A study by Dr. John Flynn and medical student Robert Miller of bone and joint specialists presented at the AAOS conference finds that thirty-five percent of medical costs (and 20% of total tests) are driven by defensive medicine. [AP] This figure isn't perfect: the study is self-reported (which can lead to exaggeration), and there was no qualitative assessment of whether the defensive medicine resulted in benefits that partially offset the marginal costs, even if an individual test was not cost-effective. (Compare: the famous Kessler/McClellan study where only zero-benefit procedures were counted as defensive medicine.) But even if the study exaggerates the problem three-fold, it goes to show that the indirect costs of our medical malpractice system far exceed the direct costs.