April 28, 2005
All insurance programs are potentially subject to costs created by the liability environment. For example, a recent CMS letter to the Medicare Payment Advisory Commission (MedPAC) indicated that spending on physician services during 2004 rose by approximately 15 percent. A significant driver of this increase is the fact that more patients are receiving more complex and more frequent imaging services, such as magnetic resonance imaging and computer tomography scans. For several years now, in fact, spending for these diagnostic services has been rising at a more rapid rate than overall physician expenditures. Based on my own research and the research of many academic experts, my interactions with other physicians, and my experience as a clinician, it is clear to me that the practice of defensive medicine is contributing to these cost increases. The evidence suggests that reforms to the malpractice system, including caps on non-economic damages and revision of the joint and several liability rules can reduce defensive medicine, which can reduce unnecessary health care expenditures. The CBO scoring of legislation in 2003 estimated that Federal expenditures would drop by nearly $15 billion over ten years. Those savings depend only on reduced premiums. My own research concluded a reduction in defensive medicine could lower overall hospital expenditures by between five and nine percent. During FY 2004, the Medicare program spent more than $133 billion on hospital fee-for-service. That would mean potential annual savings of between $6.65 and $11.97 billion dollars, just for that program, not to mention the private sector.
Even more importantly, liability reforms will improve quality and access to health care, leading to better health for Americans. I would urge the Congress to work with the Administration to formulate a plan to address the problems with our current liability system and to promote a culture of patient safety and quality within the healthcare arena. The changes in liability law have the potential not only to produce significant savings, but also to simultaneously improve patient safety and the quality of care.
This morning I would like to review some of the systemic problems in medical malpractice liability and some innovative alternatives for addressing the needs of those who have been medically injured. Specifically, I would like to highlight the Department’s “Early Offers” program as one possible way to speed resolution of malpractice claims so that patients’ needs are satisfied in an effective, efficient manner.
The Current System Does Not Work
It is well known that the vast majority of individuals injured by a caregiver do not file suit. The 1990 Harvard Medical Practice study reported that only 2 percent of individuals experiencing an adverse event due to medical negligence filed suit and, of more concern, only 1 in 14 individuals seriously injured by such an event received any sort of compensation. More recent work by some of the same researchers confirms these findings. The Physicians Insurance Association of America reports that, on average, it takes more than 5 years for an insurer to pay a malpractice claim after the date of the incident--mostly due to delays in reporting (22 months) and delays in the tort system (43 months). When an injured patient does finally successfully settle or win a case, lawyers typically take anywhere from 30 - 40 percent of those funds as compensation. In short, many of those who are injured due to negligent care are simply not receiving justice because the system does not work for them.
On the other side of the coin, the current system does not do much in terms of screening out cases with no medical merit, or in differentiating between adverse events due to negligence and unavoidable adverse events. A study published in the New England Journal of Medicine found:
…no association between the occurrence of an adverse event due to negligence or an adverse event of any type and payment. ... among the malpractice claims we studied, the severity of the patient's disability, not the occurrence of an adverse event or an adverse event due to negligence, was predictive of payment to the plaintiff.
The same study reported that 10 of 24 cases involving no adverse event whatsoever were settled with a mean payment of nearly $29,000. Six of 13 cases involving an adverse event not due to negligence were settled with a mean payment of more than $98,000. More broadly, of claims filed during 2003, only about a third resulted in some payment to the plaintiff, and of the small percentage that go to trial, more than three in four resulted in a finding for the defendant, immediately leading one to question the validity of the bulk of claims.
Rapidly rising premium rates can have a real impact on patient access to care. A study by the Agency for Healthcare Research and Quality examined how the supply of physicians varied across states between 1970 and the present. The study concluded that states adopting caps on non-economic damages experienced about 12 percent more growth in physicians per capita than States without caps. Notably, the study also found that States with relatively high caps were less likely to experience an increase in physician supply than States with lower caps. This sort of disparity can translate into very real access challenges. It means that it is more difficult for patients to find the types of specialists they need, that they must go further out of their way, and take more time from their own lives to access the care they require. In some cases, the limitations on access result in negative health outcomes as well.
