I came across this article from Jeff Lemieux who used to be the head staffer for the Commission on the Future of Medicare under Breaux and Thomas.  (The commission never went anywhere, but that surely wasn’t his fault).  The article is called The Curious, Counter-Intuitive Relationship Between Medicare Costs and HMO Enrollment and basically says that the more Medicare recipients joined HMOs the lower overall Medicare spending went, and that the relationship was partially causal.  When I first skimmed it I wanted to throw something at the computer screen.  But now I’m not so sure.  Take a look at it and I’ll give more comments in an update later in the morning.  It’s very relevant given PDIMA’s passage last year.

UPDATE:  OK, the site appears to be back up after an unexplained hosting glitch on Friday, so here goes for my take on this piece. 

Lemieux makes two major arguments. (Please read his piece if you can as this precis doesn’t do it justice) and they are well argued and (probably) empirically unprovable as the regression analysis equation required to figure out what caused what would be damn long. Lemieux himself says that there was so much going on that it’s hard to parse out.

First, he says that political pressure from conservative Republicans trumpeting Medicare HMOs as a solution to all the ills of the Medicare program caused liberal Medicare officials to subtly cut spending in response to show that the mainstream Medicare program could cut costs too.  He says that this attack on spending started in advance of the Balanced Budget Act (BBA) in 1997 that specifically went after hospital spending in Medicare and caused overall Medicare costs to actually decrease for a brief period of about 18 months. So his argument is that more Medicare recipients going into HMOs will cause the FFS Medicare program to innovate and compete by cutting its own costs.

Second, he says that the accepted wisdom that the Medicare Risk HMOs only recruited healthy seniors was wrong because if that had been the case, the increase in average cost of those staying behind in the Medicare program plus the 95% of average cost doled out to the Medicare HMOs as premium should have been reflected in an increase in the overall cost of the program. But instead at the time when Medicare HMO growth was at its height Medicare costs increases headed down not up. (I hope you’re following)

Both of these arguments may have some grain of truth in them, but equally both can be refuted.  For example, a significant amount of the reduction in Medicare cost growth came as the home health care program was frozen in place in the mid-1990s because of the massive amount of fraud in it.  Now that fraud had been going on for the better part of a decade (and was partly to do with laundering drug-trade profits in Florida).  Why did Medicare fraud become such a big deal?  Ed Hughes used to give a talk in which he said the reason was that all the FBI agents who had been assigned to chasing Russian spies in the cold war found themselves in 1992 with nothing to do, and so they seized on health care fraud as the next big boom area for their services. (If you think that’s far fetched consider that the FBI was a main mover behind the banning and demonization of Marijuana in 1937 after their work chasing bootleggers ended with the repeal of Prohibition).  So given that since 9-11 the FBI has got other stuff on its mind perhaps the increase in Medicare costs since then reflects that all the fraudsters who’d stopped taking it to Medicare in the mid-1990s are now back.

Is that the real answer? Almost certainly not, but there are several other factors and there’s actually been some real research on a related topic by Lauren Baker at Stanford. Baker found that from 1990-4 in areas where there was growing HMO penetration (which tended to be in the same places where there was highest Medicare HMO penetration too), overall health care utilization in the Medicare program fell slightly.

The standard interpretation of this is that physicians can only practice one style of medicine at a time. So when enough managed care comes to town–meaning the number of HMO patients goes from 10% to 20% or higher–they start practicing more conservatively with all their patients including those in Medicare FFS. So you could argue that Lemieux is right about HMOs causing a decline in Medicare FFS costs, but it’s all HMOs not just Medicare ones, which never got above 15% of the Medicare population and only got above 30% of the Medicare population in a few big cities, mainly in the West.

Lemieux’s other analysis is that the notion that only healthy Medicare recipients were going into HMOs was false.  The GAO report on this used data from the early 1990s.  Well it was pretty true then, hence the profits of HMOs like Pacificare that specialized in Medicare HMOs–reflected here in its stock price versus the S&P index–went up very fast until the growth in Medicare HMO enrollment slowed after 1996. After that they started getting to saturation in the markets where they were strongest out in the west, which meant that their mix probably did look like the overall Medicare populations. Their remaining growth was in the big cities in the East where Medicare payments were much higher and for a while the HMOs could still make money by hammering the hospitals on price and admission rates.

