To the Editor:

Despite research on the overall costs of the U.S. medical liability system,1 national data are limited on the costs associated with resolving medical malpractice claims2 — defense costs — and how they vary according to physician specialty. The frequency of malpractice claims and the size of awards vary considerably across specialties,3 yet the degree of variation in defense costs across specialties is unknown. Defense costs constitute an important expense for insurers, and they affect physicians and patients by raising the costs of malpractice premiums and medical care, respectively.

We analyzed defense costs associated with 26,853 malpractice claims closed between 1995 and 2005 among 40,916 physicians covered by a nationwide professional liability insurer. We have used claims from this insurer elsewhere to study malpractice risk according to physician specialty.3Each claim contained information on whether or not a payment was made to a patient, the specialty of the physician involved, and the defense cost associated with the claim. Defense costs included factors that were directly associated with the cost of defending an individual claim, such as filing and expert-witness fees, but they did not include items that could be spread across multiple claims.

The mean (±SD) defense cost associated with claims was $22,959±41,687. Claims in which an indemnity payment was made were associated with higher mean defense costs than claims with no payment ($45,070 vs. $17,130, P<0.001). Mean defense costs varied considerably across specialties, both among claims with and without indemnity payments For example, mean defense costs per claim resulting in an indemnity payment were $83,056 in cardiology and $78,890 in oncology, as compared with $24,007 in dermatology and $23,780 in ophthalmology. We found that mean defense costs in cases that did not result in indemnity claims were lower in all specialties, but there was still significant variation (from $7,283 in nephrology to $25,073 in gynecology). The correlation between mean costs in claims with and without payments was small and positive (0.39).

These findings show that although the costs of dispute resolution are higher for claims that result in indemnity payments, there is still a meaningful cost of resolving claims that never result in payment. Moreover, there is substantial variation in these costs across specialties, though this variation is not necessarily the same as the variation in the frequency or size of claims. Lowering the costs of dispute resolution could lead to considerable savings for physicians and insurers, particularly in specialties with high mean defense costs.

Seth Seabury, Ph.D.
RAND, Santa Monica, CA

Amitabh Chandra, Ph.D.
Harvard University, Cambridge, MA

Darius Lakdawalla, Ph.D.
University of Southern California, Los Angeles, CA

Anupam B. Jena, M.D., Ph.D.
Massachusetts General Hospital, Boston, MA
[email protected]