Just to illustrate, a 2004 survey of Ob/Gyns in Illinois found that in the previous two years, 11 percent had stopped practicing obstetrics as a result of medical liability concerns. Based on how many office visits physicians report in an average month (N=250), that means 46,250 office visits for Ob/Gyn services were lost across the state during those two years.
The malpractice system has important adverse effects on quality as well. In a widely read 1999 report, “To Err is Human,” the Institute of Medicine (IOM) noted that:
[R]eporting systems are an important part of improving patient safety and should be encouraged. These voluntary reporting systems [should] periodically assess whether additional efforts are needed to address gaps in information to improve patient safety and to encourage health care organizations to participate in…reporting, and track the development of new reporting systems as they form.
The IOM emphasized that fear of lawsuits deters doctors and hospitals from making reports, even when they are not negligent, because in many states such reports can be used against them in court. This very understandable concern impedes quality improvement efforts. If our liability laws do not encourage error reporting and analysis, they serve only to perpetuate the very problems that they ostensibly exist to address.
The truth is that common human decency and professional ethics are sufficient motive for the vast majority of physicians to provide the best care possible. Most medical errors today are not the result of bad doctors or nurses, but rather the result of complex or difficult systems in which they work.
You would think that we would do everything in our power to encourage the kind of self-analysis and systems evaluation necessary to identifying and addressing systemic errors. Instead, our current tort system sets up roadblocks that discourage this very important activity. This roadblock needs to be removed.
Congress should pass patient safety legislation that includes a mechanism for allowing anonymous reporting of errors and that protects databases of such information from discovery. If we don’t collect this data, we’ll never see the patterns that will allow us to make changes to improve patient safety and will never realize the concurrent savings resulting from reduced errors.
The Costs of Our Current System
The article’s findings on the impacts of liability reforms on cost and quality are supported by a substantial body of other work. In an earlier study published in the Journal of the American Medical Association, researchers found a positive relationship between malpractice claims risk and rates of cesarean sections. In a 2002 paper also published in a peer-reviewed economics journal, Dr. Kessler and I further explored the role of malpractice reforms in reducing defensive practices. Dr. Kessler and I found that malpractice reforms affect physician behavior by changing both financial measures of "malpractice pressure" (such as malpractice claims rates and malpractice insurance premiums) and non-financial measures (such as the time and hassle spent in defending against a claim).
Based on the work we did, Dr. Kessler and I concluded that if direct liability reforms had been adopted nationwide between 1984 and 1990, it would have resulted in annual savings of $450 million for each of the first two years and close to $600 million for each of the succeeding years for just the two conditions we studied. As I mentioned earlier, our study concluded that these reforms could potentially reduce overall hospital expenditures by five to nine percent. Those kinds of savings, if realized, could have a significant impact on the fiscal health of the Medicare and Medicaid programs. Furthermore, as stated above, these savings would come without any drop in the quality of care and outcomes experienced by patients.
CBO has taken issue with the estimates from the paper written by Dr. Kessler and me, contending that tort reform will not reduce defensive medicine. CBO used our work as a model, but their efforts are hampered by two critical methodological limitations. First, when CBO sought to replicate our study on a more recent sample of patients with the conditions we examined, it obtained similar results to ours. The finding of insignificant effects arose only when CBO sought to re-estimate our models on a set of patients with very broadly defined illnesses. Because hospital expenditures on patients with a broad range of illness are likely to be heterogeneous and hard to predict, the unexplained variance in hospital expenditures for these patients is likely to be large—larger than the unexplained variance in hospital expenditures for patients with clearly defined illnesses we studied. Since the standard errors of the estimates of the effects of limits on liability are proportional to the unexplained variance in expenditures, the statistical significance of estimates from models with broadly defined illnesses would be less than the significance of estimates from models with narrowly defined illnesses.
