pageok
pageok
pageok
Damage Caps and Medical Malpractice Litigation: II

Yesterday, I reviewed some background on damages caps. Today, I want to focus on the effect of damages caps on payouts. Of course, payouts are not the only thing worth studying, as the comments to my first posting reflect. Access to medical services might well be affected by a cap. So might malpractice premiums -- at least that was the hope/expectation of those who proposed caps to deal with the malpractice (premium) crisis. And, don't forget defensive medicine, which affects total spending on health care. But, for right now, I want to focus on actual payouts.

What does past research find about the effect of damages caps? The results are mixed, but most of the studies find that caps do reduce payouts -- typically by between 15% and 35% -- although some find no effect whatsoever. You can find a review of the literature, with references to the underlying studies here, in Section II.

There are at last four different ways of studying how damages caps affect payouts. One can:

1. Compare payouts in states with and without damages caps. This can be done either at a case-level, or using aggregate insurer payouts.

2. Obtain case-level verdicts and estimate how a particular damages cap will affect payouts.

3. Obtain payouts in tried and settled cases from a state without a cap, and estimate how a damages cap will affect payouts.

4. Compare payouts before and after a cap is adopted in a single state or across multiple states.

Each of these approaches has been used by researchers, and each has their own mix of advantages and disadvantages.

For example, the first approach implicitly assumes that all states with caps have the same cap. This is clearly incorrect, as yesterday's posting made clear. So, if a comparison finds no difference in payouts between cap and non-cap states, it might be because caps have no effect on payouts, or it might be because a very restrictive cap in one state had a big effect, but less restrictive caps in other states had little/no effect -- and averaging results across all states with caps obscures the fact that there was a difference in payouts, but only in the state(s) with more restrictive caps.

The second approach looks more straightforward, but it has its own complexities. For example, if payouts don't correspond to verdicts, applying the cap to the verdict gives you a misleading impression of the real impact of the cap. For example, if it turns out that defendants don't actually pay what the jury awards, then a straightforward application of the cap will substantially overstate the cap's impact -- giving the cap credit for "taking away" money that isn't being paid to begin with.

In an earlier article, we found that defendants generally don't pay what juries award -- and the larger the verdict the larger and more likely the "haircut." Overall, only 46% of the amount awarded by juries is actually paid. The most important factor explaining verdict haircuts is the amount of insurance coverage. If the doctor has $500,000 in policy limits, it doesn't seem to matter whether the jury awards $500,000, $1,000,000, or $5,000,000. The insurer will pay $500,000, and that will usually be the end of the dispute. Above-limits payouts are uncommon, and when they occur, they are virtually always paid by the insurer. (More discussion of those subjects is saved for another day).

The third approach, which is the one we use in this paper, has the virtue of relying on actual payouts (instead of verdicts), but one needs to make a series of assumptions in order to do the estimation. The main weakness of this approach (apart from the plausibility of the necessary assumptions) is that it is a static snapshot: it takes cases to which the cap doesn't apply, and assumes the same cases will be brought post-cap. That's a pretty strong assumption -- particularly if what we are interested in is the impact of a damages cap on payouts by defendants. Consider three possibilities:

1. the cap makes some cases insufficiently remunerative, so they are not brought -- decreasing the volume of cases;

2. the cap changes the economics of some (but not all) cases, so some cases are dropped, but other cases( that used to be insufficiently remunerative) are now worth pursuing, and they take the place of the cases that are dropped -- meaning the volume of cases stays the same;

3. the cap makes malpractice cheaper, and so doctors take less care and injure more people -- increasing the volume of cases.

The first two effects are likely to be realized, if at all, in the short-run, while the third is likely to be realized, if at all, in the long-run. It is hard to know how to sort out this issue in the abstract. Even though the third approach will not provide a clear answer as to the dynamic consequences of a non-econ cap on defendant's payouts, it does have one important advantage -- it tell us what the impact of a cap will be from the perspective of the current group of plaintiffs -- and if you're at all interested in the distributional consequences of a cap, that's worth analyzing.

The final approach is the best way to do these kinds of studies, but the data to do so is generally not available. (We anticipate doing one of these studies around 2011, since that is the earliest the necessary data will be available).

Regardless of which approach one uses, there are additional complexities to be dealt with, such as determining when a cap actually went into effect. That problem is harder than one might think: how should one handle a cap while it is under constitutional challenge in the state courts? How should one handle a cap that was in effect for a while, and then struck down? The answer to both questions will depend on one's sense of the factors that influence insurer behavior. For example, to what extent do insurers discount their expectations regarding cap effects by their expectations of when and whether the cap will be upheld? Do they hold up settlement of cases until it is clear whether the cap will be upheld, or settle them with the expectation the cap will be upheld -- or struck down -- or something in-between?

