American Enterprise Institute
December 19, 2005
[Edited transcript from audio tapes]
1:45 p.m. Registration
2:00 Presentation: Tom Baker, University of Connecticut Law School
2:45 Discussion: Ted Frank, AEI
Martin F. Grace, Georgia State University
David A. Hyman, University of Illinois Law School
Moderator: Jonathan Klick, AEI and FSU
4:00 Adjournment
Proceedings:
JONATHAN KLICK: Thank you for joining us today. This is an event by American Enterprise Institute's Liability Project. The Liability Project is a project that's meant to examine tort issues mostly from an empirical standpoint, starting with the presumption that both sides engaged in the tort debate have a bad ratio of evidence to advocacy. We want to try to remedy some of that.
Today we have an event that centers around a new book, a potentially important book, The Medical Malpractice Myth by Tom Baker. Tom Baker is the Connecticut Mutual Professor of Law and the Director of the Insurance Law Center at the University of Connecticut, and Tom will be our first speaker today. Thank you, Tom.
TOM BAKER: Medical malpractice premiums are skyrocketing. Closed signs are sprouting on health clinic doors. Doctors are leaving the field of medicine, and those who remain are practicing in fear and silence. Pregnant women cannot find obstetricians. Billions of dollars are wasted on defensive medicine. And angry doctors are marching on state capitols across the country, all this is because medical malpractice litigation is exploding. Egged on by greedy lawyers, plaintiffs sue at the drop of a hat. Juries award eye-popping sums to undeserving claimants leaving doctors, hospitals and their insurance companies no choice but to pay huge ransoms to be released from the clutches from the so-called civil justice system. Medical malpractice litigation is a sick joke, a roulette game rigged so that plaintiffs and their lawyers' numbers come up all too often and doctors and the honest people who pay in the end always lose.
This is the medical malpractice myth, built on a foundation of urban legend mixed with the occasional true story, supported by selective references to academic studies and repeated so often that even the myth makers forget the exaggeration, half-truths and outright misinformation employed in the service of their greater good. The medical malpractice myth has filled doctors, patients, legislators and voters with the kind of fear that short-circuits critical thinking.
Those of you who've looked at the book may recognize this as the opening from the book. Obviously I can't read the entire book to you today, nor can I summarize all the information that it contains. Instead, I plan to do three things. First, tell you why and how I wrote the book. Second, provide a taste of the myth-reality framework of the book. And third, and I think most importantly, described the evidence-based liability reform ideas that I propose.
First, why and how I wrote the book. Why did I write the book? I wrote the book after getting involved in the recent medical liability reform debates in Connecticut. I teach tort and insurance law at the University of Connecticut, I conduct empirical research on tort law and action, and I run the Insurance Law Center at the University of Connecticut School of Law, an academic research institution that's generally thought to be favorably inclined towards a free market. So it's not surprising that stakeholders wanted me to weigh in with the Connecticut Legislature.
When I did weigh in, I was shocked at the disconnect between myth and reality. Almost without exception, the hospital, doctor and insurance lobbyists and witnesses were reciting the myth with which I began. But of course they didn't recite it as myth, they cited it as reality. Yet as I know from my own research and writing and from the by now extensive empirical research on medical malpractice and medical malpractice litigation, the research refutes that myth. But most people, including most politicians and doctors, don't know anything about that research, so I decided to write the took at this time in order to bring a dose of reality to the debates.
How did I write the book? It was very easy. First, I made a list of all the myths, and then when I was halfway through the book, George Bush gave a January speech, a Collinsville address, in January 2005 that then provided a nice checklist so I can make sure that I had them all.
[Laughter.]
MR. BAKER: Second, I then systematically collected all the research on these topics, much of which I had already collected for some more specialized academic writing I'd done, and then I sat down and synthesized all this research into the myth and reality framework that drives the main empirical parts of the book, and that's Chapters 2, 3, 4, 6 and 7 for those of you who are keeping track.
In the process I realized that I needed to write a chapter that laid out of the positive case for medical malpractice lawsuits. That's in Chapter 5, which in many ways is the central chapter of the book, and not just geographically. Then I realized that I needed to set out an evidence-based liability reform program to counter the myth-based reforms that have been the main focus of the policy debates. My evidence-based reform program is in Chapter 8, and I'll describe that at the end of my talk today.
The second part of my talk: some myths and part of the reality. If you want the whole story you have to read the book and listen to what the commentators say.
Myth one: medical malpractice litigation is a bigger problem than medical malpractice itself. Reality: the real problem is medical malpractice, not medical malpractice litigation. One hundred-thousand people die every year from medical malpractice, only about 5 percent of the victims of medical malpractice bring a claim. For the rest of the story, see Chapter 2.
Myth two: the big increases in medical malpractice premiums in 2002, 2003 and 2004, resulted from an increase in the number of medical malpractice claims and an unprecedented increase in the size of jury verdicts and settlements. Reality: the big increases in medical malpractice premiums came from the boom and bust cycle in the insurance industry, not from anything, or not very much to do with medical malpractice litigation. The number of medical malpractice claims is steady or declining, and certainly declining relative to the size of our health care system. And the size of medical malpractice claim payments is increasing but roughly at the rate of medical inflation. See Chapter 3, and I'll talk more about this in a moment.
Myth three: medical professionals are overwhelmed by frivolous lawsuits. Juries and insurance companies regularly give eye-popping sums to undeserving claimants. Reality: all the closed-claim research shows a close correlation—not perfect but close—between payments and the merits of the claims. There is no such thing as a perfect system, but our litigation system does a pretty good job of weeding out weaker claims. The research also shows that the weaker claims are on the whole not frivolous. First, a significant but so far not adequately pinned down percentage of the zero payment claims that are said to be frivolous are not claims at all, at least not in the ordinary sense of that term. They're an incident that a doctor or a hospital reported to an insurance company pursuant to their contract, the insurance company opened a claim and then eventually when it was closed without payment 2 years later because nobody sued, it became a frivolous claim. Second, the best economic research shows that patients and their lawyers have to file claims in order to find out what really happened. The research shows that doctors and hospitals do not disclose errors, and they fight even valid claims. So a denial that there was an error is not credible to the patient. The patient needs the litigation process in many but not all cases to find out what really happened, and in my view, it's not frivolous to file a lawsuit when you think there might have been malpractice and there's no other way to find out for sure.
My favorite myth: medical malpractice litigation is an important reason why medical care is so expensive. Reality: medical malpractice insurance premiums and the costs of alternatives to premiums for hospitals and other large practice groups that don't buy insurance but, rather, buy alternative services, make a trivial contribution to our health care system costs, between a half a percent and 1 percent depending on how you count. There is a lot of talk about defensive medicine, but there is not research demonstrating that wasteful defensive medicine amounts to much more. Chapter 6 summarizes all this research.
Myth: medical malpractice premiums are driving doctors out of practice and depriving patients of necessary medical care. Reality: not so. Yes, in rural areas and in fast-growing regions of the country we do have a problem with supply of doctors, but this is really a long-term problem in the supply of doctors. While some doctors are saying that they are giving up some kinds of services because of medical malpractice, this so far as we've been able to show with research does not affect access to care. I'm going to just read a little section from my book that's instructive. This is from the testimony that Dr. J. Robert Galvin, who is the Connecticut Commissioner of Public Health, gave to the Connecticut Legislature. Dr. Galvin was asked, aren't we losing doctors in Connecticut, obstetricians because of medical malpractice premiums? And he says, that according to the statistical people in Physician Licensing, this is not true. He explains, physicians move around a lot, particularly younger physicians, and they move on to different climates and they move to different ends of the profession and so there's always some movement, but we have not seen demonstrable diminishment. He's impressed about obstetricians and gynecologists. “Some of it is perception, particularly with obstetrics and gynecology. In all my years as a practitioner, what I've found is as obstetricians and gynecologists get older, they all want to do gynecological surgery. They don't want to deliver babies by and large because it's difficult and time consuming and a lot of night and weekend work.” Dr. Galvin closed his testimony with a story. “I had an acquaintance who left to go to Florida and he said he didn't like it here, and he had a lot of reasons as well. What he didn't like was the group he was in. What he did like was the fact that his wife's parents lived in Florida and he had a Florida license. I had another professional colleague who left a very specialized practice to go into the ministry. It really didn't have much to do with malpractice or anything else except that physicians tend to change.” In other words, people move and leave, moving in and out of the profession, and in the time of a medical malpractice insurance crisis, it's easy to blame medical malpractice insurance as the reason, but the research doesn't show that that's a strong contribution, and Chapter 7 spells all that out.
Most of the book pulls together research that other people have done, some people on this panel. The one part of the book that I did a little work of my own on relates to what I'll call the medical malpractice insurance underwriting cycle, and I think I'm going to successfully in 5 minutes teach you how to read that chart so that you can see why I think that I've shown through this little chart that the medical malpractice insurance crisis isn't the result of sudden changes in the litigation system but, rather, a cycle in insurance.
