease.

When people dismiss "lifestyle" drugs like Viagra, I think it's important to remember that sildenafil was originally developed to treat heart disease. Similarly, minoxidil was developed as a treatment for hypertension. In the former case, it turned out not to be a blockbuster, in the latter case the hair-growth side effects were too unwanted. In both instances, lemons were turned into lemonade.

It's also worth mentioning that both sildenafil (and the "me-too" compound tadalafil) are both used to treat hypertension as well as ED.

How does this make sense in marginal-cost terms? If other people paid more, would your own fees thereby become less out of the goodness of the seller's heart?

It's a network business. You have to consider the development cost of the network. Marginal sales are pure gravy. That's why they can give it away for free in Africa (for example). "Why shouldn't we pay nothing, just like Africa!?!" makes the answer more obvious.

Maybe we shouldn't pay more.

No, that's not how it works. Monopolists charge monopoly prices. which aren't really a function of the size of the market. What would happen if we were paying more is that drug innovation would be more profitable, so we'd get more innovation. People are dying so that Europeans can have cheap pills.

So you think Apple would raise it's prices in the US because people in China pay less?

There is a case to be made that other countries are able to pay less because we pay more, but if those countries paid more I really doubt producers would suddenly start pricing below market in the US.

If we concede that you are correct, then Americans are going to get big discounts on drugs going forward since there are no new blockbusters being discovered, and the pharmacopoeia will increasingly become generic. Right?

That would be a disaster. No hyperbole.

I don't disagree, but his comment makes, seriously, no sense to me.

It makes perfect sense to Obama.

Since the first or second most profitable and prescribed (in dollar terms) therapeutic drug category has been atypical anti-psychotics for the last 4-5 years I'm inclined to doubt the loss to society of new blockbuster drugs would be as painful as you imply.

Obviously pharmaceutical research is important, but if drug companies have ceased to innovate due to a lack of low hanging fruit or because more profit is available with copycat drugs or via promoting widespread over-prescription of psychiatric medication, then perhaps we need a new paradigm?

Absolutely correct and for many medical conditions we really don't need any new drugs as the generics (and the ones to be) represent cutting edge therapy. You see many big pharma companies getting out of certain product areas such as cardiovascular (which arguably are the biggest volume selling drugs) and moving to oncology R&D. This is certainly what I observed when I retired from the industry two years ago. It also explains by Part D drug benefits are going to cost a lot less going forward. there is nothing wrong with this happening from a medical care perspective since the drugs in the formulary are all very good.

there is nothing wrong with this happening from a medical care perspective since the drugs in the formulary are all very good.

Oh, really? People stopped dying of heart disease, did they?

No, but there are no small molecule therapies that would confer significant additional benefit in treating or preventing heart disease that Part D beneficiaries don't already have access to. Gotta develop something for people to buy.

That's what he said: Medicare Part D was cheaper than expected because the old blockbuster drugs are going off-patent and there are no new ones to replace them.

Not discounts per se, but basically drug costs will go down. Americans don't pay excessive costs for generics, just branded drugs. As the patents and market exclusivity expire we will, in aggregate, pay less. That assumes no innovation, which seems to be the case in the near future.

Alternatively, we could increase our subsidies to drug companies to drug companies on the assumption (hope?) that they'll innovate. How much more money is appropriate?

Yes, unless companies innovate--which they do not appear to be (let's not count 'me too' drugs or drugs that may be patented or marketed, but which do not confer any additional benefit or safety). Right.

Alex would like to project this problem as being an FDA issue, but it is not. If it were, we would see drugmakers developing innovative products for introduction in other countries, going where the regulation allows.

But there's not enough money to be made in those other countries because of price controls, lower per capita incomes, and smaller populations. The US market is where all the money is, so the FDA has a wildly disproportionate influence on the economics of drug development.

I assume the 'life-expectancy at birth' numbers are estimated based on a range of other factors. Is this true? And if it is, doesn't that cause problems for using it as a dependent variable? (Maybe these issues are addressed in the paper - I don't have free access.)

I would assume life-expectancy at birth is based mostly on the ages people are dying at today.

One might also read this and see it as a failure, so much money spent for such tiny and unequally distributed longevity gains, especially in light of the SENS research program (and others like it).

You refer to the bona fide public good of anti-aging research?

According to the FDA's own statistics, prescription drugs kill 100,000 in the US per year. I wonder if they took that into account.

This is largely a result of medication errors (wrong dose, wrong drug, complications from multiple drugs) and not inherent in the drug itself IF it is being administered for the labeled indication. It has an impact on health care costs but is really not relevant to the Lichtenberg study.

If referring to effects as side effects or errors make you feel better about this particular kind of slaughter then you go, girl. BTW it stands to reason that the FDA number is a lower bound.

This abstract should have read:

"By ignoring the patent cliff, we conclude that the pharmaceutical industry is highly innovative."

It doesn't say that we have enough pharamaceutical innovation. It says that on the margin, more is good. Which means that we don't have enough.

But healthcare would be more affordable if we stopped innovation, and truly enlightened world citizens know that's the best outcome for everyone.

I'm wondering if Tyler meant to title this post "Pharmaceutical Innovation WAS Very, Very Good" since larger increases in drug vintage (the length of time the drug has been in the market) were responsible for the good effects.

On reflection, it is entirely unsurprising to find that pharmaceutical innovation often helps people live longer. I'd also be curious to know how much of the living better (especially when much older) can be explained by this too.

It is a wonderful thing to know that the general quality of pharmaceutical options is better, but this empirical demonstration nevertheless makes me uncomfortable because it so easily lends itself to the notion that these innovations can perhaps be attributed to specific features of the American medical and pharmaceutical industries (such as market structure, surrounding regulatory structures, public and private health coverage) that emphatically work very poorly for too many Americans, as evidenced by pathetic average outcomes for important health indicators.

Could these innovations have happened in an environment that provided significantly better health and drug coverage to all Americans, or if the regulatory and tax environment had perhaps been conducive to some variant of market and regulatory structures in the pharmaceutical industry in the USA? Does this sort of stuff help explain why the USA spends so much more on health care despite its comparatively poor average results for some key indicators?

I note that the research was supported by Pfizer. While there are the usual caveats that Pfizer had no influence on results, there are plenty of meta-analyses out there showing that funding source matters. There are a myriad of detailed but important choices in correlation studies of this sort, each of which may be subject to bias. In particular, it is important that the author tried hard enough to control for everything else: the motivation to do so is of course lower when you already have eyecatching and sponsor-pleasing result. In this case, why keep searching for better alternative controlling factors, non-linear effects etc.? Model building is an art, and details matter.

Comments for this post are closed