Now and again there’s a real world case that reminds you why the only solution for keeping private health plans is managed competition, with the emphasis on managed. Remember as you read this story that Alain Enthoven always said that there should be community rating, with standardized benefits between plans and risk adjustment between them to override the impact of chance-driven uneven risk selection.

So the story (hat-tip to Rick Byrne for this) begins with Part D, which allows significant disparity in benefits between plans—something that it’s claimed causes few problems for enrollees. What happens next is that one plan with particularly rich benefits finds that it is adversely selected against. But this doesn’t become clear until late in 2006. Meanwhile Sierra Health Services, a Nevada based for-profit HMO with a pretty good record at cost containment, thinks that it can launch a PDP (stand alone drug plan) that covers the donut hole and has rich benefits. It has to file the paper-work by mid-2006, it charges a hefty premium and it waits for the money to roll in. 42,000 seniors sign up.

Fast forward to last week when on this conference call (audio here) (transcript)
Sierra Health’s CEO Anthony Marlon told Wall Street that on that one
PDP it lost $3m in January alone (nearly $80 each!) and is on course to
lose a lot more over the rest of the year. What the hell happened? Well
the plan was so rich in benefits that, even though it was a high
premium plan, the people who signed up were even sicker than they
expected!

DR. ANTHONY MARLON (Sierra CEO): Utilization
assumptions that actuaries used did not anticipate the level of adverse
membership selection we experienced. It now appears that the actual
level of adverse selection is driving significant variations from the
utilization assumptions we used in our bid.

And now we get to the fun part. Remember that other plan that
started in 2006 and found that it was losing money? It apparently
suggested to its more expensive members that they might like Sierra’s
plan better!

DR. ANTHONY MARLON (Sierra CEO): We
believed that our level of adverse selection was in part due to certain
high utilizing members being referred to us by another PDP provider. We
did not agree to these referrals. We have been told that some, if not
all, of these referrals by another PDP provider, we approved by CMS;
again, without any notification to us, or approval from us.

And perhaps CMS wasn’t entirely neutral here!

MATTHEW BORSCH
(Goldman Sachs analyst): Got it; okay. And, just on the question of the
activity of referrals from the competitor into your plan, if you can
just refresh me, if somebody knows, what are the rules that are
stipulated about identifying members and encouraging them to move to
specific competitor plans? Are there stated rules around that?

DR. ANTHONY
MARLON (Sierra CEO): No, there are not. But the only thing we have
learned is that one region of CMS gave permission to the competitor to
do this without our knowledge and, apparently, without the knowledge of
the central office. We will investigate this and try to find out more
about it.

MATTHEW BORSCH:
And just on that last point, is there – is there a region where this
has been particularly an issue? Or, is it pretty much throughout where
the product is sold?

DR. ANTHONY
MARLON: Well, I think one region gave the competitor permission to do
this and it has affected our offering in all of the areas where it’s
offered.

So
you’d think that a smart plan like Sierra would know something about
making money off Medicare patients! And apparently they still do,
except for this one little wrinkle.

CHARLES BRADY: I see. And, was there an enhanced product available with your MA plan as well?

DR. ANTHONY
MARLON: We have – well, for the years that we have had an MA product,
we have always had an enhanced product. We have always had a full drug
program that covered drugs in their entirety through the doughnut hole.
So, we’ve
been at this business for the better part of a dozen – no; 15 years.
And, you know, that’s what gave us at least some comfort, together with
the outside actuaries, when we priced this other product. Obviously, we
were wrong. And we’re trying to find out why we were wrong.

CHARLES BRADY:
And, as you look at the population, those individuals that you enrolled
that were, you know, riskier than expected, those individuals are in a
stand alone PDP and, as you look at the drug trend for your MA PD
customers, you’re not seeing unusual selection bias there. Is that what
you’re saying?

DR. ANTHONY
MARLON: The second part of your question is absolutely correct. That’s
a fact. In none of our other programs are we seeing this kind of
adverse selection. But
let me explain something to you. I know nothing about the 42,000,
40,000, whatever it is, members in this PDP enhanced product . All I
have, at this point in time, is the claims information. And that’s what
I’m relaying to you. If I had some of the information, I may have been
able to take some action. But, we’re trying to find out. What
we’re relaying is that the claims information looks bad for January and
half of February. And, we’re trying to be transparent about it. We’re
trying to get the information out there.

CHARLES BRADY:
All right; I completely appreciate that. That’s very helpful, your
doing this call. And, you are able to tell – well, you say you know
nothing about them. Are you able to tell that they’re in a standalone
product and not.

DR. ANTHONY
MARLON: They’re in a standalone product. They’re in our standalone
product. But when you ask me, you tell me if they have high utilization
of MRIs or Cat Scans, I don’t have that kind of background on this
population.

CHARLES BRADY: Right.

DR. ANTHONY
MARLON: Frankly, I don’t know, given HIPAA, that I’ll ever get this
information on this population. I just know that they use an
extraordinary amount of drugs and, when I look at the list of drugs, it
is unfamiliar to me. It is nothing like I have ever seen before on any
population. <snip> No, what I said was, when I looked over the list of the top 25 drugs, it’s not like a list I’ve ever seen before.

CHARLES BRADY: But is that – is it more biologics or -?

DR. ANTHONY MARLON: More biologics.

In other words a bunch of sick people using very expensive
biotech products (presumably to treat cancer, kidney failure et al)
were shunted over to the new PDP. And the worse news for Sierra is that
they are holding this bag, and can’t drop it until 2008. Which they
will do then!

GREG NIRCESSIAN, ANALYST, CREDIT SUISSE FIRST

BOSTON

:
I just had a couple of quick questions. Tony, you mentioned the action
plans; just wondering, two questions. One, what is the – what are the
rules around, you know, shutting enrollment, or closing enrollment to
new members in this product? And
then secondarily, you mentioned, you’re evaluating your legal options.
Would those legal options be targeted at the referring plan? Or, at the
agency that, you know, authorized the plan referrals? Or, what are you
suggesting there?

DR. ANTHONY
MARLON: I’m not suggesting anything. What I’m suggesting is that we’re
examining all alternatives. I can’t tell you what we’re going to do,
how this is going to come out, you know. The
issue is there are no written rules by CMS on any of this. We cannot
close enrollment to the plan. We have to live with this. We can’t shut
it off. We can’t turn these people out loose. The issue is, we’ve got
work with CMS to try to mitigate the damage. And that’s what we’re
doing. And, I have no idea of what I’m going to be able to accomplish.

So likely there is no recourse and a certain competitor just passed on a huge financial loss to Sierra.

Which brings us back to the problem. If you run a guaranteed issue
plan like Part D but with voluntary enrollment, if you have
differential benefits you will attract adverse selection. And you even
if you think you priced it right, you might just find that your
competitors will take advantage!

The simple way to fix this would be to have compulsory enrollment
(so healthy people as well as sick ones are in the universal pool), and
standardized benefits, so everyone gets the same thing and no insurer
can play games on benefits or with formularies. Then you need risk
adjustment payments on the back end in case one plan does get lucky or
unlucky with enrollment. (You’d do that by testing a sample of all
enrollees as they go into the plan).

Then the insurers could concentrate on improving care management and
not worry about making money playing games and shifting risk. After
all, that’s apparently not a game they always win!