There are three lines on the chart. The first one to look at is red, and on the red one you should look at the left axis, and that's operating profit. It's got a technical definition, but what you need to know is that when it's above zero that's good for the insurance industry, and when it's below zero that's bad. This is a calendar year picture, it's not perfect, but it gives you a picture in 1995 that was the second most recent insurance crisis, and then starting in 2000 and 2001, the most recent one. You can see there's a profit valley followed by higher profit years, and if you would have looked at 2004, the profit numbers are now above zero.
So the next line to look at is the purplish line that I call “predictive losses.” The technical term for this is “initial incurred losses.” This you should read on the right-hand side, and what that number reflects is the amount of money that the insurance industry as a whole predicted would be paid on policies sold in the indicated year at the time they sold their policy. For example, in 1980, when the insurance industry sold policies covering the year 1980, they predicted they would pay about $2 million worth of claims. Then similarly, if you look at the final year, 2007, when they sold those policies, they predicted that they would pay about $7 million worth of claims. The absolute dollar numbers here are not significant. What matters is the relationship between the lines.
What you can see, and I have a second chart in a moment that will illustrate this more clearly, if you look at the purple line you will see a more or less steady line that jumps twice. In other words, at two periods there's a big change in what the initial predicted losses are, otherwise the line is fairly smooth and the chart we'll look at next will show that more clearly.
The next line is the green line. I call that “actual losses,” and the initial term there is “developed losses.” That's a line that shows what we know 10 years after the fact that the insurance industry actually paid on policies sold in that year. For instance, in 1980, that's a number that you couldn't know until 1990. In other words, there are no crystal balls in the insurance industry, it's an uncertain business, but looking back after the fact 10 years later, we see that there was somewhat more than $2 million paid. In the years 1980 to 1985, the predicted losses were lower than the actual losses, then from 1985 to roughly 1997, that predicted losses were more than the actual losses, and the closer we get to the present, that green line becomes more a prediction because, obviously, we don't know in 2004 for sure.
But the thing to look at is the relationship between the lines, and what that shows is, in the years leading up these hard markets, these high price increase periods, the insurance companies were not charging enough money for their policies, and then in the years after there was some make-up that goes on. I do not mean to suggest anything nefarious about this, but that what happens when the premiums spike in 1985 and then happened in 2000, is a readjustment of the market to reflect the fact that prices had been too low, and because insurance companies pay claims out over time, when the market shifts in 1985, you have to, in a sense, make up for the fact that you have not charged enough for the last few years. In 1985, I can't go and charge a new premium for the policy I sold in 1982, but yet I'm still paying claims under the policy, so the only way that I can deal with that is either take money out of my assets or to collect new premiums.
To my mind what this shows is that what happened when the premiums spiked is that the insurance industry had to make up for the fact that it hadn't been charging enough both in 1985 and 2000. Here's another slide that shows this somewhat differently. The red line is the same. It's the operating profit. You've got the same valley in 1985 and 2001. The purple line here is still the predicted losses or, again, the technical term is initial incurred loss. But here what I've done is I've shown change in predicted losses. For instance, it shows that for each year I show the change in the predicted losses this year as compared to last year, adjusting for medical inflation. What this shows is that, except for in the hard market periods, the predictions of losses basically track medical inflation or are even a little below medical inflation. Translation, they're cutting their price every year because the main driver of medical malpractice insurance costs is medical inflation, for three reasons. One: health care costs are the largest component aside from pain and suffering, but it's the largest hard component of medical malpractice damages. Two: health care inflation is not a bad proxy for the exposure because, what causes health care inflation? Increasing growth of the health care sector so that in the health care sector we should expect more losses as the health care sector expands.
The basic point to take from this is you can see a picture of the insurance underwriting cycle, I've explained it in more detail in the book. The big picture point is that increases in medical malpractice insurance premiums in the 2001, 2002, 2003 period come from changes in the cycle, not from sudden changes in litigation behavior.
So far I've explained to you why and how I wrote the book. I've gone through some of the myths and part of the reality. And I'd like to spend the rest of the time making the case for evidence-based medical liability reform.
The evidence that I summarize in the book shows that the fundamental problem with medical malpractice lawsuits is almost exactly the opposite of what the medical malpractice myth would have us believe. The problem is not that there are too many claims, the problem is that there are too few. And because our health care system does such a poor job of giving injured patients the information they need to tell whether their injuries were due to malpractice, filing a lawsuit is too often the only way to find out.
An evidence-based medical liability reform program should have goals to track the real problems that the research has identified. The research tells us there is too much medical malpractice; therefore, the first goal should be to reduce the amount of medical malpractice. The research tells us that injured patients cannot adequately evaluate whether their injuries are the result of medical malpractice. Thus, the second goal should be to give patients the information they need to make that evaluation. The research tells us that most people injured by medical malpractice are not compensated. Thus, the third goal should be to improve compensation for medical injuries. Finally, the research tells us that the boom and bust insurance cycle causes the medical malpractice insurance crises that are so disruptive for doctors. Thus, the fourth goal should be to moderate the effect of this cycle on doctors. Just as there are four goals, I have four parts to my program.
First, a medical injury disclosure and enforcement process to give patients information that they deserve, and also to build a database of adverse events so that we know more about what happens. Second, an apology and early restitution procedure to speed up the resolution of the most serious existing medical malpractice claims. Third, a supplemental no-fault compensation program to address the low and moderate value injuries that the current system completely ignores. The research shows that it is almost impossible to get a lawyer to take a medical malpractice case that is worth less than $150- or $200,000, so the small dollar claims are completely ignored. Fourth, an extension of an existing private-market-based approach to medical malpractice insurance that will protect doctors from the worst effects of the insurance cycle, and I call that enterprise liability insurance.
Few of these ideas in general are truly unique to me. I've just pulled them together and put them in a book with a pretty cover. But doctors and health officials must disclose orally and in writing, and the writing part is important. If you are going to get a mortgage, they have to tell you in writing what your interest rate is. Believe me, interest rates are a lot less complicated than what wrong in a medical procedure. Orally in writing any adverse and possible adverse events: Here this isn't disclosing, just an admission that the outcome wasn't what it was supposed to be. Not any admission of fault or anything, so it's a disclosure of an adverse event and a possible event to the patient and to the Department of Public Health. That would be the right agency in my state, it might not be the right agency in another state. And to the patient's health insurer, and the disclosures to the agency and to the health insurer is because they're information aggregators and they can provide useful feedback as to what kinds of problems there are.
I have strong enforcement powers for the disclosure requirement. If you have a disclosure requirement without serious enforcement, you might as well not have a disclosure requirement. Then here, an added enforcement mechanism is that if you don't disclose an adverse event, that if you are sued, the adverse event will be treated as negligent in any medical malpractice claim. The enforcement and audit powers are designed to make sure that hospitals and other practice groups set up systems for reporting, but no system can make sure that everyone is reported because if I really did a negligent one I might not want to report it because once I report it, of course, the patient knows about it, and nondisclosure presumption helps that.
Apology and restitution procedure: This is a procedure that I borrowed and adapted from a legal ethicist called Lee Taft who used to be a plaintiff's lawyer and saw the better of his ways. People have probably heard of "Sorry Works!" It's a little different than "Sorry Works!" but the basic idea is the provider apologizes and offers restitution within a reasonable time after the injury. The key thing that I don't have on the slide is that the apology is an admission of liability. This is a naked apology that is an admission of liability. If the offer is refused and if the verdict is less than 20 percent more than what the offer was, the plaintiff receives the lesser of the offered amount and the verdict, minus the attorney's fees the defendant incurred from the date of the offer.
Part three, supplemental no-fault: These are provider-financed benefits and provider-financed through liability insurance for any patient injured by an adverse event caused by the provider, with basic benefits capped at $50,000. The idea here is to address entire classes of injuries that are completely ignored at present and some kind of administrative process modeled on Worker's Compensation or on Social Security Disability. Tort rights continue if you want to exert them.
Finally, enterprise liability insurance: Here the idea is to take the obligation to buy the insurance off the high-risk specialists and place it on a hospital or other institution that's better able to bear those risks and can negotiate with the insurance industry or other organizations for alternative risk transfer arrangements. The basic point is to place the obligation to purchase the insurance on an entity that provides a wide variety of services so that when the hard market hits, the high-risk specialists like the obstetricians or neurosurgeons are not hit with bearing all the costs of that hard market.
Let me just wrap up by explaining why all this is evidence-based. Each part of my program is evidence-based. The disclosure requirements are based on two kinds of research. First, research shows that doctors and hospitals do not in fact disclose adverse events. Second, on research that shows that if you provide patients better information, they will increase the accuracy of their claims. Patients who have better credible information that there was no negligence do not file claims as frequently as patients without information. Of course, conversely, we hope that patients who have good information that there was negligence will file a claim. After all, the incentive is the whole idea here. So it's evidence-based in that regard. Even more importantly, a mandatory disclosure will produce the kind of evidence that we need to reduce patient injuries and to improve the quality of health care.
The apology and restitution incentive is also based on research. First, the closed-claim research shows that well over half of the medical malpractice insurance payment dollars are paid in cases in which the insurance company's own experts determine that there is strong evidence of liability that the doctor was in fact negligent. That suggests two things. One, that these cases are ripe for an apology. Two, that a system that makes those cases move along more quickly will provide a great benefit. The second research supporting the apology and restitution incentive is a research project that I did with Northwestern University Professor Albert Yoon where we looked at fee-shifting statutes and we found that this sort of fee-shifting incentive actually works. In other words, if you place the burden of potentially paying the attorney's fees on the other side, that will encourage them to settle the case.
