THE TREATMENT OCTOBER 21, 2009
This is the second installment of our new feature: Curbside Consult. For the uninitiated, curbside consults are a venerable medical tradition, whereby a doctor seeks informal advice from an experienced colleague in treating a patient with a complex condition. In covering or understanding complex health and social policies, we need sometimes help too.
Today’s interview is with Katherine Swartz, PhD. She is Professor of Health Economics and Policy at the Harvard School of Public Health. Author of Reinsuring Health: Why More Middle-Class People Are Uninsured and What Government Can Do, she has researched state efforts to cover the uninsured and to help people with costly conditions who fare poorly in the private health insurance market.
Swartz is a particular expert in the economics of high-risk pools (HRPs--sorry for the acronym) and reinsurance. If you follow health reform, you’ve probably heard these terms thrown around without a lot of discussion of what these terms actually mean.
To put it simply, HRPs are special insurance plans for people who couldn't get coverage on their own, because of costly pre-existing medical conditions. Reinsurance is probably best explained as "insurance for the insurers." It's a special fund that reimburses insurers for the super-high claims of catastrophic medical expenses. The idea is that this spreads the burden of those expenses as widely as possible, while removing (at least partly) the incentive insurers have to avoid risky beneficiaries.
Both parties propose HRPs and reinsurance to address the immediate and long-term challenges in financing care for people with costly conditions. President Obama mentioned HRPs in his address before Congress to offer quick help to people facing the dual challenges of uninsurance and serious illness. Such provisions are contained in the Baucus bill and in recently-proposed Republican amendments. The Bush administration made modest investments in HRPs. Senator McCain’s 2008 health plan made heavy use of them, as do current Republican proposals presented in the House.
I am no fan of HRPs, although the Senate Finance bill envisions them only as a stopgap measure. For one thing, the financing seems murky at best. Finance would allocate $5 billion to support HRPs before insurance exchanges can get up and running. A Republican proposal would allocate $2 billion per year between 2010 and 2019. No one really knows how many Americans have costly illnesses that would occasion such help. We do know that the budget numbers don’t add up.
An August 2009 Government Accountability Office (GAO) report indicates that about 200,000 Americans are now covered in such HRPs. These men and women have an average household income of only $41,000. They pay higher premiums than private insurers typically charge healthy individuals. Individuals face average deductibles exceeding $1,500 and lifetime expenditure caps. People also endure waiting periods before they can enroll.
To serve 200,000 people, HRPs expended about $1.9 billion in claims last year, about $9,400 per person enrolled. The GAO estimates that another four million Americans would be eligible because they are uninsured and experience costly health problems. The leading bills allocate, at-best, $1,000 per year per person.
Given everything else in health reform, you probably haven’t thought or heard much about these issues--hence the need for this curbside consult. What follows is a shortened and edited excerpt of my September 17 interview with Professor Swartz.
Pollack: I should start by asking a basic question. Few readers really understand the difference between reinsurance and HRPs. Can you just say a little bit about what the difference is between those concepts?
Swartz: High-risk pools, as they have been used, serve people who are perceived or predicted by insurance companies to be likely to have very high medical costs in the next year, usually because they've had high costs in the last couple of years. These people are ceded by the insurers to the HRPs…. Sometimes people are denied coverage because of the prediction. Other times, people are charged such a high premium (in states where insurers cannot deny coverage) that they can't afford it – so in effect, they've been denied coverage. The point is that high-risk pools currently are used to cover small numbers of people who are predicted to have very high medical costs in the coming year.
The reinsurance idea, which hasn't been used much, is more of a proposal. The idea here is that after a year is over, we will know with certainty if someone has had extremely high medical costs. And at that point, you use a pool of money to offset much of the costs of these people who definitely had extraordinarily high costs. So, in the first case with the HRPs, people are there because it’s anticipated they will have high medical costs. In the second case with the reinsurance, we know for sure that a person had medical expenses that landed them in the top one percent or the two percent of the entire population.
I want to add one more thing. HRPs to date cover fewer than 200,000 people each year. And they exist only in 35 states. A number of those states actually have closed enrollment to new people.
Pollack: Is that because they've just run out of money?
Swartz: Yes. And they've been closed for a long time. So, any idea of using HRPs in a major way to cover many of the uninsured people even if you just wanted to cover people that you think are going to have very high costs, that's a lot more than 200,000 people. And we need to restructure and refinance how those pools would be run. I think that's part of why the Senate Finance Committee put $5 billion in there. But I agree with you. I don't think $5 billion is enough money.
Pollack: If you ask how many people these risk-pools can realistically serve, the numbers get very small. That was something that came up in the campaign with McCain's proposal, which was also underfunded.
Swartz: Yes. I coauthored a piece with Sherry Glied, Thomas Buchmueller, and Anne Royalty where we argued that HRPs won’t work without a lot of money.