However, even if the HMOs were not recruiting healthier people, claiming that risk selection in Medicare HMO recruitment had anything to do with the overall costs of Medicare is a big stretch.  Don’t forget that Medicare spends 50% of its money on 10% of its people and 80% on 20%.  And the numbers in HMO’s never got much beyond 15% of the whole Medicare population.  My guess is that the vast majority of very expensive cases got treated the same way in both FFS and HMOs, as the costs of the really expensive cases were probably re-insured out by the Medicare HMOs, and they never had a sufficient number of the really expensive ones to try to change what happened with their care.

So what was happening with the care of the expensive folks?  Well costs went up much faster for Medicare than in the private sector in the early 1990s because it took HCFA longer than the private sector to figure out that they could beat their suppliers down on price.  The employers and HMOs figured it out in 1993-4, HCFA didn’t start being so aggressive until 1996 and really not until the BBA cuts took effect in 1997-8–hence the sharp fall in Medicare cost growth in 1997-9 as shown in Lemieux’s chart. By then of course the AHA got their people in Congress to adjust the rates, first on the side issues like rehabilitation care, and then overall.  At the same time America’s hospitals were merging like crazy to gain market share to enable them to raise prices to the private sector, while the HMOs were being beaten up in the court of public opinion, and gave up trying to cut expenditures. Of course drug costs (which as you know if you’ve been awake in the last 2 years) are not covered by FFS Medicare were going up like a rocket and were a big part of making Medicare HMOs’ margins shrink, which is one reason that their enrollment stopped growing as they often stopped offering cheap or free drug coverage. Not surprisingly they found that their members left in response.

So the artificial price controls of HCFA didn’t last long, and the artificial price controls of HMOs didn’t either.  And when market power switches back to suppliers, costs go up fast. Public costs and private costs take turns in going up fastest–but they all go up in the end. That’s the same pattern that Uwe Rheinhardt has been talking about in his core speech for years. My contention is that the Medicare HMO issue is somewhat irrelevant to the overall expansion of Medicare costs, and will stay that way for some time.

However, Lemieux is a supporter of innovation in Medicare and so am I.  He thinks that a combination of private plan innovations and mainstream response to that innovation will create the chronic care management, improved patient care and process innovation that I think we both agree is necessary to change the program. To that end he’s arguing that Medicare HMOs and other private plans will over the long run save costs. 

He may be right and they may be the best source of innovation but I think what he’s missing is the real reason that lefty Democrats are so opposed to the private sector expansion into Medicare. They are all decrying the privitization of Medicare, but I don’t think that’s the real problem.

The real issue for the future of Medicare is whether it will be defined benefit or a defined contribution.  As the AAHP’s own propaganda survey shows more money from the government means a higher "contribution", and hence the ability of the private plan to offer more at a lower premium cost to the senior. The real fear of the those in the left who care about equity isn’t so much that the private plans themselves will destroy Medicare or not provide promised services, or even that the people left in FFS Medicare will really be that much more expensive.  Scratch them very hard and you’ll see that their real concern is that once a large number of seniors are in a private plan and there’s a widespread acceptance of a voucher-type defined contribution system, at that point, defined contribution could be mandated. Then the mainstream plan will just become another option where the senior can spend their voucher. And of course the "contribution" from the government will be means tested as of 2006 anyway (rich seniors will be getting less of a "contribution" to Part B premiums after then).

That’s when things get really worrying, as the voucher now becomes a form of government welfare, and of course that can be cut if things get tough on the budget side.  So eventually it’s not impossible to foresee a time when the Medicare recipient gets a voucher that can only buy the overloaded public FFS plan, or a bare-bones private plan.  And of course the better-off seniors can trade up with their own money to a better class of plan.  That’s the nightmare scenario for the Democrats, and it’s hard to see  a way to definitively avoid that happening in the current legislation.

To paraphrase a concept Lemieux introduces elsewhere, we shouldn’t be focusing on whether the public sector MacDonalds or the private sector Burger King makes a better cheeseburger unless we’re damn sure that the senior of the future will be able to eat in the same resturant as everyone else and won’t be abandoned outside on the sidewalk as many were before 1965–and as the uninsured are now.