If we don’t completely replace the current system (in the way I was suggesting), then your ideas are ones I generally like. In your model, since you speak of “premiums” and the “gatekeeper” function, I take it you are imagining HSAs not completely replacing insurance, but rather, serving as a complement to insurance (so that people can use their HSA funds to purchase insurance). Is that correct? If insurance is still going to be part of the picture, I do, as I suggested earlier, like Rand Paul’s idea to allow people to form associations (whether through church groups or clubs or any other form that works) that buy insurance together, and this then addresses the big issue posed by pre-existing conditions and adverse selection and, thus, significantly reins in premiums (it’s the same thing that happens when you get insurance through work, essentially). Both the HSA idea and the Rand Paul idea can co-exist.

I completely agree with you about the need to bring back some sort of system for authorizing testing and follow-up with specialists, etc. (what you call the “gatekeeper” idea) … though, of course, if someone wants to ignore the gatekeeper and pay more, they should be free to do that.

One major concern I have that remains unaddressed by what you propose is my issue of doctors authorizing all sorts of unnecessary procedures because they have the incentive to do so (due to reimbursement). I think this is a critical issue to fix, because it not only makes the system more expensive but also, as I described (citing data), is a leading cause of death in the U.S. Here is a Harvard Business Review article presenting some ideas in addressing this issue. One thing it links to is this white paper which has a bunch of interesting suggestions on this front, including reimbursing physicians at a higher rate when they authorize procedures/interventions in circumstances where their use is evidence-based, bumping up physicians to a higher reimbursement level when they show a pattern of achieving quality results (obviously difficult to gauge) and a few other ideas. One measure that might be useful here is forcing patients themselves to absorb more of the cost of treatments (i.e., higher co-pays), especially when those treatments are not evidence-based in the circumstances at issue. I’m not sure what the best solution is here, but I do think it’s essential that this issue be addressed.