One final difficulty, which is common to all four approaches, is the problem of obtaining data. When money is transferred from defendants to plaintiffs, it is almost always the result of a settlement -- and it is extremely hard to obtain case-level information on settled cases. It is somewhat easier to obtain case-level information on tried cases, but trials are rare, and, as noted above, the jury verdict does not necessarily indicate the actual payout. The most common source of information on jury verdicts (commercial jury verdict reporters) are systematically skewed toward larger verdicts -- and they usually don't contain information on payouts.

Researchers have used different strategies to address this problem. Some have simply used information on verdicts, while cautioning readers as to the limitations of this approach. Others have obtained information on payouts in tried and settled cases from individual insurers, the National Practitioner Databank, or state closed claims databases. Several states maintain such databases, but not all of them are public. For example, Illinois has a database of all malpractice claims dating back to 1980, but the enabling statute prohibits public release of the information, even if it is de-identified. A slightly dated list of such databases, which is Table 1 in this article, is reproduced below.

The National Association of Insurance Commissioners is working on developing guidelines for states that are interested in creating their own databases. Not surprisingly, one of the flashpoints has been the degree of confidentiality of the reported information. Physician groups have generally opposed public release of the information, even on a de-identified basis. That said, the American Society of Anesthesiologists has used closed claims to identify areas likely to lead to malpractice claims, and to improve the quality of the services they deliver.

In our study we rely on the Texas closed claims database, which includes case-level information on all commercially insured closed medical malpractice claims in which there was a payout > $10,000 nominal. More detailed information is available on cases in which the payout was greater than $25,000 nominal. The database is updated annually, and currently includes the years 1988-2005. The data is here. ("Closed Claim Data"

As my posting from yesterday indicates, the Texas non-econ cap varies from $250,000 to $750,000, depending on the number and type of defendants. The Texas cap is not indexed for inflation.

This post has once again gone on longer than I intended, so I'll just summarize our findings, and provide some more detailed analysis tomorrow. We find that

• The Texas cap reduces the mean (median) "allowed verdict" (the allowable portion of the jury award, plus interest) by 37% (36%). The mean allowed verdict drops from $1.28M to $800k.

• The Texas cap reduces the mean (median) predicted payout in jury verdict cases by 27% (23%). The mean payout drops from $696k to $512k. The reduction in mean payout ($184k) is substantially smaller than the reduction in the mean allowed verdict ($480k). In total, the non-econ cap reduces adjusted verdicts by $156M, but predicted payouts by only $60M.

• Settled cases account for 97.5% of the cases and 95% of the dollars in the dataset. Predicted aggregate payouts in settled cases decline by 18%. The mean settlement payout declines from $313k to $257k. The total reduction in payout is on the order of $780M.

• The non-econ cap has a disparate impact across plaintiff demographic groups, with the larger percentage reductions borne by deceased, unemployed, and (likely) elderly plaintiffs, relative to non-deceased, employed, and non-elderly plaintiffs.

That's enough for today.

J. Aldridge:
I suppose you could say excessive damages rewards are largely symbolic.
12.4.2008 12:14am
Bruce McCullough (mail):
Focusing on mean awards and similar measures does not model the system as it actually functions. These measures are easy to calculate, but they yield little insight into the problem. Agents in this model respond to extremes, not means.

Consider two economies, both with mean award of $100 and n=100 suits. In Economy A, there are 100 awards of $100 each. In Economy B, there are 99 awards of zero and one award of $10,000.

The actions taken by agents in each economy will be markedly different. I give but one example.

In Economy A, someone afraid of a lawsuit only has to prepare for a small award. Since this amount can be paid out of petty cash and will not bankrupt the agent, neither insurance nor defensive measures will be necessary in Economy A. In Economy B, someone afraid of a lawsuit has to prepare for a large award. Insurance and defensive measures will be necessary in Economy B.


Bruce McCullough
Professor of Decision Sciences
Drexel University
12.4.2008 12:56am
JoelP:
Your "third possibility" is framed somewhat oddly. Perhaps it would be better to frame it
'the cap makes malpractice cheaper, and so doctors practice more evidence-based medicine as opposed to defensive medicine; this could increase or decrease the volume of cases'
12.4.2008 1:03am
Anonnnn:
This is really interesting stuff, and I'm looking forward to seeing more of your postings on VC.

Although, I should say that my feelings are mixed since your most recent paper partially pre-empted something I was planning to write (especially since I had already started the number crunching). :(
12.4.2008 2:12am
Patrick216:
The most important factor explaining verdict haircuts is the amount of insurance coverage. If the doctor has $500,000 in policy limits, it doesn't seem to matter whether the jury awards $500,000, $1,000,000, or $5,000,000. The insurer will pay $500,000, and that will usually be the end of the dispute. Above-limits payouts are uncommon, and when they occur, they are virtually always paid by the insurer.