Second, we used more comprehensive data, while CBO used data from a 20 percent random sample of beneficiaries for most (1991-1996) of their study period. Third, there was very little variation in states’ tort laws during the CBO’s entire study period (1991-1999)—according to CBO staff, only 6 states changed one or the other of the two liability system variables under analysis. In the period that we studied (1984-1994), 33 states changed one or the other of the liability system variables under analysis. These two differences—the less comprehensive data and the smaller number of “experiments” in the CBO analysis—would also lead the statistical significance of estimates reported in their brief to be lower than the significance of our estimates.
It is important to put the differences between myself and Dr. Kessler, and the CBO, in the context of what we focused on. CBO has not made estimates of savings from reductions in defensive medicine. They have, however, concluded that reduced premiums would save the Federal government billions of dollars. My own research shows the potential for billions more in savings as a result of reduce defensive medicine. What we both end up saying -- along with numerous other researchers – is that reforms will lead to billions of dollars in savings each year.
Liability Concerns Reduce Physician Productivity
The perceptions of practitioners themselves back up these statistical results. A 2002 poll by Harris Interactive found that the fear of litigation impacts healthcare administrative issues. Well over three-fourths of all physicians and nurses (84% and 81%, respectively) reported that they spend more time on paper work, such as medical record documentation, because of malpractice concerns than they would based solely on the patient's clinical needs. Additionally, nearly all physicians (94%) believe that written descriptions of cases are very often or sometimes influenced by the fear of litigation.
In a 1997 paper, Dr. Kessler and I investigated how the intrusiveness of the liability system affected physician perceptions of medical care. We estimated the impact of liability reforms on objective measures of malpractice pressure -- such as claims rates -- and on perceptions of the effects of malpractice pressure on practice patterns. The study found that malpractice pressure affects physician perceptions of two important dimensions of medical practice: propensity to make referrals, and the ability to spend time with patients.
More generally, the legalistic atmosphere in which physicians practice warps the physician-patient relationship. Hauser et al. give a good example of how fear of litigation can reduce the trust in the physician-patient relationship and actually become a barrier to clear and effective communication.
A woman went to a gynecologist for a problem and a minor surgical procedure was recommended. At the beginning of the discussion of this procedure, the physician commented, “The law requires me to inform you of certain facts about this operation.” And then, in a perceptible alteration of his normal patterns of speech, the gynecologist began to chant a litany of side effects, risks, morbidity, mortality, percentages, probabilities, etc. The patient later reported that after about ten seconds of listening to this, her mind shut down entirely. “This appears to be some sort of arcane ritual! The communication was not directed to me for any benefit of mine whatsoever.”
High-quality medicine requires effective communication with patients. The various tests and procedures available to us provide a tremendous amount of useful information, but often, a diagnosis, or the type of test to utilize, is prompted by something the patient shares with the physician in conversation. If, because of liability concerns, physicians are unable to discuss the inherent ambiguities and complexities of medical practice, and the variety of potential outcomes to a given procedure or service, in a manner to which the patient can personally relate, then the patient’s ability to make informed decisions is compromised. Our current system, because it recasts this relationship in legalistic terms does not promote mutually beneficial exchanges of information.
Innovative Private Sector Approaches
Some time ago, Johns Hopkins hospital began requiring non-emergency patients who came to them for elective procedures (individuals who had the option of going elsewhere if they so chose) to sign an agreement to take any malpractice claims to mediation prior to going to court. In 2003, Hopkins mediated 24 cases and resolved 21 of them. As a result, Hopkins 2003 claims expenses decreased almost 30 percent. Mediation is typically much faster than a court case and involves far lower attorney’s fees. In short, patients who are injured get compensated at a higher level and in a shorter amount of time. Furthermore, this reform has helped the hospital communicate more freely with the patients, and probably with the professional staff, in order to be sure the mediation is successful and the highest possible quality of care is achieved.
