September 2006

Grant Results

National Program

Improving Malpractice Prevention and Compensation Systems

SUMMARY

From 1998 to 1999, researchers at Northwestern University studied the impact of the National Practitioner Data Bank on the process of resolving medical malpractice claims. The data bank is the repository for mandated reports of medical malpractice payments and other adverse actions against physicians.

The research team analyzed closed claim files obtained from insurance companies as well as claim files maintained by Florida to determine the validity of two hypotheses:

  • Physicians may be unwilling to settle some cases because they do not want a report made to the data bank.
  • Hospitals and other corporate defendants may be shielding physicians by agreeing to settle suits on condition that codefendant physicians are dropped from the case.

The project was part of the Robert Wood Johnson Foundation's (RWJF) Improving Malpractice Prevention and Compensation Systems national program.

Key Findings

  • The results of the analyses were consistent with the first hypothesis.
  • The results of the analyses provided "only limited support" for the second hypothesis.

Funding
RWJF supported this project through a grant of $322,030.

 See Grant Detail & Contact Information
 Back to the Table of Contents


THE PROBLEM

The National Practitioner Data Bank (NPDB) is an information repository created by Congress in 1986 to restrict the ability of incompetent physicians and certain other health care providers to move from state to state without disclosure of their records. Physician malpractice payments in any amount must be reported to the NPDB as well as adverse actions regarding licensure, clinical privileges, and professional society membership. Hospitals are required to query the NPDB when granting or reviewing clinical privileges or medical staff membership, and other health care entities are encouraged to do so. The NPDB began operating in 1990. Its information is not open to the general public.

From its inception, the NPDB generated controversy. Some critics questioned the reliability of its information as an indicator of incompetence. Others contended that the punitive potential of an NPDB listing impeded the voluntary resolution of malpractice claims.

There was also concern that the focus on the deficiencies of individual practitioners undermined efforts to improve care on a systematic basis. Another suspicion was that corporate defendants — hospitals, health maintenance organizations, and group medical practices — might be agreeing to settle malpractice cases on condition that any individual physicians named as codefendants are dropped from the suit. Such "corporate shielding" would be possible because settlements against corporations were not required to be reported.

In November 1996, the IMPACS national program office convened a meeting of experts to discuss potential research to address questions of this nature surrounding the NPDB. As a result of the meeting, in April 1997, the national program office issued a Call for Proposals inviting studies in two principal areas: (1) the sources, sizes, and implications of gaps and errors in NPDB information; and (2) the response of health care institutions to the NPDB, particularly decisions on physician licenses, credentials, privileges, and discipline.

A collaborative proposal by researchers at Northwestern and Harvard universities was selected for funding; the funding went to Northwestern, which subcontracted work to the team at Harvard. The same Harvard researchers oversaw the IMPACS-funded study of medical injuries in Colorado and Utah (see Grant Results on ID#s 022603, 023685, 029907, and 032865).

 Back to the Table of Contents


THE PROJECT

This grant funded a study of the impact of NPDB reporting requirements on the process of resolving medical malpractice claims. Specifically, the study examined the validity of two hypotheses: (1) physicians may be unwilling to settle some cases because they do not want a report made to the NPDB; and (2) hospitals and other corporate defendants may be settling cases on condition that codefendant physicians are dropped from the suit. The Harvard University researchers collected much of the study data under the subcontract. Additionally, CNA was paid for access to its case files and research space, and the research director of the Physician Insurers Association of America was hired as a consultant.

The project team obtained information on a total of 3,406 closed claims from four insurance companies (CNA, the Utah Medical Insurance Association [UMIA], Copic Insurance, and the Controlled Risk Insurance Company [CRICO], the self-insurance program for Harvard University Affiliated medical institutions, employees, and physicians). The claims included pre- and post-NPDB cases and, while not a random sample, were considered by the project team to be representative of claims in more than 20 states. The claims data were merged with relevant ancillary data sets and analyzed to determine the impact of various factors, including NPDB reporting requirements.

Additionally, the team obtained and analyzed data on closed claims in files maintained by the states of Florida, Indiana, and Maryland. Initially, the team planned to merge the different data sets. But because of varying claims patterns, the team decided to make the insurance company information its primary data set and the Florida data a secondary set. The Indiana and Maryland files, because of difficulty in reconciling the information, were not used in testing the NPDB hypotheses.

For each hypothesis, the project team established sub-hypotheses that could be tested by the claims data. If the first hypothesis (physicians may be unwilling to settle some cases because they do not want a report made to the NPDB) was correct, the team reasoned that the following should result:

  • An increase in the number and proportion of claims dropped without payment, especially among cases involving small dollar amounts or non-permanent injury.
  • Delay in the settlement of claims, resulting in an increase in resolution time and defense costs.
  • Prevention of settlements, resulting in an increase in the number of cases reaching trial.