I'm looking forward to this section. I practice mostly in the area of bankruptcy and creditor's rights, and can tell you my bank clients are pretty aggressive about pursuing $500M to $4.5MM in claims, even if unsecured. I have a hard time believing a plaintiff with a significant overage from insurance coverage would walk away, esp. given that doctors are pretty collectable people. I have dealt with debtor physician practice groups and doctors before--they're usually terrified of me, because they typically have enough assets and a high enough income stream that I could inflict a lot of pain if my clients so instruct me. If a doctor had injured me to the tune of $2MM, I'd have no qualms about pursuing the guy into bankruptcy.
12.4.2008 7:30am
Tyger River:
How about treating it similar to bad faith, procedurally?

If P makes an offer, that is within the damage cap, and D refuses, the damage cap is busted.
12.4.2008 8:38am
DavidBernstein (mail):
the cap makes malpractice cheaper, and so doctors take less care and injure more people -- increasing the volume of cases.
Doctors don't pay malpractice awards, insurance companies do. Individual doctors aren't going to change their behavior, most likely; they don't like getting sued, but I doubt getting sued for 1 million versus 1.3 million makes a difference given that the award comes out of someone else's pocket. It might very well make practice much less stressful, however, if you know that your policy limits can't be breached by a runaway jury.

Hospitals, however, might respond to THEIR premiums going up by instituting better practices to avoid malpractice, or insurance companies might try to require better practices for covered physicians to reduce payouts. But if you want to encourage better practices, malpractice awards hardly seems like an efficient way to do so, especially given the Harvard study that shows that few victims of malpractice sue, and few who sue are victims of malpractice.
12.4.2008 9:53am
Anderson (mail):
Prof. Hyman, the VC has a widget whereby previous blog posts in a series or on a given topic are linked at the end of each post.

Since your posts will likely be an internet resource for some time to come, you might want to ask Prof. Volokh etc. about adding this feature to your existing &future posts on damage caps? I'm sure it would be helpful to many readers.
12.4.2008 10:10am
Jon Roland (mail) (www):
What seems missing from this analysis are subjective perceptions. People do not, in general, make economic decisions rationally, based on objective risks and benefits, but on subjective impressions. Even lawyers and insurance companies do that. Thus, if available coverage is exceeded by caps, the fear of overage can be enough to discourage physicians from practicing in a state, and cause him or her to move or go into another field. I have seen this happen with obstetrics and anesthesiology, resulting in shortages of physicians.

I have also seen here in Texas how caps have discouraged lawyers from taking cases on a contingency, even if there was profit potential, because of the difficulty (now increased) of financing a case until payout might be attained. Many have, perhaps irrationally, decided not to take such cases at all, leading to a shortage of available lawyers.

The potential, objective or subjective, risk of either treating or not treating patients is pricing medical care out of reach of people in ways we do not see in countries where litigation is less likely. For my medical care I am covered by VA, but go to Mexico for dental work, where the costs are lower than the co-pay for dental insurance.
12.4.2008 10:45am
einhverfr (mail) (www):
DavidBernstein wrote:

But if you want to encourage better practices, malpractice awards hardly seems like an efficient way to do so, especially given the Harvard study that shows that few victims of malpractice sue, and few who sue are victims of malpractice.


That's the problem, is it not? Defensive medicine is aimed at the latter group (those who are likely to sue whether or not they are victims of malpractice). Since settlements are extremely common, it rewards those who sue regardless of actual malpractice at the expense of the rest of us.

IMO, we need to re-evaluate what we want out of the medical malpractice system. Damage caps strike me as a fairly absurd reform on the basis. If we want to compensate people for loss, let's build that into the insurance system and avoid this sort of abuse of the law. Let's raise the bar for lawsuits to go forward regardless of whether we also build an automatic (no-fault) payout system for things which are currently heavily litigated absent actual malpractice.
12.4.2008 11:26am
Tatil:

If we want to compensate people for loss, let's build that into the insurance system and avoid this sort of abuse of the law.

I don't think the main objective should be compensation for loss. The society in general benefits if those providing medical services, i.e. hospitals, doctors etc., use the best practices to avoid medical errors. If they don't think it will cost them too much even if worse possibility happens, they will only pay lip service. The medical malpractice awards needs to be high enough to concentrate the minds. If they are too high or results too random, it will not add more benefits, but cost a lot more.

Free market works only if buyers have easy access to information about the sellers. However, medical professionals resist providing any information about how successful they are. There was a podcast by CEO of Kaiser Foundation. He told of huge discrepancies in success rates between different hospitals even in treating well known diseases, such as diabetes. Of course, he seemed to be allowed to say which hospitals were doing well, but not which ones were failing.
12.4.2008 12:33pm
NowMDJD (mail):
Prof. Hyman,

None of your methods address the cases that aren't filed.