The no-fault concept is based on research showing that medical malpractice lawsuits almost always involve high-value injuries. Thus, a no-fault system with modest benefits would fill, in my view, a huge deterrence in compensation without interfering with the current tort approach.
Finally, the enterprise insurance concept is based on research on the medical malpractice insurance underwriting cycle. That research, as I said, shows that insurance crises result from the boom and bust of the insurance market, not from short-term changes in medical malpractice lawsuits. The research also shows that the burden of these hard-market shifts falls disproportionately on doctors with hospital-based practices. Then, finally, the idea of enterprise liability insurance is consistent with the emerging expert view that a systems approach is a better way to prevent medical malpractice.
The tort reform proposals that emerge out of the medical malpractice myth stand in sharp contrast to my evidence-based reforms. My favorite place to see these tort reform proposals is the White House website. The White House proposes, and I quote, "Curbing lawsuit abuse with needed medical liability reform." The reforms proposed include imposing a $250,000 cap on non-economic damages, unspecified limits on punitive damages, eliminating lump-sum damage awards in favor of periodic payments, shortening the period after an injury in which the patient may bring a claim, and eliminating the legal rule that makes each defendant in a case responsible for all the harm in the event that the other defendants do not have enough insurance or money to pay for their share.
With one exception, the White House proposals have a clear and well-targeted aim, to reduce the number of high-severity, high-damage malpractice lawsuits, and to reduce the amount of damages that a severely injured patient can collect. The one exception is punitive damages. Punitive damages reform does not have very much to do with true medical malpractice lawsuits, which almost never involve punitive damages. Drug companies and medical device manufacturers are the real constituency for punitive damages reform, not doctors and hospitals. This is the "hiding behind the doctor's white coat" phenomenon that I discuss in the book.
These reforms attack tort law rules that English and American judges developed over hundreds of years in a carefully considered effort to balance the interests of plaintiffs and defendants. Plaintiff's lawyers did not dream up these rules and impose them on the rest of the world. The rules emerged out of a long deliberative process in which both sides of every case had full opportunity to be heard. If anyone was at financial or tactical disadvantage in that process, it was plaintiffs and not defendants.
It's my not so humble opinion that not one of the White House proposals would improve the accuracy of medical malpractice claiming, increase the number of injured patients who will be compensated, or improve patient safety. Not one of the White House proposals will protect doctors from the next medical malpractice insurance crisis or provide real immediate relief for doctors who deliver babies. What they will do instead is increase patient share of the medical malpractice burden.
Of course, the White House proposal does not say that. Instead it says, "Frivolous lawsuits and excessive jury awards are driving many health care providers out of communities and forcing doctors to practice overly defensive medical. This reduces access to medically necessary services and raises the cost of health care for all. For these reasons the President has proposed proven reforms such as common-sense limits on non-economic damages to make the medical liability system more fair, predictable and timely." As I wrote in my book, this is the medical malpractice myth at work.
As the research demonstrates, not one of these factual predicates is true, not frivolous lawsuits, not excessive jury awards, not driving many doctors out of communities (some I would have to acknowledge), not overly defensive medicine, not reducing access to medically necessary services, and certainly not raising the cost of health care for all. I would be the last person to say that our current approach to medical liability insurance is perfect. It's not. There are too few claims to provide an adequate safety incentive, not enough patients are compensated, patients get reliable information about what happened to them far too late in the process, and doctors are asked to bear too much of the medical liability insurance burden. But if we want to address these real problems, we should start with the evidence and not with the myth. Just as we need evidence-based medicine, we also need evidence-based medical liability reform. Thank you.
[Applause.]
MR. KLICK: We're now going to hear from a panel of three discussants, and then that will be followed by short comments by me. At that time we will let Tom respond to the discussants, and then we'll open the floor to questions from the audience.
Our first discussant is Ted Frank who is a Resident Fellow here at the American Enterprise Institute. And he is also the Director of the AEI Liability Project. Ted?
TED FRANK: Professor Baker has written a book with a great title. There is certainly a lot of misinformation around, and I think it's important to view the data that's out there with skepticism and debunk the claims that distort the debate.
My problem with the book is that it doesn't quite do that. I can't go through the entire book today, and, as it turns out, the talk I've prepared doesn't quite match up with what Professor Baker has discussed, including what I think are some tremendously punitive measures on doctors and health care providers as part of his reform proposal, but I do want to focus on some areas.
There are good things about the book. There's a good look at the insurance market. Unlike many trial lawyers, Professor Baker isn't out there claiming that this is a conspiracy of insurers to raise the prices, and there are real economic factors behind this. But what he's really saying is that medical doctors were paying too little for many years and now they're paying closer to what the real costs are and that in fact they should be paying even more. If you agree with that, then Professor Baker has a lot to say for you, but if you feel that doctors are faced with paying too much in the way of insurance premiums, you might disagree that some of the things Professor Baker discusses as myths.
I have a problem with the way the book debunks things. Litigation reform claims are attacked with withering skepticism, sometimes justly, sometimes not so much so, while antireform claims are taken at face value or sometimes even stretched to reach conclusions that the data doesn't support. Studies that have mixed conclusions have the anti-reform conclusions trumpeted and the pro-reform conclusions swept under the carpet. I'm thinking here mostly of the Harvard study which we'll get to later.
Reformers are attacked for relying on anecdotes, but what Professor Baker calls the centerpiece of his book, Chapter 5, consists almost entirely of anecdotes, including one that characterizes reform prompted by a newspaper expose as an example of litigation-caused change.
One issue I want to talk about is the issue of type-one versus type-two errors, and it's the question of malpractice versus defensive medical. Measures that are taken to reduce malpractice will necessarily increase the amount of defensive medicine practiced and vise versa. Where are we on that? That's an interesting question. But what surprised me about the book was that it really attacked the type-two error studies, the defensive medicine studies. For example, the section on C-sections notes that these procedures which are often defensive, and C-sections have increased dramatically over time without any concomitant increase in the number of cerebral palsy cases that they're supposedly meant to prevent, but since they vary greatly from hospital to hospital, there must be other factors causing them rather than malpractice concerns.
But then on the other hand, you look at the Harvard study and there you see 40 to 1 ratios in what the Harvard study determines to be malpractice conducted by the doctors at these different hospitals in much greater variance in what individual hospitals do for C section and we're ready to insist that that malpractice litigation at the margin is having the effect of deterring negligence. Why discount one but not the other? And this sort of thing is unfortunately more typical in the book than I would have liked.
The evidence of type-one improvements is anecdotal. There is nothing inherently wrong with that. Anecdotes can be very important and illustrative. But the book is equally eager to explain away defensive medicine evidence, but on the other hand, if there is a reform it's because there is litigation pressure. That doesn't seem to be right to me. We can't be at a sort of inflection point where the type-one errors are being discouraged by the medical malpractice system, but type two errors aren't going to occur by changes in the system.
I will fully acknowledge that as one changes the rules to encourage or discourage litigation, one will get more or less type-one errors and an inverse relation of type-two errors. Purely in terms of the health care system, we might be better off if we moved the arrow in one direction or the other. I don't have evidence where that pointer should be, neither does Professor Baker, and neither does anybody. Nobody has done that study. What we do know is that we have a malpractice problem and we have a malpractice litigation problem, and I think the evidence is there that these two are not anywhere nearly as connected as Professor Baker would argue and it's that disconnect that I think is the real weak point here. There are real second-order costs to litigation that justify reforms by themselves, and the book disregards these for the most part.
We have a wealth transfer of over $10 billion from patients and doctors to attorneys. The medical malpractice system has other costs that aren't counted at all, the time doctors are spending in depositions and court and dealing with attorneys, the change in the world where the plaintiff's bar has developed an effective stable of for-hire experts over the last quarter century. These depositions aren't especially effective in terms of getting to the truth, and I urge you to spend some time with a doctor who has actually dealt with a malpractice suit and they will tell you that a major goal of these depositions—and I will tell you from doing depositions—is to trick the witness into something that can be taken out of context or capture them losing their temper on videotape, or some other measure to encourage settlement or to move the case towards settlement, not so much to discover the truth.
The proposals we're seeing would increase these administrative costs dramatically. We are talking many more lawsuits, many more administrative hearings because of the whole network there to resolve claims under $50- or $80,000, new layers of bureaucracy for adverse event disclosure, and I think you have huge problems there. What is going to be determined an adverse event? Is a doctor going to have to issue three sets of paperwork for every single patient that he or she sees? If I prescribe cholesterol medicine and my cholesterol goes down 10 points, is that an adverse event because it was supposed to go down 20 or 30? And on and on in that vein. And because of the huge punitive risks of failure to disclose, there are going to be tremendous overdisclosures, a tremendous amount of administrative cost getting the disclosure done, and we have seen no consideration of that. So at the same time we're hearing calls for evidence-based reforms, there is no analysis what the additional costs are going to be and whether they're going to justify the benefits that may or may not come from these reforms.