Tort Reform and Liability Insurance Premiums
A number of possibilities exist for improving our medical liability system. Tort reforms include actions such as capping awards for pain and suffering, so called non-economic damages, as well as capping punitive damages. In addition, suggestions have been made to reframe rules for joint and several liability, such that each actor involved in a given episode of care, including the physician, hospital, and payer, all bear a level of blame proportional to their share of fault or responsibility. Liability for damages would not be joint. As another option, attorneys’ fees could also be capped, so that more of the dollars won by a plaintiff with a meritorious case actually go to that individual to address their health needs, and large awards could be paid as an annuity, or over a number of years, instead of as a lump sum, so that the money is available in the future when the individual needs it to pay for care. Collateral source rules, taking into account funds coming from health, automotive, or workers’ compensation insurers, could also be modified to allow reductions in settlements or jury awards commensurate with insurers’ payments. Alternatively, mandatory pre-trial screening by an independent medical expert to weed out baseless claims could reduce the number of baseless suits faced by physicians. President Bush supports securing the ability of injured patients to get fast, unlimited compensation for their economic losses, including the loss of ability to provide unpaid services like care for children or parents, but has urged the Congress to support a cap of $250,000 on non-economic damages, limit punitive damages, eliminate joint and severable liability, create a uniform statute of limitations, and provide for the structured payment of future damages.
According to the GAO, the greatest driver of increases in physician liability premium rates is losses suffered as a result of malpractice claims. They also concluded that states with tort reforms that include certain damage caps had lower growth in liability premiums than did those without such caps. Another study by Stephen Zuckerman et al. concluded that capping medical liability awards reduced premiums for general surgeons by 13 percent in the year following enactment of that reform and by 34 percent over the long term. The reforms resulted in similarly lower premiums for general practitioners and Ob/Gyns. A 2002 HHS study found that during 2001, states with meaningful caps on non-economic damages saw average premium increases of 15 percent, while states without such caps saw increases of 44 percent.
Like many academic studies, my own research has demonstrated that direct tort reform, including capping damages, abolition of mandatory prejudgment interest, and collateral-source rule reforms reduce premium expenditures significantly. The 1997 paper with Dr. Kessler I mentioned above showed that in states adopting such reforms, within three years physicians saw substantially and statistically significant lower trend growth in their real malpractice insurance premiums of approximately 8.4 percent.
Amounts paid on malpractice claims, either in settlement or because of a jury award, have been growing substantially in the past few years. The Physician Insurers Association of America (PIAA) reports that the median jury award in medical liability cases nearly doubled from 1997 to 2003, increasing from $157,000 to $300,000. The PIAA’s as yet unpublished report on 2004 indicates that the median jury award during that year was $439,400; a one-year increase of more than 46 percent. It is notable that PIAA found a 2004 mean payment on a jury verdict of $606,907. Such a large difference between the median and the mean indicates the existence of a significant number of large awards. The size of settlements has similarly increased. Median settlements increased from $100,000 to $200,000 between 1997 and 2003. As previously noted, these increasing losses drive increases in premiums. However, physicians must also pay legal fees. Physicians who win at trial have average defense costs of $87,720 per claim and in cases where the claim was dropped or dismissed, their costs averaged $17,408.
There is substantial evidence on the positive effects of tort reform to provide a basis for congressional action at this time. Not only will tort reform result in lower premiums, but, much more importantly, it will help foster an environment in which physicians do not feel the need to engage in defensive medicine and we will see our costs drop as a result. Tort reform will increase access to healthcare and it will result in improved quality as providers feel the freedom to openly discuss systemic improvements that will lead to a higher degree of patient safety. I would urge the Congress to take this issue up and act on it.
The Impact of Market Forces
Insurers typically invest the bulk of their revenues into bonds. Some people argue that during the stock market rise of the 1990s, insurers realized profits from their investments that allowed them to reduce premium rates. They contend that as the stock market has suffered declines, insurers have raised their premiums to make up for investment losses. In addition, many insurers purchase reinsurance from larger entities. Some say that such reinsurance has become increasingly expensive in the past few years, particularly after the tragedy on September 11, 2001, and that it is also common for insurers entering a new market to provide lower introductory rates in order to obtain market share, and then raise the rates once they have an established client base.