If the second hypothesis (hospitals and other corporate defendants may be settling cases on condition that codefendant physicians are dropped from the suit) was correct, the project team hypothesized the following results:

  • A change in the composition of the defendants in settlements, resulting in a decrease in the percentage of cases with a named physician defendant; the absolute number of defendants per suit; and the proportion of defendants per suit who are physicians.
  • An increase in the dollar amounts of settlements against hospitals and other corporate entities and a decrease in the dollar amounts of settlements against physicians.

The team tested the first set of sub-hypotheses separately using both its primary and secondary data. Because of difficulty in identifying codefendants in the insurance company claims, only the Florida data were used to test the second set of sub-hypotheses.

 Back to the Table of Contents


FINDINGS

In a report made to RWJF, the project team outlined the following findings on the two hypotheses:

  • First Hypothesis: The results of the insurance company claims analysis were consistent with the hypothesis that the NPDB generated an unwillingness to settle on the part of physicians. The study team reported to RWJF that initially after NPDB's implementation (1991–92), trial rates became "significantly higher" and the proportion of claims that were dropped without payment became "slightly higher." In the longer term (1995–96) — "perhaps as plaintiffs and their lawyers developed an understanding of the new litigation dynamic" — trial rates returned to pre-NPDB levels, but the proportion of cases dropped without payment remained higher than before the NPDB requirements. Defense costs also remained higher (in real terms). However, instead of an increase in the resolution time of cases, there was weak evidence of a decrease.
  • Second Hypothesis: The analysis of the Florida claims data provided "only limited support" for the hypothesis that corporate defendants were shielding physicians. The project team reported no change in the percentage of cases with a named physician defendant, and only an initial "transitory" decrease in the number of defendants per suit. However, the team told RWJF that "there does appear" to be a long-term decrease in the proportion of defendants per suit who are physicians. The team also reported "strong evidence" that dollar amount of corporate settlements more than doubled in the post-NPDB periods, while physician settlements increased less than 20 percent.
  • The increase in the proportion of cases dropped without payment has "competing implications" that warrant further analysis. It may indicate that the NPDB is having "an adverse impact on some consumers with legitimate claims," especially consumers bringing small claims that are not likely to be worth the cost of litigation," the project team said. On the other hand, raising the settlement threshold may also have the effect of "screening" out frivolous claims.

Limitations

Use of the Florida data to test the first hypothesis produced differing results from the test using the insurance company data, including a decrease in the post-NPDB proportion of cases dropped without payment. Also, the second principal hypothesis was tested solely with the data from Florida — a state that the project team acknowledged has unique characteristics.

Communications

Members of the project team wrote and submitted for publication an article reporting the findings on the first hypothesis. An article on the second-hypothesis analyses was in preparation at the date of this report. Additionally, project staff presented their findings as part of a panel discussion at an annual meeting of the Association for Health Services Research. (See the Bibliography.)

 Back to the Table of Contents


LESSONS LEARNED

  1. Researchers collecting primary data from company files with which they are not familiar should be prepared to encounter problems that they did not envision. The Harvard researchers had previous experience working with files of three of the four insurance companies (UMIA, Copic, and CRICO), and the abstraction of their claims data went relatively smoothly. However, team members were unfamiliar with CNA and encountered obstacles in abstracting that data.
  2. To create a well-rounded project, it is necessary to actively seek and bring together the necessary components. The national program director took a proactive role; he worked with the teams at Northwestern and Harvard to nurture their collaboration, and proposal to the IMPACS program.

 Back to the Table of Contents


AFTER THE GRANT

The project director participated in a subsequent NPDB-related project funded by the federal Health Resources and Services Administration; that project included a survey of NPDB users and non-users on their perceptions of the NPDB program. She did additional work on the NPDB for the US Department of Veterans Affairs.

 Back to the Table of Contents


GRANT DETAILS & CONTACT INFORMATION

Project

Research on the Impact of National Practitioner Data Bank Reporting Requirements on the Resolution of Malpractice Claims

Grantee

Northwestern University (Evanston,  IL)

  • Amount: $ 322,030
    Dates: January 1998 to December 1999
    ID#:  033494

Contact

Teresa M. Waters, Ph.D.
(901) 448-1189
twaters@utmem.edu
Peter P. Budetti , M.D., J.D.
(847) 491-5643
p-budetti@law.northwestern.edu

 Back to the Table of Contents


BIBLIOGRAPHY

(Current as of date of this report; as provided by grantee organization; not verified by RWJF; items not available from RWJF.)

Articles

Waters TM, Studdert DM, Brennan TA, Thomas EJ, Almagor O, Mancewicz M and Budetti PP. "Impact of the National Practitioner Data Bank on Resolution of Malpractice Claims." Inquiry, 40(3): 283–294, 2003. Abstract available online.

Presentations and Testimony

Teresa M. Waters and Peter P. Budetti, members of a panel discussing "How Can Consumers Be Protected in the Health Care Market Place?" at the Annual Meeting of the Association for Health Services Research, June 28, 1999, Chicago.

 Back to the Table of Contents


Report prepared by: Michael H. Brown
Reviewed by: Robert Crum
Reviewed by: Marian Bass
Program Officer: Joel C. Cantor
Program Officer: Beth A. Stevens
Program Officer: Judith Y. Whang

Most Requested