The question is, are courts good decision makers? This is where I want to mention the disconnect between malpractice and malpractice litigation. If courts are good decision makers, then more malpractice litigation will have an effect on malpractice, but I think what we're finding is that the courts are not good decision makers and because of that, you can fix problems with the malpractice litigation system without having an effect on malpractice and you can fix malpractice without having effects on the malpractice litigation system. There are really two separate problems here. I think there is good empirical evidence that the courts are not good decision makers because the way lawyers act shows the revealed preference that they don't believe that courts are good decision makers.
We have of course the Harvard study litigation files, and it is mentioned in passing in the book that the Harvard studies relied in great detail elsewhere to show how much malpractice there is, but there is very little discussion of the fact that when these malpractice cases were followed-up in the litigation system, what happened was that the majority of money that was paid out was paid out to people who didn't suffer malpractice, or at least under the definitions of the Harvard system.
The other way we know that lawyers think that they can manipulate juries is how much in the way of resources is spent on forum shopping. If juries are these neutral, equal decision makers, if all courts are equal in this regard, there would not be the resources spent for malpractice claims in Western Pennsylvania to rush to Philadelphia to try to get their cases dealt with there, there wouldn't be the battles in other cases between state and federal courts. And the fact that lawyers at least think that they can get better juries by going to different places shows that maybe these juries are not the best decision makers out there.
But more importantly, it doesn't take that much of an error rate to really create gigantic distortions in the system. What plaintiff's lawyers are looking for is the big pay day, and they can settle for winning 1 out of 10 cases where the damages are $60 million or otherwise gigantic. So many of the studies out there are looking at medians, or are looking at how most of the juries get it right, and judges agree with the juries most of the time; but the real problem with the system are the outliers, that one jury that goes out there and issues a $250- or $300 million award or the $20 million award, or even a judge as recently happened in Florida issuing a $60 million award for a cerebral palsy case. Insurance companies have to adjust for these amounts. The variance of these issues is extremely important and hopefully Martin or Jon will talk about that.
Another reason we can expect that courts get things wrong is because we know experts get things wrong. In a fascinating study, we see the ASA study mentioned in the book, but we don't see the follow-up study that was, I don't have the cite there, but I think it was the Journal of the American Medical Association, Kaplan and Posner, and they took a set of case studies from the closed claims and for each they varied one fact, whether the injury was temporary or permanent. They gave the doctors one set of identical facts with the only variance being the significance of the injury. What they found was that the professional suffered from an outcome bias. With identical facts, they were twice as likely to find that the anesthesiologist acted negligently when the injury was permanent. That is consistent with other studies that shows that juries suffer from outcome bias, it is consistent with how attorneys talk about cases, the sympathy of the plaintiff, and how attorneys act at trial. You are more likely to find negligence if there is an adverse result, and that is a further sign that the court system is not rational in that regard.
There is another great anecdote in the book that isn't drawn to the full conclusion, and that is the ASA reforms’ 95 percent reduction in deaths. That's a marvelous litigation-promoted reform where the anesthesiologists got together, said what's causing the problems, what can we do to reduce the number of deaths, and they did reduce the deaths by 95 percent and their insurance rates went down 40 percent. To me that shows that there's malpractice out there and there are malpractice litigation problems out there because 50 to 60 percent of their malpractice costs have nothing to do with the malpractice that they are incurring.
Other studies out there, I'm not going to go through this, a great must-read article in the Journal of the American Medical Association by Merenstein over his experience trying to defend evidence-based medicine in court.
Let me just get to the question of incentives. What we hear is, who should pay the biggest share of the medical malpractice burden? What struck me about this book was there seemed to be an attitude that medicine would be great if wasn't for all of these doctors. It's important to look at what's being deterred. We take the case of Dr. Thomas Hurley of Illinois who recently had a malpractice case against him. He performed one thousand of these very risky spinal surgeries that have a definite risk of paralysis and he was very skilled. Only one of the patients was paralyzed, but unfortunately that patient sued and won $2.3 million. The plaintiff was probably hoping for more. There are certainly eight-digit paralysis verdicts out there, and if in fact the plaintiff hadn't already been confined to a wheelchair before he had the surgery, it might very well have been more. So we're talking a cost to the average patient of $2,300 to $2,500 in paralysis insurance once we include the legal costs which are probably greater than $200,000 because of the number of other parties involved in this litigation.
I don't know how many patients can't get the spinal surgery because of that increased cost, but consider that not every doctor is as skilled as Dr. Hurley, maybe some of them paralyze three or four out of one thousand, or that the jury could very well have easily been more generous, and we can quickly see that we're getting to a cost per surgery out there if everybody gets their compensation that is dramatically higher and the only thing it might deter is not so much malpractice, but the surgery itself. Are we really better off if the economic signal to Dr. Hurley is not to perform any spinal surgery? That's a question I think you should ask yourself. If the answer is no, then perhaps we should be reevaluating the concept that patients should be compensated for adverse results rather than adopting a proposal that increases these transfer payments several-fold. And it comes down to that question, do you really feel that doctors aren't paying enough in medical malpractice insurance.
MR. KLICK: Thank you, Ted. Our next discussant is David Hyman who is a Professor of Law at the University of Illinois. David is also a medical doctor as well as a lawyer, so he's on both sides of this problem. David?
DAVID HYMAN: I thought I was on both sides of this problem until I spoke here in the spring and suddenly I found myself not on the side I thought I was.
[Laughter.]
MR. HYMAN: It's a real pleasure to be here as a discussant this time. One of my discussants from last time was Randy Bovbjerg who started off by saying the purpose of a discussant is to first say something nice, then say something nasty, and then close with some useful suggestions. So I'll let you all score how I do on those three criteria.
This is a cover of Regulation relatively recently that I had an article in, and I can't tell you what a huge achievement to get a question mark at the end of “Saintly Doctors and Demonic Lawyers?” I think the conventional wisdom has in most circles picked out who's wearing the white hat and who's wearing the black hat. I like using cartoons. Here are just a couple of cartoons to sort of make the point about the public's view of what's going on in medical malpractice. The caption for those of you in the back says, "Too bad you had a plane accident. Instead of being a malpractice victim, you could have been a rich man," that sort of resonates with the notion that we've got a lawsuit lottery here where the type of defendant makes a huge difference as well as the type of plaintiff. From the lawyer's side, a sort of snide version of father talking to his son, "Let me remind you, frivolous lawsuits put you through college." So rather different views from the medical profession and the legal profession as to what's going on, and I think Tom has done a nice job of capturing the competing perspectives in the book. I like using this quote from Daniel P. Moynihan, one of the virtues of these sorts of things at AEI is, as was put previously by Jon, we're trying to change the evidence to advocacy ratio here, so let's start with what some of the facts are.
As Tom I think pointed out nicely in the book, there are really polar competing perspectives, different conventional wisdoms as to what's going on both regard to medical effort and with regard to medical malpractice litigation. And if you ask the doctors or the insurers, they basically have straightforward explanations, there are more claims and we're spending more dollars per claim, and it's the fault of those damn lawyers, not to mention the juries that say how many zeroes do you want after that when dealing with an injured plaintiff. If you go talk to lawyers, they have an equally persuasive (at least to them) narrative that blames the insurance companies for engaging in collusion, and a bunch of bad doctors who if they just stopped committing malpractice the lawyers would find something else to do. The politicians just pick a side and the side they pick depends on their political party. One party has concluded that it's better to be nice to doctors, and the other has concluded it's better to be nice to lawyers with campaign contributions flowing accordingly.
Then academics; I think Tom has reflected quite accurately that the wisdom among academics who have studied the medical malpractice system, which is that the insurance cycle has a tremendous amount to do with what's going on in the pricing of malpractice premiums, but there's lots of evidence that health care isn't nearly as safe as it should be and there's lots of work to be done on patient safety issues.
Bill Sage, who has studied this area a lot, ran the program that the Pew Charitable Trust had for studying malpractice experience in Pennsylvania and has this quote. Can people read it in the back? What it says is that the real difference here is not between the doctors and the lawyers, it's between the academics and everybody else because everybody else seems to think what we should be fighting about is whether we have caps or not. And the academic position is that's really not going to get you very far. Now a cap of zero would probably get you pretty far, although it wipes out the insurance market, so the insurers are not very keen on that as a solution.
As a representative of the doctors, the lawyers and the academics, I've torn here as to who I should pick sides with. Instead of trying to resolve that for you, let me just focus on what Tom has done here and give you a couple of complications.
The first is that Tom does a nice job of talking about both the closed-claim studies of which there are a number, the medical review studies of which Harvard is the leading study, first of New York and subsequently of Colorado and Utah, and the original study done in California done over three different decades and four different states. But he also cites and gives considerable credit to an observational study done at a single hospital in Chicago where the main publications out of that were in The Lancet and the DePaul Law Journal. My personal bias on this is I think that that study has a lot of merit, but you need to be very cautious in reaching a conclusion as to the rate of negligence from out of that study because of the way in which that study was conducted. You had trained graduate students hanging out in the wards going to rounds and listening and writing down everything. Everybody said something to the effect of, you shouldn't do that. I overstate slightly the looseness of the criteria, but nonetheless, it was a very different study in its design than the medical practice study which looked at comprehensive hospitalization, medical records and a series of staged analyses to try and determine what the likelihood of negligence in a given study was. So I think that that study is a useful complement to the inpatient studies, but I am much more hesitant than Tom is to give it the kind of weight that he does.
