Yeah, the HSA + high-deductible combination definitely has an effect on low-hanging fruit like primary care stuff. I don't know if it's been measured, but it intuitively makes sense. In terms of health care inflation, however, this is small potatoes. When a heart attack runs ~$80K for a course of treatment... well, primary care is pretty small. A normal course of cancer treatment (chemo + radiation) can run $200K and up. In this respect, controlling costs at the PCP's office has very little effect on overall health spending trends.
Information asymmetry is significant, but it's most significant at the specialist level. How is a patient to know that long-run (5 yr) outcomes between surgery (we're doing something!) and physical therapy (hard, time-consuming, no sense of immediate satisfaction) are identical for something like herniated discs? And perhaps even better if you factor in potential surgical complications? So there's definitely that aspect to consider. Also, an orthopod is incentivized to do the surgery because we exist in a fee-for-service cottage industry. This doesn't mean they'll always recommend surgery, but there's always the financial incentives to consider. We reward more care in the US, not better care. But this isn't exactly news.
To go back to the primary care type information problem... this is solvable to some extent. The Internet has certainly brought a huge volume of distilled knowledge to the masses. WebMD and the Mayo Clinic site are excellent examples of this. Unfortunately, they aren't very good at providing physiological context when explaining mechanisms of action (whether we're talking disease processes or their treatments), and this results in poor consumer understanding of what's going on in the big picture. When I was in pharmacy school, this was a common problem when counseling patients. It was quite common for an engineer to understand how this one drug worked on this one problem, but they didn't know how it fit into the big picture that is their body and how it might affect their body's homeostatic balance, especially if they did X or Y while taking the med. It didn't even occur to them to find out. This is where these information sources aren't enough, and where expert knowledge from someone who thinks about this stuff all the time comes into play. People have a frighteningly cavalier approach to taking pills... even smart, educated people. I suspect it's a cultural thing.
Anyway, this lack of context problem is particularly exacerbated in psychiatry, neurology, and any other specialty where what is unknown is greater than what is known, especially when you start taking into account non-physiologic factors like cultural context.
WRT guidelines: they're always changing as new information becomes available. The definition of obesity is broader today than it was 20 years ago. (And accounts for some of the increase in incidence.) Even guidelines for "simple" stuff like ear infections (otitis media) change with some regularity, if you follow organizations like the AAFP.
The fundamental difference with health care in the US is that we ration based on ability to pay, rather than need. I know more about health care from both the ground-floor clinical side and big picture economic side than I know about probably anything else. If I were an investor, I wouldn't invest in health insurance startups or pharmaceutical startups. There's simply too much risk relative to the potential upside. I think this is to be expected in an industry where the low-hanging fruit has largely been plucked, though. What's left are niche diseases which aren't huge profit centers, as a rule.
I'd actually love to know if there's any meaningful data on health care startups vs pure technology startups WRT return on investment, broken down by type of health industry.