The University of Iowa
The University of Iowa News Services Home News Releases UI in the News Subscribe to UI News Contact Us
5137 Westlawn
Iowa City IA 52242
(319) 335-6660; fax (319) 384-4638

Release: July 12, 2002

NOTE TO EDITORS: For a copy of the research article discussed in this release, contact Renee Wilson at the Journal of General Internal Medicine at (410) 955-9868. For interviews with Arthur Hartz, the study's primary author, contact Becky Soglin at (319) 335-6660 or

Physician Surveys Could Be Useful In Medical Malpractice Cases

In medical malpractice cases, defending and prosecuting attorneys commonly hire experts to provide testimony about whether the way a physician cared for a patient was negligent or met customary care standards. However, testimony of expert witnesses may be biased or uninformed.

A better way of providing expert opinion on standard of care in medical malpractice cases may be to randomly survey a group of peer physicians and use their collective responses to identify customary care, according to a University of Iowa study that appears in the July 12 issue of the Journal of General Internal Medicine. The survey method shows promise as an objective method to assess usual practice for specific medical problems. Results from such surveys can be used in court or in settling a case before trial.

The UI investigators focused on seven primary care medical malpractice cases: three actual closed cases, three variations of two of those cases and one active case. In all the cases, the facts of a patient's symptoms were not disputed, but the physician did not make the correct diagnosis or failed to refer the patient to a specialist.

The important question was whether the majority of peer group physicians would have made referrals or ordered diagnostic tests, necessary steps to make the correct diagnosis, said Arthur Hartz, M.D., Ph.D., the Quality in Medicine Professor in the UI Department of Family Medicine and the study's principal author. Cases where the issue is failure to diagnose in a timely way make up about one quarter of all medical malpractice cases nationwide.

"Underlying such cases is the question of customary care -- whether the defending physician practiced mainstream medicine," Hartz said. "Because referrals and tests are costly and sometimes risky to the patient, they cannot be ordered in all cases. The question in medical malpractice cases is whether most physicians would have ordered the tests for patients who are similar to the plaintiff."

Jurors decide how most physicians practice based on the testimony of a few medical experts, who may give conflicting opinions. Because jurors lack the medical expertise to judge the substance of the testimony, they must rely on the experts' style and credentials, which are fallible indicators of accuracy.

"The most common criticism of expert witnesses is they are biased because they are hired and prepared by each party's attorney. To reduce bias it has been suggested that judges, not lawyers, bring in one or two medical experts. However, unbiased witnesses may also be wrong," Hartz said. "Few physicians know how others practice. Our research suggests that most physicians think others practice the same way they do."

"The difficulties with adversarial expert witnesses are well known and have generated significant efforts at reform," said Michael Green, a former faculty member of the UI College of Law who contributed to the study and now is at Wake Forest University School of Law.

The physicians surveyed in the UI study all practiced in Iowa. The surveys included responses from 219 of 350 queried community family physicians (a nearly 63 percent response rate) and 110 of 216 queried community specialists (a 51 percent response rate). In addition, responses were received from 91 percent of 54 academic family physicians and 51 percent of 54 academic specialists.

For the three closed cases, 47, 78 and 88 percent of the family physicians surveyed said they would have made a different referral decision than the charged doctors did. For these three cases, testimony was available from the opposing medical experts on the actual case. In addition, for each closed case, 65 percent or more of the surveyed physicians disagreed with testimony from one medical expert.

The three variations on two of the actual cases helped the investigators determine that physician judgments depended on the quality of the medical management more than the severity of the outcome.

In one case the patient died of a heart attack but had atypical symptoms of heart disease. The majority of physicians determined that the physician charged with malpractice had acted adequately even though the patient died.

In another of the closed cases, a young woman did not receive a timely diagnosis of appendicitis. She survived but suffered due to the missed diagnosis. In that case, the vast majority, 80 percent, of the physicians surveyed determined that the defending doctor had provided inadequate diagnostic care, even though the case did not result in death.

"That type of information may be more helpful to the jury than the conflicting testimony of two adversarial experts, or it may help the jury break a tie when confronted with conflicting testimony," Green said.

Joshua Lucas, now a fourth-year student in the UI Roy J. and Lucille A. Carver College of Medicine, began this study as a summer student research project. He played a key role in summarizing the closed cases into vignettes and then randomly selecting and surveying physicians. To survey physicians, the researchers summarized each medical malpractice case with a one-page abstract.

"We were able to simplify the cases without distorting them," Hartz said.

In addition, the study showed that physicians were interested in medical expert testimony. Hartz said he expected a 25 to 35 percent response rate; instead, 63 percent of the family physicians contacted responded to the survey.

"This high response rate without compensation shows that physicians believed the study was important and not burdensome," Hartz said.

Hartz said that response rates may be even higher if the surveys are sponsored by medical societies and physicians are compensated for their time. However, the authors also recognize that it is a big step from testing this method in a research study to having it widely adopted in the courts.

"The research shows that the surveys are practical and provide valuable information. The next step is to find out how they can be used in practice. We want to find lawyers who will use this method," Hartz said.

He added that a serious concern, usually held by plaintiff's lawyers, in medical malpractice cases is whether physicians are prone to professional allegiance and will, unconsciously or not, strive to protect one another. The UI study revealed that physicians overall are willing to criticize each other.

"There may be a small group of physicians who will not criticize care they know to be bad. However, the majority of the physicians in our study were critical of the physician defendants. When surveyed physicians supported the defendant, the support was usually well reasoned," Hartz said.

Hartz said the current malpractice system is unfair to both physicians and patients.

"If physicians have a chance to improve the system, they will. They would only hurt themselves by defending bad care," he said.

Green also sees promise in the survey method.

"We are acutely aware of the problems with the medical malpractice system," he said. "Surveys are not a panacea, but they make a contribution to improvement."

In addition to Hartz, Green and Lucas, the investigation included Timothy Cramm, M.D., a recent graduate of the UI College of Law, and UI investigators Suzanne Bentler, research associate in family medicine; John Ely, M.D., associate professor of family medicine; Steven Wolfe, M.D., associate professor (clinical) of family medicine; and Paul James, M.D., the Iowa Academy of Family Physicians Chair in Rural Medicine.