Faulty Data and False Conclusions

Faulty Data and False Conclusions: The Myth of Skyrocketing Medical Malpractice Verdicts
by Lewis Laska and Katherine Forrest, Commonweal Institute, October 2004

A new report from the Commonweal Institute concludes that the notion that medical malpractice awards are rising dramatically in both frequency and size is a myth.

The report says:

“While healthcare costs have been rising and some physicians have curtailed their practices because of rising insurance costs, there are insufficient data to conclude that this is the result of rising medical malpractice awards. In fact, the preponderance of available relevant evidence indicates that this notion is a myth…. The recent spike in medical liability premiums was caused by the insurance cycle, not by increased numbers of new claims or ‘skyrocketing’ jury verdicts.”

According to the report, “Most of the data used to support the notion of ‘skyrocketing malpractice awards’ come from one source – Jury Verdict Research, Inc. (JVR).” JVR’s database is intended for lawyers to understand the value of specific cases. But, as the report points out, “JVR has never asserted that its data should be used for public policy purposes.”

JVR’s reluctance to be associated with public policy purposes is well founded given the problems with its data, as the report makes clear:

• Only jury awards – jury awards comprise only 4 percent of malpractice cases thus JVR data does not account for the 96 percent of cases that do not go before a jury.
  
  
• Not all jury awards – JVR estimates that it only captures about 60 percent of awards.
  
  
• Ignores zero verdicts – JVR sample is biased because it ignores zero verdicts, verdicts reduced by judges or overturned verdicts, rendering its medians and averages meaningless.
  
  
• Disproportionately high awards – JVR relies on inappropriate sources such as newspaper reports, newsletters and voluntary reports. Lower verdicts are much less likely to make the newspaper or to be voluntarily reported. JVR does not purport to collect all trial data.
  
  
• Not the true payout number – JVR does not track reductions in awards, nor appeals that often overturn the initial judgment.

Beyond these fundamental flaws there are other less obvious faults in the data. In 1998, JVR combined both physician and hospital verdicts. Trends in physician liability could therefore no longer be separated out from trends in hospital liability – a fact most journalists fail to understand. It also takes no account of so-called “high-low” agreements, in which the final award can be just a fraction of the compensation JVR records. Finally, JVR’s averages are not qualified by the removal of outliers – the infrequent large awards that drag the average disproportionately (and erroneously) higher. JVR itself warns against the use of its averages in public policy debates because of the “distorted view of data” they give.

There are few sources other than JVR because insurance companies are reluctant to release data publicly. However, when insurance company information has been released, for instance in New Jersey, Massachusetts, Mississippi and Missouri, it has undermined claims of skyrocketing compensation verdicts. “Severity has not gone up, and in fact, it has gone down a little,” according to one insurance executive quoted in the report. There is even more evidence contradicting the notion that the number of cases is skyrocketing. In fact, in most states, the numbers are actually falling, and nationally, National Association of Insurance Commissioners (NAIC) data show a four percent reduction in medical malpractice lawsuits between 1995 and 2000.

The report concludes that there is a clear need for better data to allow a more responsible public policy debate. Insurance companies, healthcare providers and medical accreditation boards must cooperate and take advantage of NAIC’s uniform data collection standards in an effort to provide a better understanding of the truth surrounding the medical malpractice debate. Only then can public policy decisions be based on facts.

November 1, 2004