Science! (Clinical Trials)

You know I'm a 100% "science guy".

Part of that, in my opinion, is being able to self-assess.  To say "nope, this isn't working", evaluate alternatives, and change... even if it's a deep, core change.

I see a problem with clinical trials.  (And, more broadly, the same logic could be applied to general science, but I leave that to the reader.)

There are a lot of possible therapies out there.  Billions of compounds, biological and otherwise, that have yet to be tested against tens of thousands of human (and livestock, and agriculture) ailments.

As we sciencey (that should SO be a word) people know, in order to conduct a trial with any kind of legitimacy, it must be done double-blind, randomized, and must have a sufficient sample size (depending on the rarity and uniformity of the condition, this should mean between several dozen and several thousand subjects).  Proper clinical trials are complex* and expensive. They are slow to start, difficult to recruit for, take a long time to run, take much longer for proper followups, and--perhaps worse than all of these things--are mired in political motivations and opinions.

There is no way to systematically and categorically test every drug for every potential use.  ...Even if you hone out the clear losers, there are still billions of possible trials today, and as we learn more, that number can only increase.

Thus, there needs to be discussion about which therapies get priority over others.  ...This is where the system is flawed: it is largely dictated by the superior capital of giant drug companies.  What they want to bring to market will surely get undue focus.  This is also to the detriment of "cheap" solutions (think exercise, relaxation techniques, dietary changes): there is nothing to "gain" (monitarily) from these trials, so there is a lack of funding for them.

Compounding the problem is the small trials that take place before the larger ones: they are doomed to failure if they don't produce positive results, so are particularly prone to fudging of numbers, or less rigorous techniques, or poor randomization.

In a nutshell, what I'm saying is that the selection process for the scientific method governing clinical trials is anything but objective, and much less than effective.  This is not an area where "free market mechanics" is properly applied: it's science, and there needs to be methodology to it.

* Not to appeal to authority, but I worked three years on programming software for clinical trials.

No comments: