A Different Way to
Regulate Drugs?
Many of you probably know already that I am a cynic when it
comes to the drug industry. The problems
emerge from several factors:
- The incentives are royally screwed up. Pharma R&D spends tens of millions developing drugs that may or may not work. So not only is there a huge temptation to push the boundaries with the results, there is also a big risk of self-delusion that the results are better than they really are.
- The research publications industry has problems of its own, generally only publishing significant results. So all of the studies that find a drug doesn’t work never make it into the public eye. We are biased from the get go that a drug is better than it really is.
- The health care industry has problems too – doctors are so rushed seeing patients that they definitely don’t have time to read a ton of research studies. So they have to rely on the pharma companies to tell them what a drug does.
- Drugs are sold by salespeople working on commission. They clearly have incentives to push drugs a little harder than perhaps the drug merits. Not necessarily deceptive sales practices (although unfortunately that happens too), but the unconscious power of incentives is extremely strong.
- The real world is complex. We never really know if a drug is working once it is released. Controlled trials are controlled for a good reason.
- There are several self-delusion effects among patients. Our passionate hope to get better leads us to want drugs, even if they have questionable effectiveness. Even to the point where doctors will prescribe drugs just to shut us up (yes that happens, perhaps also to prevent lawsuits from dissatisfied patients). The placebo effect makes us feel like the drug is working even when it is not.
- The FDA is not without some of the responsibility. FDA rules get bogged down in politics. Regulations get watered down in the promulgation process. FDA inspectors are too few and not paid enough. And if it is missing data that concerns them, there is no way to inspect for that.
All of this results in billions of dollars of spending on
drugs that may or may not work and we will never really know. For example, one of the few follow-up
field studies found that anti-depressants might reduce depression by 10-20% in up to
35% of patients (the numbers are based on memory, but it was something
“depressing” like this). And that is if
the patient is willing to try them all until they find one that works for them. If not, the numbers are even lower.
The Libertarian Mind
This situation came to mind as I was reading David Boaz’
book “The Libertarian Mind: A Manifesto for Freedom.” I believe that small “l”
libertarian principles are often the most effective ways to govern and live.
Boaz is a bit of an extremist along this line and I don’t think I recommend the
book unless you just want 400 pages of go-go libertarian sermonizing.
The extreme position when it comes to things like drug
regulation is that the market and Adam Smith’s invisible hand will take care of
it. Pharma companies can release drugs at will.
Doctors know this, so they will not prescribe just anything. Drug
companies will have to build up reputations that their drugs work. If not, no
one will buy their drugs and they will go out of business. Unethical behavior will be made public and
those companies will disappear too. By
eliminating the millions of dollars in testing, companies can try out many more
drug candidates and we will end up with better drugs.
But this depends on two things that I don’t think exist.
First is that he assumes that the drug companies have read Adam Smith’s earlier
book “The Theory of Moral Sentiments” that instructs people that they should
behave ethically as well as in their self-interest.
All of the reasons I described above make this pretty unlikely even if
drug companies want to.
The second is the asymmetry of information. The efficacy and
safety of drugs is not discoverable by the market and its invisible hand could
not work its magic.
A New Way?
So what if we turn the whole process on its head? What if drug companies were required to
publish all of their data in real time into an open access database? The FDA’s budget could be redirected on
creating the database so that it is open source and platform independent so that
it would be easy to create analytics for. A
new industry for the big data analysis would emerge to parse the data and
demonstrate which drugs work and which don’t.
Deep, sophisticated, extensive programs would be expensive, but would be
worth it for a hospital or large medical practice to use. Smaller systems that
are more like search engines could be developed for patients through smartphone
apps. Independents could create trusted
repositories that list what drugs work, how well, for whom, and what the side
effects are.
It would take time for this to emerge so the FDA would have
to keep its current system in place in parallel for a while. But there is
already an open source drug data movement that is growing in strength so some
of the work has been done. As the cost
of health care increasingly strains the economy, there is more of a desire to
do something different.
Thoughts?