Second, on defensive medicine, this has been studied very extensively and the personal accounts of it are very different than what the studies have revealed. Part of the difficulty is that many of the studies were done when malpractice premiums were low, and so those are the circumstances when you would not expect defensive medicine to have the sort of front of mind awareness as if you studied it in the middle of a malpractice crisis. And there's a relatively recent paper that I think didn't make it into the book done by Studdert et al. in JAMA. I think it was July 2005 that looked at the rates of defensive medicine using a survey approach in Pennsylvania, and it asked people something that the previous studies hadn't. The previous studies gave people clinical scenarios and said, what would you do? This study said, what's the last defensive medical thing you did, and provide some detail. It allowed people to more or less frame for themselves the kinds of defensive medicine practice that they were engaged with an open response strategy.
The other thing that I think this study adds is it slices defensive medicine a little bit differently by describing it as first assurance type defensive medicine. That's where you do more things to lower your liability risk and hopefully be clearer about what's going on, versus avoidance behavior where you avoid certain kinds of patients and you avoid certain kinds of procedures because the liability risk is too high there. This is useful both for understanding that defensive medicine is not a single entity, but also for seeing that both of these types of defensive medicine can have benefits for patients but also have costs for patients. It's a mistake to view defensive medicine as net positive or net negative. It's a mix. It wouldn't be a mistake, but there are both positive and negative aspects of defensive medicine that need to be attended to. That said, the study doesn't attempt to quantify the total dollar cost, and I think this is something we're not going to be able to come with a number on anytime soon.
Third is I think Tom is extremely economically oriented, something he would not necessarily view as a compliment, in how he deals with the framing of this malpractice crisis which has been primarily access driven. In Illinois you've heard a lot that there are no neurosurgeons south of Springfield and family physicians are not delivering babies anymore. Tom's response is essentially, who says you need to have a neurosurgeon south of Springfield? And if there ought to be one, maybe we ought to deal with that directly rather than tinkering at the margins with malpractice. He also has an interesting discussion about delivery patterns and whether you actually want family physicians delivering babies, and whether OB-GYNs' complaints are part of what's going on in the background. So I think he has a sensible response in economic circles, but it's probably not one that the political sphere is going to be happy about.
The other issue is obviously if you've got a tenuous practice situation and your malpractice premiums go up and you can't raise your rates, that may be the straw that breaks the camel's back that has the access consequences that are commonly discussed here. One of the difficult questions you have to confront is, if the choice is between a bad doctor and no doctor, is it so clear that the right answer is no doctor, which is the way that the malpractice feedback loop seems to be ending up operating.
Tom talks a little bit, and I think he has a single paragraph that talks about loss prevention. This is something people expect insurers to do in well-developed markets. My sense is there isn't that much of it going on, and it's actually very hard to see it going on while we're cycling between hard markets and soft markets because in a soft market nobody is much interested in your loss-prevention efforts, it's really primarily price driven. Now, it's possible that loss prevention may result in lower prices over time, but this is a year-by-year decision. And in a hard market it's price driven as well. You're going to spend your time about your actuarial risk more than lowering the risks in the loss.
I think the other factor to keep in mind, although Tom doesn't try and fix the insurance cycle directly, it turns out to be very hard to fix, particularly given the kind of regulatory framework we have, because the problem is essentially under-pricing in the troughs and overpricing in the peaks, and to fix that it's easy to order the insurance companies not to charge so much in the peaks, it's very hard politically for an insurance commissioner to order an insurer to charge more than it wants to in the troughs, and you need to do both of those, you can't just fix one of them. So that turns out to be very hard to solve, a different set of complications.
I think it's here where I come to the real areas where I part ways with Tom on his solutions. He's got this Patient Protection and Health Care Responsibility Act, and it's pronounced PPHRA. PPHRA. Great acronym. I think PPHRA has some benefits. It's a useful response to what I would call the chutzpa defense. For those of you who are not fluent in Yiddish, the technical definition of chutzpa is someone who kills both of their parents and then throws themselves on the mercy of the court because they're an orphan. The chutzpa defense in medical malpractice is health care providers complaining that the malpractice system doesn't compensate patients enough or as quickly, when the solution to that is in their own hands. They could come up with ways of writing more checks to more people more quickly. And when you look at the research which indicates about 40 percent of the cases that go to trial are zero offers, that is, the defendants says we won't pay you a cent and we'll make you prove up everything, some efforts at switching the incentives a little bit seem to me to be potentially useful in dealing with that.
The second point is that Tom in talking about verdicts makes two observations, the first, that punitives are nonexistent in medical malpractice, and the second is he ties his fee-shifting provisions to doing 20 percent better on your verdict than the offer. In some research we're doing right now using Texas closed-claims data, we actually find punitive damages in 6 percent of our jury verdict cases. That's not to say they get paid, but we do see them, and the not-get-paid point I think is important. One of the things we see in the same study is verdicts, you can award as much as you want, it doesn't mean that you get paid. That doesn't necessarily apply outside of medical malpractice, but we're talking about physicians who don't have lots of assets typically, hospitals have more, but unless you want a hospital, you don't want most of their assets. And so what you end up seeing is the cases end up being settled often after verdict for the limits of insurance coverage regardless of what the actual verdict is. So I think that complicates matters a little bit.
Tom likes public health departments, and I'm sure the Connecticut Public Health Department is extraordinarily fine. It's not so clear that the other 49 public health departments are the best place to do this or that they have the skill, interest or budget to do this. And forcing a reporting system when the people at the other end just say, that's very nice, and then shove them in the closet, is probably not what Tom has in mind, so I'm a little concerned about that strategy. I wanted to mention on Tom's verdict versus the larger framework of this. This is somewhat similar to a proposal Jeffery O'Connell has made in a number of forums including at this very table.
The final thing I want to focus on is enterprise insurance with the hospital as the organizing entity. This is also not a completely new idea. Sage and co-authors had a paper on this in 1994. Tom actually cites a 1997 paper by a similar set of authors. But there was an ongoing debate at that time whether hospitals were the best organizing entity for what was then enterprise liability as opposed to enterprise insurance. Abraham and Weiler were on the side of arguing for hospitals, Sage and co-authors thought insurers made a much better organizing entity. My next point is that given the unbundling of inpatient services and the changes in where health care is delivered, I think there are good reasons to be very concerned about using hospitals as the basis for organizing this system even if you think enterprise insurance is a good idea. I have some concerns about whether enterprise insurance relatively quickly becomes enterprise liability. So let me show you a couple of data slides that will make the point of why I think hospitals are a problematic basis for the proposal.
This is Medicare hospital utilization from 1985 on in terms of days of care per thousand enrollees. What this shows is a pretty steady and consistent decline from 3,000 days per 1,000 enrollees per year, to 1,860. It's down a very significant amount over this time period. Not surprisingly, spending tracks number of days in the hospital, although not perfectly. Medicare used to spend 66 percent of its budget on inpatient care, it's now down around 40 percent. The population in general started at 38, peaked a little bit and is now down at 31 percent. So dollars follow days, and both of them are reflective of the fact that a lot of health care has moved out of the hospital. And if it's moved out of the hospital, the question is why would you set up your system to tie coverage and physicians to the hospital when it's seeming to assume a less and less significant part of the system.
This just looks at four different types of health care providers, one inpatient which is hospitals, the red line that sort of starts at the top on the left. I apologize because I thought this would be easier to see. But we had about 6,800 hospitals in 1980, we're down to about 6,000 now and that's a pretty significant decline. The other three are home health agencies, outpatient physical therapy, and ambulatory surgery. Business is booming for all three of those outpatient providers of services consistent with more dollars flowing to them. The one exception is home health agencies which was doing extremely well in 1998 and has dropped significantly since then for reasons I think most people in the room know, has everything to do with fraud and abuse and government reimbursement patterns and not changes in demand.
Let me just go back. I think there are reasons for thinking the unbundling of inpatient services will complicate Tom's proposal, a couple of different complications. It's not just general hospitals, but we see specialty hospitals now, further unbundling of the general community hospital, and there is some reasonable evidence, I used to wear the Federal Trade Commission hat, of alleged anticompetitive conduct by local hospitals that had attained market power and the question is whether you want to use Tom's liability insurance proposals to solidify that power, because once people are tied in to the local hospital, the physicians as well as the inpatient services, Katy bar the door when it comes to pricing.
The last point is the old Yogi Berra line: if people don't want to go to the ballpark, nobody is going to stop them. There is nothing that keeps individual hospitals from trying to adopt this kind of strategy, and they haven't, so revealed preferences are telling you something right there.