Hypothetical: A woman earning $100,000/year and who has no children has a colon injury during a gynecologic procedure and requires a colostomy for 2 months. It is reversed at a second operation without complication and she lives happily ever after.

Pecuniary damages are loss of wages (est. $40,000, factoring in recovery time) and excess medical costs, from which (depending on the state) insurance payouts may be deducted) (est. $40,000). Give her husband $50,000 for loss of consortium and loss of domestic services. An attorney may have a low level of enthusiasm for taking this case, as the fee (say 1/3 of 130,000, or $43,333) will be largely eaten up by the cost of trying the case. Or else, a relatively ineffective bottom-feeding lawyer will take the case to keep himself busy.

But in the outer boroughs of New York City, it is not inconceivable that a jury would award $1,000,000 for the pain and suffering. Now you are talking about real money, with a possible award of #1,130,000. A capable lawyer will take the case.

I would guess that the principal economic effect of non-economic caps, especially at a relatively low level like $250,000, would be to keep cases out of court. As I understand your methodology, I don't think you're measuring this effect.
12.4.2008 12:53pm
einhverfr (mail) (www):
Tatil:

I am not saying that we should do this. Hence "if we want...."

However, the problem right now is that there is only a small intersection between those who suffer from medical malpractice and those who sue for it. If that is acceptable because, geeze that baby was born with a birth defect and the doctor neglected some unrelated element of the standard of care (we know how predictable juries are), then we need to build a system which acknowledges this. However, let's stop pretending that the current system really is protecting real victims of medical malpractice rather than (sometimes) victims of bad luck.

Either way, we should be trying to clear the court of bad cases, but still allowing for those which have a clear merit.
12.4.2008 2:51pm
einhverfr (mail) (www):
Tatil:

Also your proposal of medical provider information would probably result in doctors refusing to treat higher risk patients simply because they want to show good numbers.

We saw big problems with Medicare and Medicaid care accessibility under Reagan because DHHS under his administration made these patients too risky to treat.
12.4.2008 2:55pm
Cityduck (mail):

I have a hard time believing a plaintiff with a significant overage from insurance coverage would walk away, esp. given that doctors are pretty collectable people. I have dealt with debtor physician practice groups and doctors before--they're usually terrified of me, because they typically have enough assets and a high enough income stream that I could inflict a lot of pain if my clients so instruct me. If a doctor had injured me to the tune of $2MM, I'd have no qualms about pursuing the guy into bankruptcy.


Plaintiffs attorneys want quick easy money. Even in sophisticated practice areas like securities class actions, it is highly unusual for the plaintiffs' counsel to pursue an individual's assets. That's one reason why Lerach's Enron settlement made such big news: It was a very rare example where D&Os actually paid personal funds.

The correlation between insurance and settlement values is well established. Another reason is that insureds want their defense counsel to sell out their insurers to make cases go away. Prof. Baker &Griffith relayed in "How the Merits Matter: D&O Insurance and Securities Settlements (Mar. 2, 2008) (available at SSRN: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1101068)" how they interviewed securities class action defense counsel who told them: "You know the [insured] doesn't care about the insurance company's money."
12.4.2008 5:17pm
alamac (mail):
Nice post.

I work with a lawfirm in Texas and can promise you that the caps have had some effect, but the real change has to do with the new law (Ch. 74 or our Texas Practices &Remedies Code) and its tightened requirements to successfully sue, the most onerous of which is the necessity of filing an "expert report" which is judged according to the Robinson/Daubert "analytical gaps" theory (there cannot be any "analytical gap" in the chain of causation establishing a breach of the standard of care on the part of the physician). The effect of this is that a lawfirm who wishes to file a case must be ready to spend a minimum of $20,000, and more often nearly $50,000, to retain forensic experts before going forward; and this means that many perfectly viable but small-damages cases are simply not filed. (I turn down several a week precisely for this reason.) I suspect that this self-censoring activity on the part of Texas trial lawyers has a great deal to do with the figures you quote. The damage awards and payouts may not be that much different, but there are many fewer med-mal cases filed now than before. Is justice served? NO. Do insurance companies profit? DEFINITELY. Welcome to Texas.

Just FYI. I enjoy the site!
12.4.2008 6:11pm
JoelP:

I don't think the main objective should be compensation for loss. The society in general benefits if those providing medical services, i.e. hospitals, doctors etc., use the best practices to avoid medical errors.


In this case, you want the people judging to have medical degrees; a courts-based system is a poor judge of "best practices". It is significant that doctors who wish to reduce their malpractice risk tend to focus their efforts on improving documentation, moving their practice to mirror FDA documents rather than current data, and avoidance of high-risk patients. When doctors try to reduce their malpractice risk, it rarely improves patient care.
12.4.2008 7:12pm