The PwC study with AHIP looked at all forms of defensive medicine costs stretching across physicians, outpatient care, hospital, drugs and other medical services (e.g. imaging centers, laboratories, etc.). In doing so, they arrived at a considerably higher figure of almost 10% of total healthcare costs, or $210 billion (Chin, D, et. al., The Price of Excess: Identifying Waste in Healthcare Spending; Health Research Institute; PricewaterhouseCoopers; Boston, MA; April 2008.). This figure constituted the single largest component of the $1.2 trillion in annual waste that these organizations found running rampant throughout the U.S. healthcare system.   But even assuming that this high figure is anywhere close to being accurate, might the practice of defensive medicine actually improve the quality and safety of care? Also, from a risk management standpoint, might this practice either prevent litigation or provide for a better defense of it should it still occur?   Quality and safety of care Certainly in some instances, it is admirable to “pull out all the stops” when diagnosing and treating patients. Patients’ presenting symptoms and initial findings could be very deceiving in the absence of more definitive imaging studies and other interventions. Treatment in their absence could be either ineffective or even harmful.   The problem lies when getting a test, procedure and/or consultation becomes a shorthand way of diagnosing and treating patients. This is especially true when physicians give in to the fear of potential litigation if something may have been discovered from a test, procedure and/or consultation, which was not completed, resulting in patient harm or even death.   Physicians have all heard the horror stories of multimillion dollar verdicts or settlements, especially with young patients where a test or procedure could have uncovered a rare medical problem prior to its catastrophic effects.  Physicians are taught in medical school to diagnose on the basis of possibilities, not probabilities. The fear of litigation may turbo-charge physicians’ ingrained inclinations to work up as many diagnoses to rule in or rule out as possible. But when ordering of tests, procedures and consultations really does step over the line to become defensive medicine, such practices may be perceived as acting more to protect the physician legally, rather than to help the patient medically.   It is difficult to conclude that defensive medicine ever really improves the quality or safety of patient care. For one thing, according to Six Sigma process improvement methodology, the more steps that are introduced into a process (in this case working up a patient), all other factors remaining equal, the greater the opportunities for and actual occurrences of error (even when exercising the same degree of care in executing each process step).   More significantly, the extra steps introduced may carry their own independent risk of harm, whether short term in the form of an invasive procedure, or longer term – such as increasing a patient’s cumulative radiation dosage, which consequently increases chances of cancer down the road (Brenner, DJ, Hall, EJ, Computed Tomography—An Increasing Source of Radiation Exposure; N. Engl. J. Med. 2007; 357:2277-84).     Also, there is the problem of false positive test results that compel physicians to order even more unnecessary tests and procedures. Depending on the sensitivities and specificities of the tests and procedures ordered (as well as the prevalence in the community of the condition being investigated), there could be a high percentage of false positives with a given test or procedure.     To illustrate how the prevalence of a disease, and the sensitivity and specificity of a test or procedure might drive the number of false positive results which could lead to additional unnecessary tests and procedures, consider the following hypothetical example. Suppose a disease has a prevalence of 1% in patients similar clinically and demographically to a patient being considered for a CT scan.  Assume that CT scan designed to detect this disease has a sensitivity of 95% and a specificity of 90% for doing so.  For every 100,000 individuals having that particular test or procedure to rule in or rule out that condition, there will be 950 true positive outcomes, but more than 10 times as many (9900) false positive outcomes. So, for each patient accurately diagnosed in this scenario, more than 10 patients will undergo further unnecessary tests and procedures, experience associated increased anxiety, and produce considerably greater costs to the healthcare system and society (Hunink, MGM, Gazelle, GS, CT Screening: A Trade-Off of Risks, Benefits, and Costs; J. Clin. Invest. 111(11): 1612-1619 (2003). http://www.jci.org/articles/view/18842).   The dramatically increased number of process steps introduced into what could become a cascade of unnecessary tests and procedures will also significantly increase the opportunities and actual occurrences of error, all other factors remaining equal. Such a cascade certainly doesn’t enhance patient satisfaction or cooperation with their physicians (and perhaps the many consultants that become involved). This may not only compromise the quality and safety of care, but also the physician-patient relationship, itself.