Let me just close with my suggestions on getting there from here. I agree with Tom that there are lots of things we could do to improve the quality of health services in this country, and I think medical malpractice has an important role to play in doing that, as I've written myself, so a couple of translational issues. The first is politics and allies matter, and just before he describes his proposal Tom says I don't expect this to come about any time soon or in the form that I'm describing. Maybe you ought to change your proposal just a bit in order to get it more likely to come about soon and in a form that you would like, and to try and look for some allies to help make that happen. Otherwise, it's just us propeller heads in the academy who are proposing these things and people on the Hill are ignoring us.
The second thing is acronyms matter. PPHRA is not a good acronym.
[Laughter.]
MR. HYMAN: Maybe he was trying to emulate HIPAA which actually got passed into law, but I think you need to come up with something that's got a little more zing to it.
Then the last point I want to make is if you're thinking about dealing with health care quality, you need to think about complementary strategies to just tweaking malpractice, and my favorite is always information-forcing strategies. If every hospital had to have a sign like this outside it, you wouldn't need enterprise insurance or enterprise liability. First of all, the number would be 1 instead of 172 at least for a while, but in relatively short order people would get the message.
Thank you very much.
[Applause.]
MR. KLICK: Thank you, David. I'd like to correct myself. As David said, he was at the FTC, so that means that he actually hits the grand slam in terms of groups, he's a doctor, lawyer, academic and politician.
Our next discussant will be Martin Grace who is the James S. Kemper Professor of Risk Management at Georgia State University.
MR. GRACE: Thank you, Jon. I have the privilege I guess of being a lawyer, an economist and an academic, so we're hitting all the bases today.
I want to thank everyone for bringing me here today, but I'm going to be talking about insurance. Is everyone asleep yet?
[Laughter.]
MR. GRACE: That generally puts everyone asleep. I'm going to repeat some of the things that Ted has said and some of the things that David has said, but I hope to give a couple of other things to think about.
What I liked about the book: So I'm going to start off with a nice, pleasant set of facts. I thought this book had a nice set of evidence in there, and maybe you can disagree with the evidence, but it's one of the first times you see someone make an effort to bring it all together and put it in one book. As I said, you can disagree with some of the evidence, but it's all there. And we're looking at empirical studies, not necessarily anecdotal studies, and that's what some people claim are studies where you get a couple of anecdotes and then you have data. But this goes beyond that.
My favorite part is that this book actually describes the insurance market pretty well. Most attempts at describing the medical malpractice insurance market are either biased or uninformed, and this one I think does a really good job of getting at exactly what the problems are from the insurance company's perspective. So what I'm going to talk about is some of the things I might have quibbles with. I'm not sure how strong my critique is, but they're quibbles with how things are framed in the book going forward.
One of the first themes that Professor Baker talks about is that we have an epidemic of malpractice, not a lawsuit epidemic. An economist would say that you like to have lawsuits because these lawsuits internalize the mistakes that a hospital or a physician might make, the cost of these mistakes, and therefore they will make better decisions in the future and safety will increase and everybody will be happy. But we don't really see that necessarily, that more lawsuits will actually make this occur than malpractice insurance.
We see that lots of people don't sue. We saw a discussion of evidence from the Harvard study and other studies that no one sues, but the problem is, there may be some rational reason why people don't sue. I'm going to talk about that in just a second.
What else is interesting is that physicians or defendants in medical malpractice cases seem to win a heck of a lot of the cases when they go to court. So this makes medical malpractice a lot different from other types of liability insurance or liability markets or liability litigation.
One of the things, I guess it maybe was a throw-away in the book, was that Professor Baker said everyone who has an automobile accident makes a claim, but not everyone who has a medical malpractice accident makes a claim. He was talking about that and it made me think. What I was thinking about is that these things aren't really comparable. Who's to say that the medical malpractice claim rates should be the same as the rates for other types of injuries in other types of negligence cases? Auto insurance is given as an example, and this is going to crack me up when I say this, but auto insurance is a well-oiled ex post compensation system. That sounds kind of high-level academic, but compared to medical malpractice, the automobile compensation system is heaven. So we have to think about these are really two different types of compensation systems.
Just to give you an idea of some of the costs, and you can go to A.M. Best and get zillions of different ratios, and I just happened to pick two, and there are more. But what's interesting about the ratios is they're always going to look something like this. What I did was I looked at the direct adjustment services expenses. That means these are the amounts of money spent directly on a particular claim in medical malpractice or automobile insurance. We're not talking about overhead allocated or anything like that. The second number there, the total loss adjustment expenses incurred, that has the allocated overhead, everything else. So for medical malpractice, just trying to adjust and figure out what the claim is worth and going through all of the processes leading up to the claim being paid, that's 84 percent of the cost of resolving this dispute, where for automobile it's 25 percent.
Another way of looking at it, and I'm only going to give you two because if I had all day I would give you the rest of them and you would for sure be asleep. If we look at just the expenses that go to adjusting these claims to premiums, for medical malpractice it's 34 percent, and for automobile insurance it's 14 percent. So if you are a rational actor and you think about bringing a medical malpractice suit and you have full information which I'm suggesting is a lot of information to have, people rationally don't bring suits because it's too expensive, and that's quite possible. That doesn't mean that these people shouldn't be compensated, but our system that we have set up for the compensation of medical malpractice injuries is not an efficient system. It's not clear that it deters, doesn't increase patient safety, and it's not clear that the people who are truly injured are getting the money. So that's one thing that I think can be talked about some more.
The thing that Professor Baker brings out that is sort of a theme in his book is that we're having an insurance crisis of a kind, not a tort crisis, and I'm paraphrasing what he said slightly differently. But I think that the insurance market, the way it works is actually harder and more difficult to get a grasp on than Professor Baker describes, and it has to do with a lot of idiosyncratic things that have happened over the last 20 or 30 years in medical malpractice. We don't have large carriers selling malpractice insurance, and if we do they are the exception. They tend to be small, physician-owned mutuals, which means they are nonprofit, or they are captives, that means they're wholly owned subsidiaries of a hospital or some other kind of provider, and a captive is just generally a savings account. It's not really insurance in the classic sense, but you get to save over time for your medical malpractice liability claims. And risk-retention groups are very similar in that regard, but instead of being a subsidiary, groups of people get together and form an RRG like all doctors in a particular county or in a particular state, maybe.
These markets also may have 50 percent of their coverage supplied by alternative means. That means they may be coming from outside the country, from these captives, rather than traditional insurance companies. So we've eviscerated the insurance market, and what we're left with are a lot of alternative mechanisms that tend to be small, less sophisticated in many respects, than a large company like we're used to dealing with for our automobile or homeowner's claims.
What makes this worse is that because of the cumulative malpractice climate in the last 30 years, we're left with single-state companies writing a single line of business. They have little access to capital markets. That means they can't go to Wall Street and get an infusion of capital, they have to go to reinsurance markets or banks to get money, and the bank only wants to lend you money when you don't need it, right? Well, the reinsurer only wants to give you money when you don't need it, too. These companies have more difficulty in underwriting. There aren't 50,000 medical malpractice accidents like there are rear-end collisions that happen every year. It's hard to price these things. Pricing is not an exact science like it is for a homeowner's claim or it is for an automobile claim. That makes estimating the frequency and severity of these types of injuries very difficult. So insurers, in addition to being subject to this cycle which I'm going to talk about in a minute, also have a difficult time to begin with. It's just not the type of rocket science that we're used to dealing with for other types of liability.
Here is my discussion of the cycle. I was telling Professor Baker at lunch that you can't be a real insurance professor until you write a paper about the cycle, and now he's a member of the club. So congratulations.
We normally think of cycles, and I found on the Internet that looks really nice. You have these ups and downs over time. They're smooth, there are sine waves, and they're predictable. That does not describe the insurance cycle at all. In fact, I don't even use the word cycle in my seminars anymore. I call it a thingy.
[Laughter.]
MR. GRACE: It's not really a cycle, it is a set of events that happen that are resolved and they might happen again, but we can't tell at any particular time when they're going to happen. In fact, there were articles in the professional press, the insurance press during the 1990s, where is the cycle? We're expecting it. It's time. Where is it? And it didn't come. So it's not really the way we think it is.
A proper characterization is that we might have these random shocks and a return to some sort of equilibrium. But what if you have a random shock before you get back to an equilibrium, so it looks kind of disrupted? So this is not something that's predictable. It can't be managed by firms, and it's really important, it can't be managed by regulators. Think of the Georgia Insurance Commissioner and think of the power he has. Can he stop the market? No, he can't. Nor can the Illinois or the Connecticut Insurance Commissioners. And even acting in concert they can't do this. So firms acting in concert would be illegal, and they can't do it either even if they could.
What Professor Baker does state and I think is kind of interesting and actually accurate, and it's important, is that the increased losses do not cause the crisis. However, the increased losses concurrent with an external shock, that's what the crisis is caused by, and I have a chart in just a second that hopefully will illustrate it somewhat.
This is just a quick shot of the cycle and all lines from 1967 to 2004. We see a long-term up and down, but each of those little ups and downs at one time people were calling the cycle, and they're just not repeatable in any regular sense of the word so I tend to call it something different.