In addition to business cycle factors, the St. Paul Company, one of the larger physician insurers in the country, decided to cease providing malpractice coverage at the end of 2002. This action reduced competition among insurers and allowed them to, at least temporarily, increase their premium rates.
Critics of tort reform efforts point to all of these factors as relevant to the malpractice debate. They argue that we should not engage in tort reform if it is not the only driver increasing premiums and expenditures.
The GAO concluded that although none of the companies it examined experienced a loss on their investments, a 1.6 percent decline in investment return from 2000 to 2002 would have resulted in premium increases of 7.2 percent over the same period. Such a decline would not have been outside the realm of possibility given market movement during that period. Studies like these lend credence to the argument that a component of liability premium increases may result from factors other than rising settlements and jury awards.
That said, the present problem in many States is not the result of the so-called “insurance cycle,” or reckless investments by insurance companies. Although we have been on an “up” part of the cycle, that does not explain extremely high premium increases in the last few years in some States that have not reformed their liability system, compared to much smaller increases in most of the States that have implemented significant reforms. The insurance cycle is not a phenomenon that occurs in some States but not others. But the growth in liability premiums and even the availability of liability insurance has clearly varied substantially across states, in association with differences in liability laws.
Consequently, reforms in insurance would not address the underlying causes of the problems of unnecessary costs, lower quality, and less access to care that result from our current liability system. Insurance market reforms will not change physicians’ perception of the liability environment in which they work and market reforms will not reduce the level of defensive medicine. Furthermore, market based reforms will not produce swifter settlement of claims, or improve the equity of injured patients’ compensation.
Additional Steps to Improve Our Liability System
When a patient who has been served at a federally funded health center or Indian Health Center facility files a medical malpractice liability claim against HHS, we send a standard notice explaining our early offers program. Both sides have 90 days to submit a confidential offer to a neutral third party who will compare the offers and notify both sides only if a match is made. Not only are offers voluntary, their amount and existence remain confidential forever if no match is made. So neither side tips its hand or loses leverage if the case goes to court.
The program is up and running at HHS and we’re hopeful that it will show promising results in the months to come. In the meantime, any doctor or hospital can set up an early offers program. Because an early settlement only occurs when both parties agree, you’re not losing any options by setting up a program, and no government action is required.
Evidence on how we can improve quality of care for patients should drive our reform efforts. We should be sure that if doctors take steps to encourage quality, for example, installing and using electronic medical records so that they can more easily track adverse events and thereby prevent them, that these physicians are not then punished by our legal system. If a physician who is considering such a system has in the back of his/her mind the fact that some day an attorney might use his data to bring suit, that physician may abandon the idea altogether. We should be looking to create systems that support quality care, that provide the data that are needed for good decision making.
To illustrate what can happen when physicians are able to be more open with their patients about medical errors, I would point you to the experience of the Lexington, Kentucky Veterans Affairs medical center. In 1987, after losing two malpractice cases with judgment totaling more than $1.5 million, this facility adopted a policy of radical honesty. They began to openly and immediately discuss with patients and/or their families any errors that occurred during treatment, including giving the patient information about their right to file a claim or an application for compensation. Furthermore, the facility disclosed medical errors when the patient or family had no reason to know one had occurred. A 1999 study of the Lexington facility’s claims experience during the years 1990 to 1996 concluded that the facility did not pay any more in malpractice claims than comparable VA facilities, and had concurrently avoided significant legal expenditures. Partially due to the success of the Lexington policy, the VA adopted this practice system-wide in 1995.
The VA is not entirely analogous to the private market, but I bring up this example because it demonstrates how the real needs of patients who have been injured can be addressed more adequately when systems are in place to encourage patient-physician communication.
Mr. Chairman, the current medical liability system simply does not address the needs of patients, and it’s costing those patients, the Federal government, and other payers billions of dollars every year because it adds to costs and encourages care that does not improve health. More importantly, our liability system reduces access and reduces quality of care. I would encourage the Congress to take action on this issue and would be happy to work with you as you move forward. I would be pleased to take any questions at this time.
Last Revised: May 2, 2005