Pollack: The Baucus proposal requires people to be uninsured for six months before they can gain access to it. What happens to people? If you think about this prospectively, it means that we are requiring that people lose their insurance or not have insurance and go six months with potentially costly and deadly diagnoses before we are willing to address their needs.
Swartz: There's so much evidence that when people don't have health insurance, they are not getting the same level of care as people who have health insurance. Suppose somebody has liver cancer, or let's say she had ovarian cancer 20 years ago. There are a lot of women who if they were diagnosed with ovarian cancer starting about 20 years ago, are lucky because we really changed how we treat it and many are now survivors. But, for the last 19 or 20 years, they are viewed as very high risk people. And if they were suddenly to lose their insurance coverage for whatever reason, and then have to wait six months, but during that time, a lump was discovered or some other evidence that maybe there's a new cancer, they are in trouble.
Pollack: Are there any positive lessons from the states about things that have worked well that we can learn from? If you could design this thing and make it as well crafted at least as feasible in the current structure, are there any lessons that we should draw from this from states?
Swartz: We learned a lot in Massachusetts from the requirement that people have to show evidence of coverage and working through how we define “affordable” and who would get exempted from the requirement. I think there are lots of good lessons out of what's happened in Massachusetts. It's gotten a bad press. In fact, the Massachusetts Taxpayers Foundation came out with a report in June saying that it's working pretty well, actually – the medical inflation in Massachusetts are not due to the increased number of people with coverage…. If you are going to get rid of adverse selection in this marketplace, you do have to have this type of personal responsibility requirement in place.
I think Vermont is another state where there are good discussions going on between representatives of different stakeholder groups. I don't know a lot of the other states quite as well, in terms of what they have been able to do in the last couple of years. I think people are much more thoughtful now about the risks that are involved in health insurance. For example, how do you share risk among everybody, rather than having one or two insurers end up with a lot of adverse selection.
I also think the other lesson that has come out, chiefly from Massachusetts and from Vermont… thinking through what is a minimal benefits package everybody should have. We’re trying to balance out the fact that if you add more required services to it, it's going to cost a lot more. What is it that we are really trying to insure? I think we are working our way towards coverage against catastrophe, where catastrophe is defined relative to somebody's disposable income, along with some cost-effective primary care services basically.
Pollack: How about specific lessons of state risk pools?
Swartz: Minnesota and Oregon are the only two that have large enough numbers of people that have been covered, but again, I don't think that the current structure of the risk pools is what we should be looking at if we are going to greatly expand them. They weren't set up for this. They were set up, really, to take very small numbers of people out of the insurance market. They weren't meant to be a substitute for public or private insurance.
Pollack: How about the reinsurance provisions in the various leading bills? I take it you believe that reinsurance would be useful in a state insurance exchange to address the really extreme cases that are going to come up.
Swartz: I think reinsurance is a way of more fairly and widely spreading the burdens of people who have extraordinarily high costs. It’s pretty random who lands in that top one or two percent in the population in terms of healthcare costs in any given year. So, having a broadbased population paying most of their costs makes a lot more sense to me than placing the burden on others who happen to be covered by that person's particular insurer or insurance policy.
Pollack: I was going to say there's a tension in that. People affected by reinsurance are, of course, the really high cost people. How do we manage the costs of their care without either having them dumped from the insurers or creating other unwanted incentives. I don't really know of a good answer to do that.
Swartz: Reinsurance offers some real advantages in managing the care of people who have clearly crossed some tripwire heading down the road to requiring very expensive care. You really do want somebody being the quarterback about what's happening with a person's medical care. I can tell you this from personal experience, watching what's going on with my parents.
For example, suppose you have someone with Parkinson's disease who has shortness of breath (a common issue with Parkinson’s). A 911 call causes the person to be taken by an ambulance to a hospital emergency room. Hospitals are generally terrible for people with neurological illnesses. Most of the ER people have little experience with their special needs or problems, and they respond to an older person coming in with shortness of breath by assuming it could be a heart attack. As a result, many expensive tests can be run during the first 12 hours to determine if the person had a heart attack or stroke. Meanwhile, hospital pharmacies generally don't stock all the meds that somebody with Parkinson's or other neurological diseases have. If the family or the person has forgotten to bring along his meds, the person just starts sliding downhill – and then the hospital is even more likely to suspect a stroke and to keep the person there for observation and more tests. Unless someone is alert to the Parkinson’s issues, a lot of money can be quickly spent on what I would argue are unnecessary tests.
I think when you have a reinsurance program--and you put that together with effective electronic medical records—there are better incentives to manage the care. Instead of calling 911, the family can be urged to call a nurse who knows the specifics of the case and can respond more effectively. That sort of thing could make a big difference in controlling costs.