This is a little bit harder to see. This is the loss ratio over time for all lines coverage, malpractice and reinsurance. In the reinsurance one I just wanted to point out some things. That high point in 2001 is 9/11, and there's a smaller high point in 1992 which is Hurricane Andrew. That has a tremendous effect on the price of getting high-capacity for everyone, not just medical malpractice insurers, but for everyone. But med-mal insurers are in that same market. They're trying to buy the same capital from the same companies, so they have to compete with everybody else in these shortage times for reinsurance protection. If you happen to have high losses and a history of slightly under-pricing your losses, you're going to be hit with a small or a big shock when it comes to repricing next year.
If you go back further I have a slide that I didn't bring, but it shows the oil shock having the same kind effect in 1970s, and we see that's when we had our first medical malpractice crisis. So it's losses plus it's this external market that has an effect on pricing over time.
I'd like to talk about the problems of pricing again. Professor Baker actually does what I think is a very good job describing the insurance accounting. I think I'm going to actually have my students read it because it's easier to read and understand than some of the more technical books, and it's just as accurate. But the price of medical malpractice insurance depends on losses, it depends on investment return, and it depends on the risk that the company is taking. A company could say we're only going to take a certain kind of risk, we're going to take the pediatrician. That's it. They're not going to very many doctors, they're not going to have a big enough pool of doctors. Or they'll say we'll take the next category, or the next category. But even still, even if you take all the doctors in the state, you don't really have big enough risk pools to price easily. That's why we don't have what's typical in automobile insurance, if you get a ticket or you have an accident, your prices go up. We don't have that experience rating because there's not enough good links between a medical malpractice lawsuit with a verdict or a judgment against a doctor and the fact that he or she is a higher or lower risk. It's not necessarily there.
But the risk I want to point out is actually also not just the doctor, but it's an uncertainty about future losses. If you have relatively high investment income, that's going to create a certain pricing scenario. If you have investment income that drops precipitously like it did after 9/11, actually it was before that, but you're going to have a different kind of pricing reaction. And if you have losses that change significantly, you're going to have a different pricing reaction.
But if you think that the future is going to be even worse, you're going to want to protect yourself against that, and that's what I think right now when we have declining investment income, we have potentially higher, I say this anecdotally, verdicts. Companies who are small and don't have the capital base are going to try and raise and prices to try and remove that risk of insolvency, so I'll talk about that once more.
If you can understand this chart, you are an actuary. I want you to be able to go home and say, yes, I now understand actuarial science. The blue curve, if you will, that could be what actuaries think is the distribution of medical malpractice losses. But if we start seeing a whole bunch to the right of the black line, it could be on the blue one or it could be on the red one. Did I say it backwards? I did say it backwards. The red one is the current one we're thinking about, so I hope I don't lose my actuary license now that I messed this up.
The idea here is that if you start having hits—verdicts—above where that black line is, we don't know if we're on the blue distribution or on the red distribution. So on the safe side, we'll take the more conservative approach, and if we start to see large verdicts, more of them, we're going to assume that the distribution of losses has shifted. This is what the actuary on the ground, the guy that's pricing these products in Kansas, Texas or Illinois, is having to figure out. It's not just the lawsuits that happen in your own state, it's big lawsuits that happen in other places because like a contagion effect, the losses from next door can actually affect your premiums and losses in the state. So we want to think about that, too.
As I mentioned before with the increased litigation rates, that doesn't seem to be an issue right now. Increased payouts may or may not be an issue. But the supposed increased jury verdicts, and these are the anecdotes that the actuaries have to actually listen to as real data and put them into their pricing which also causes insurers to reserve more conservatively. So the largest verdicts seem to be increasing. We hear about things increasing all the time, and this may be a myth because these things actually aren't paid, but to an insurer who has to actually pay if a claim is actually due, they'll reserve in this manner.
One thing that's interesting to think about is that if you look at the purple line, the purple line is the loss ratio and that's the losses that a company incurs divided by the premium. As that gets lower and lower, the company tends to make more profits. If the losses relative to the premiums are lower, then the company is doing better. We see the losses to relative to premiums increasing in 2001, and as companies increase premiums since that time, this loss ratio had gone down. I think in 2004, maybe the first half of 2005, medical malpractice insurance has broken even, including their investment income, for the first time in it seems like centuries or something like that. That's good except for the fact that the variance or the standard deviation of these loss ratios have gone up. Companies are more uncertain about the future. In fact, from 2001 to 2002, it didn't go up very much, but from 2003 to 2004 it's gone up even more. This is a measurement of the uncertainty in the market there and it's a hangover, if you will, and we were talking about this at lunch, why do companies still over-reserve or increase their reserves during this time and I think it's because they're afraid about the future.
Finally, I'd like to talk briefly about the enterprise insurance proposal that Professor Baker brings up, but I want to make people think about this. Society wants medical malpractice lawsuits to act as both a deterrent and a compensation system. It doesn't do either one of those very well, we've heard evidence and the book has plenty of evidence, and this is the reason we have the big push for tort reform in many respects. We see a lot of evidence that Professor Baker puts in the book and he says it's not the tort system and it's not the insurance industry, but it may be the insurance market. So there's something about how insurance is sold to physicians that is important here.
He proposes some potential solutions. The disclosure one, and I'm not going to really talk about it because Ted has talked about it, but the enterprise liability, the enterprise insurance aspect I think is the most intriguing, but I think it's also troublesome, and as an economist it's kind of bothersome in some ways, and I'll mention that also.
Because some physicians are practicing in high-risk fields, this may make hospitals pay for medical malpractice insurance, is the argument that Professor Baker is making. This means the physician, according to him, is not the best person to bear this residual risk. Just as an aside, if you ask me, I'm not the best person to bear any of my risks. But getting back to this, there is some fairness and equity in that because society demands physicians to be placed in their neighborhoods and if the physicians don't want to be there, you have to do something for them and that's the argument for tort reform. Professor Baker's approach is to say we won't charge the doctors any premium, we'll make the hospital pay for this. I think David mentioned this too, but this will cause changes in hospital behavior. I'm not sure how far you can push this, but hospitals that have a portfolio of practices will maybe take out the high-risk practice, so we'll have the same problem, but now it will be an employment problem, there will be no places for the high-risk doctors to practice because the hospital is now acting as the insurance market and it is excluding high-risk physicians from that market. That's something we need to think about, that if we're going to push something this we have to think very carefully if we want to make the problem one layer higher.
In sum, I want to reiterate that the book really describes insurance markets quite well. My only quibble is expanding insurance to the enterprise rather than the individual physician. Thank you.
[Applause.]
MR. KLICK: Thank you, Marty. I'm going to make fairly quick points. Then we'll give Tom just a couple of minutes to address broadly our comments, and then we'll have a couple questions from the audience.
My two points are, one, Tom puts an awful lot of faith or credence in the studies that purport to show, including Professor Hyman's study, there is not a very good correlation between losses and insurance rates. When we had an event here last spring on David's paper, we talked about a lot of technical reasons why those studies might not be very powerful. Quickly, as Marty said, if I'm an insurer pricing insurance policies, I'm not just going to look at liability experience in my state, I'm going to take all the evidence including evidence from other states, and how you include liability exposure and liability data from other states in terms of with in-state pricing, how you can control for that econometrically is not exactly simple and no one has actually figured out how to do that, or no one has actually done that yet in any of the existing studies, but it presumably is a very important issue to look at.
Also as Marty pointed out, we see changes in variance of the losses that insurers are experiencing, and none of the existing studies include that. So I think in general, Professor Baker puts maybe some unwarranted reliance on these studies, and if you're interested in the more technical points, check the AEI website which has a video of the event that we had on Professor Hyman's paper.
Perhaps the more important of my comments because it has to do with my own work, in Chapter 7 Professor Baker talks about the studies including my own that look at the effect of tort reforms on where doctors choose to practice. He says these studies are kind of interesting, but at best the results are kind of small. My study, for example, shows that when states pass caps on non-economic damages, their number of per capita physicians goes up by about 3 percent, and Professor Baker says that's a relatively small effect. I don't know whether or not it's small, but I know in some new research that I've done when I focus not on just total number of per capita doctors but on per capita high-risk specialist doctors, looking at OBs particularly, looking at neurosurgeons, looking at radiologists, the kind of guys who get sued the most and then using the kind of guys who don't get used at all as a control group, it turns out that the marginal effect of tort reform is much bigger than I had previously estimated. It turns out that you actually see about a 10-percent increase in per capita doctors practicing in these high-risk specialties when tort reform is passed.
Interestingly enough, most of the folks who have looked at this area find that these increases come in those communities that are most underserved. So you're seeing increases in the number of docs practicing in rural areas, places where people in general don't have good access to health care, and so the marginal effects on health outcomes due to increased access to care could possibly be quite substantial. Professor Baker writes this off and says if anything, the tort exposure, liability exposure is the straw that broke the camel's back, that it's all these other things, that people don't want to live in rural areas and what have you. Guess what? We can't exactly turn the panhandle of Florida where I live into the Upper West Side of Manhattan. That's just not a policy option in terms of attracting new doctors to the area, but tort reform may be an available policy option.
