Via Kevin, Ezra Klein points to a large body of evidence contradicting the notion of frivolous law suits against doctors:
As proof, Baker marshals an overwhelming array of research. The most impressive and comprehensive study is by the Harvard Medical Practice released in 1990. The Harvard researchers took a huge sample of 31,000 medical records, dating from the mid-1980s, and had them evaluated by practicing doctors and nurses, the professionals most likely to be sympathetic to the demands of the doctor's office and operating room. The records went through multiple rounds of evaluation, and a finding of negligence was made only if two doctors, working independently, separately reached that conclusion. Even with this conservative methodology, the study found that doctors were injuring one out of every 25 patients—and that only 4 percent of these injured patients sued.Since many of the GOP's claims about frivolous lawsuits has been shown repeatedly to be without merit, why do they insist on repeating themselves?
The Harvard study stands for a large body of literature. On their own, however, the results don't disprove the Republicans' thesis that many medical malpractice suits are frivolous. Maybe badly injured patients don't sue, while the reflexively litigious clog up the legal system, making tort reform a viable solution. But a new study, released in May, demolishes that possibility. Dr. David Studdert led a team of eight researchers from Harvard Medical School and Brigham Young University who examined 1,452 medical malpractice lawsuits. They found that more than 90 percent of the claims showed evidence of medical injury, which means they weren't frivolous. In 60 percent of these cases, the injury resulted from physician wrongdoing. In a quarter of the claims, the patient died.
Nor is there evidence to show that the level of jury awards has shot up. A recent RAND study looked at the growth in malpractice awards between 1960 and 1999. "Our results are striking," the research team concluded. "Not only do we show that real average awards have grown by less than real income over the 40 years in our sample, we also find that essentially all of this growth can be explained by changes in observable case characteristics and claimed economic losses."