Changes in Physician Supply and Scope of Practice During a Malpractice Crisis : Evidence from Pennsylvania

A major point of contention in the policy debates over tort reforms during the most recent malpractice "crisis" has been whether rising liability insurance costs lead physicians to relocate to lower-cost states, restrict the scope of their practice (for example, by eliminating high-risk procedures), or stop practicing altogether. We investigated the extent to which these dynamics occurred in Pennsylvania, one of the states hit hardest by rising insurance premiums, by analyzing 1993-2002 Pennsylvania insurance department administrative data on liability insurance policies held by 47,366 physicians in 18 specialties with high and low liability risk. Our analyses addressed three questions relating to the malpractice "crisis period," 1999 to 2002: (1) Did the per-capita supply of high-risk specialists in Pennsylvania decrease? (2) How frequently did high-risk specialists stop practicing in Pennsylvania? (3) How frequently did high-risk specialists narrow their scope of practice to exclude high-risk procedures? We used low-risk specialists during the crisis period and high-risk specialists in the years preceding the crisis period as benchmarks for comparisons. For nearly all specialties, the proportions of high-risk specialists restricting their scope of practice were not significantly different during the crisis period than before. Overall, less than 3% (average annual percentage) of physicians performing major procedures shifted to minor procedures only (0.7%) or no procedures (1.8%) during the malpractice crisis period, and 8.2% of specialists performing only minor procedures shifted to no procedures. In the analysis of physicians leaving Pennsylvania practice, most specialties saw 10 to 20% of their ranks exiting practice each year. However, there were no statistically significant differences in the proportion leaving between high-risk and low-risk specialties, or among high-risk specialists during versus before the crisis period. The overall supply of physicians in Pennsylvania (including new entrants to the Pennsylvania market) grew 5.8% between 1993 and 2002, from 181 to 191 per 100,000 Pennsylvania residents. The total per-capita supply of high-risk specialists increased slightly during the crisis period. The only notable decrease in physician supply was in Obstetrics-gynecology: the supply of Ob-gyns dropped 8% during the crisis period. However, the decline continued a trend that predated the onset of the crisis. Our analysis found much more modest effects of the liability crisis on physician supply than have been suggested by physician survey studies. We discuss several methodological issues which may explain the disparate findings regarding physician supply effects in studies that administrative datasets and studies that rely on physician survey data. AN ELECTRONIC COPY OF THE PAPER IS AVAILABLE FROM THE FIRST AUTHOR