It turns out that if you look at my results or some results estimated by Dan Kessler, an economist at Stanford that published this year in JAMA, they suggest that the improvement in terms of number of high-risk specialists located in underserved areas when tort reform was passed is the equivalent of anywhere between a 10-percent and a 40-percent increase in wages. So for Professor Baker to just say if anything, it's a small effect and tort reform is probably a pretty inefficient way to get there, I don't think there's any basis for that welfare comparison. We really don't know. It may well be that tort reform is pretty much a bargain as compared to a 40-percent increase in wages.
The question that might be raised is why don't we know? Why don't we know what's the better choice or whether this improvement in access really leads to improvements in welfare that justify the costs of changing the reform system? The answer is that relatively few of us have actually looked at the effect of tort exposure and tort reform on some kind of health outcome. It's only fairly recently that any of us have started to look at the effect of tort reform on doctor location, and there are only one or two papers including my own that looks at the effect of tort reform specifically on some kind of health outcome measures. I find that some kinds of tort reforms improve significantly infant mortality, and again I find that there is all this concentrated in fairly underserved communities, particularly in the black community, and a handful of other folks have found some similar results with different kinds of methods.
Again, this evidence isn't either numerous or particularly powerful at this stage, but it's the question that we need to answer before we start making the kind of welfare comparisons that Professor Baker is offering.
With that, I will turn it over to Professor Baker. I will give him at most 10 minutes to give some sort of broad responses, and then we will go to the audience.
MR. BAKER: I'm going to talk about phoenixes, hydras and hammers. I used to think that the tort reform movement was a phoenix, you blow it up and it comes back. Now I've decided that it's a Hydra, you chop it off and it sprouts 50 new heads. In this book I have tried to nail down in a common-sense and I think fairly disciplined way all of the evidence that there is out there. Some of it is not as much as we would like, some of it is not as unequivocal as we would like, but you'll have to read for yourself and decide whether I'm balanced. So tort reform, not phoenix, but hydra.
The second observation, and then I'll have some sort of specific response, is hammers. Why do I talk about a hammer? For me, the classic tort reform remedies being proposed are hammers. What do you do if all you have is a hammer and you have a plumbing problem? What you do is you beat on the pipe with the hammer. A nice example would be Jonathan's last example. The thing about tort reform as a solution for high-risk specialists is it benefits not just the guy on the panhandle, it benefits the guy with the high-end practice in Miami, Dade County and Boca. If you want to do something about doctors who are having problems in the panhandle, do something that's targeted to them, not that is indiscriminately affecting all the doctors.
Now some more targeted responses. Ted doesn't like my evidence. You're just going to have to read and see what you think. I'll do one observation. He cites the 1996 NEJM article. I've got an article in a peer-reviewed journal that was put together as part of an edited collection by a person who's from that team which blisters that study as to why you can't cite the Harvard medical practice study anymore for the proposition that malpractice claims are not valid. It stands alone against all the closed-claim studies as showing a lack of a link between the merits and payments, and you can read my article and see if I'm right.
With regard to the anesthesiologists. I'm actually not surprised by the fact that there's a 95-percent reduction in deaths and only a 40-percent decrease in malpractice costs over a 20- or 30-year period. That's because what we can do to take care of people who are really profoundly injured has dramatically changed. In other words, the cost of taking care of very injured people is up. It's not that all these people who didn't die are complaining about the fact that they didn't die, what it is, if they didn't die, they have to live with a lot of expensive treatment and that's expensive so that killing someone is actually a cheap thing to do rather than permanently disabling them. That's one of the reasons why tort costs are so expensive and we just can't forget that.
With regard to David's observations about the PPHRA, I'll plead guilty to it not being a good acronym.
[Laughter.]
MR. BAKER: At a blog, actually one guy suggested it should be called PHARMA-C which is the Patient Health Care and Responsibility Medical Act of Congress and that that would be better. Probably it is. It's easier to spell. The criticisms of the particular reforms, I say in the book I'm not particularly invested in these particular reforms. They're evidence based which is more than I can say for traditional tort reform. The problem of the unbundling of hospital services is a problem that's bigger than medical malpractice, and it's a real problem, and maybe the hospital isn't the perfect risk bearer. I think it's better than the individual obstetrician or neurosurgeon. Bill Sage still says that the health insurer is the right place to place that risk, and in an era of managed care where the health insurer was directing care more, I might have been more sympathetic to that perspective. Again, most of David's criticisms about the particular proposals I would say are well taken and we need to keep plugging at thinking about what we can do to fix pipes rather than bang them with hammers.
Professor Grace's comments on the medical malpractice market, I mostly agree with them and I didn't take them as being fundamental criticisms of the paper. I don't think that auto insurance is the same as medical malpractice insurance. I use automobile insurance as a comparison situation. It's not surprising to me that case-specific loss adjustment expenses are a larger percentage of medical malpractice claims than auto claims because medical malpractice claims are big and in part that's because of prior efforts to cripple the tort system. This is getting close to what I think David calls the chutzpa defense which is that because of all these earlier tort reforms, you made it to difficult and so complicated to get it recovered through the medical malpractice system, therefore, it doesn't work. Automobile insurance is not the same as medical insurance. I think I'm going to quote Martin at some point that he won't like about auto insurance being a well-oiled system. Those are my observations.
The goal of writing the book is to make the debate be about the evidence and it's to change the debate to expand the policy options. Whether these are the particular four correct, perfect policy options, I don't make that assertion. But I think that they're more likely to be successful at addressing the problem than hitting pipes with hammers.
MR. KLICK: Thank you, Tom. We're now going to turn it over to questions from you. Please wait until the microphone gets to you, and before you ask your question, state your name and your affiliation.
MR. VAIL: John Vail from CCL. This is an observation and a question for anybody on the panel who cares to comment. Tom alluded to the fact we can measure claim frequency and Tom and Martin talked about that, and we can measure claim severity as a gross number, and Tom said that claim severity tracked medical cost inflation. The question I have is, what do you see, as researchers, as the quality of data available to you regarding the components of medical malpractice verdicts? What pieces of them go for medical costs, what pieces for other economic injury, what pieces for non-economic injury? And then if that's important, how could you get better data? The third piece of it, again with an observation, take severely injured baby claim. Maybe you'd have a $20 million verdict. My guess is that if you had that, about $15 million of the verdict would be for future medical costs. When you're looking at the effects of these transfers on the medical industry, how do you account for the fact that you have at one point this $20 million cost to the industry, so you should have $15 million flowing back into it in those future medical costs?
MR. BAKER: I'm going to respond to the second part and buck the first part to David because the study is working on has probably the best answer to that question.
With regard to the second part, I think that we really can't have it both ways. In other words, we can't say that medical malpractice costs are a small part of the health care system, which they are, and then think that by plugging back in the medical cost to the health care system that somehow medical malpractice doesn't really cost that much. I think the bottom line is that medical malpractice, the research shows, certainly the direct insurance costs don't add much to this system. And we can debate for a long time about defensive medicine. I will say this, I feel like people have been fair about making clear that when you talk about defensive medicine you've got to talk about the benefits and the negatives and that one of the difficulties in the research is separating the beneficial effects of defensive medicine. After all, the whole idea of tort liability is to change behavior from the wasteful aspects, and that's difficult to do.
But as regards to what's the best evidence on separating out types of damages, David can talk about that.
MR. HYMAN: Yes, I was going to defer that to Tom.
[Laughter.]
MR. HYMAN: There are a couple of complications with that. The first is the sources of data, the second is the verifiability of the data, really in that order. You can either get closed-claim information from individual insurance companies or you can get reported information from state insurance commissioners. There are two public databases, Florida and Texas. There are also other databases where the underlying data is not publicly available, but they prepare reports whose quality varies. There are probably half a dozen states that have closed-claims databases that don't release information.
The second complication that I suspect most people in the room know is the overwhelming majority, we're talking 95 to 98 percent of claims which admittedly there's a technical complication with what we call a claim, it's not the same thing as a lawsuit, but the overwhelming majority of claims are closed without trial. The consequence of that is the source of information about the allocation between economic and non-economic damages and medical versus other economic damages then is typically the result of the insurer that has its own interests on whatever allocation gets made. It may pay a lot of attention to that, it may pay very little attention and only care about the gross outcome, that is, how much they ultimately have to pay.
So if you're asking where do we have solid information about the ratios of economic to non-economic that's more general than a given case, you have to look at cases that are the result of a jury or bench trial, which in Texas is 2-1/2 percent of the total claims, we're talking 24 jury cases per year over the last 16 years, so it's not a huge data set. We're talking a grand total of about 375 cases. I should probably pull out my laptop and double-check, but my recollection is the distribution is about 45 percent economic, 55 percent non-economic, comparing only those two. There is also exemplary or punitive damages which as I said previously you see in some of the cases. And there are both pre-judgment and post-judgment interests that need to factored into that as well. And you have every conceivable variation. You have cases with staggeringly high non-economics and no economics, and the reverse situation as well.
What we don't have in the Texas database and I don't believe is available in the Florida database, although a team from Duke is primarily handling that with Neil Vidmar at the head, is a breakdown between medical economic damages and others. I'm not aware of that information. I think if you wanted it you'd have to get it from individual insurers.
One last brief poi
