American’s obsession with safety. It’s even more important than quality. But how do we measure it? Safety metrics are easy when it comes to fields like transportation or infant formulas where the expected rates of death and injury are supposed to be zero. So we make this the expected benchmark and try to get these rates to be as low as possible.
But with medical care, the picture is far more complicated. In this case, the outcome as a result of an illness or condition in the absence of medical treatment is EXPECTED to be death and injury (i.e., permanent disability). This is your benchmark. Therefore, it’s often difficult to determine whether a death or injury in this context was the result of treatment failure or it was iatrogenic (as a result of treatment itself) or a combination the two and whether, in retrospect, things could have been done differently that may have improved the odds for the patient.
Notice, I didn’t use the term “preventable”. Medical care does not prevent anything. It maximizes the odds ratios for patient morbidity and mortality. I prefer terms like “this treatment will reduce the likelihood” instead of “this treatment will prevent” such and such. Treatment with bisphosphonates will reduce the likelihood of a fracture. The use of daily aspirin will reduce the chances of a heart attack or stroke. There is however, a remaining chance that a fracture or heart attack will occur despite these treatments.
And every single medical treatment has the potential for iatrogenic harm. The use of blood thinners to reduce the risk of clots in the lower extremities increases the risk of bleeding. The use of invasive monitoring equipment will improve that ability for physicians to make treatment decisions but at the risk of infections, bleeding, and other complications. Ideally, multiple complex treatments are applied in various combinations that try to maximize the odds for recovery while minimizing the likelihood of a poor outcome though this is never assured.
Decades ago, people understood these simple concepts. But through a combination of the success of medical care, the growth of the tort industry, and American obsession with safety, we have come to have unrealistic expectations regarding safety in medical care. One of the biggest reasons for this delusion is that far too many people have lumped medical errors, iatrogenic complications, poor outcomes, and non-medical (i.e., administrative or systemic) errors together into one big messy pile.
This was evident when the Institutes of Medicine (IOM) released a study in 1999 that found that up to 98,000 patients die each year as a result of medical errors (they even included “to err” in the title of the report). But this figure was significantly revised when the methodology of the IOM study was re-evaluated. The IOM study simply looked at hospital charts and depended on the conclusion of the reviewer to determine whether an error was present. However, the accuracy of such methodology was found to be highly variable on further evaluation. It’s like watching two Monday morning ESPN analysts slug it out. Was the interception a result of a bad throw, a bad decision, good defense, a mistake by the receiver, or bad luck, or a combination? Different “experts” can see it differently. This type of study is not helpful (except for lawyers who still quote the study on their web sites).
Amazingly, few people make the distinction between systemic errors and poor outcomes as a result of medical treatment. Systemic errors are errors that should never happen due to things like poor equipment design or defects, improper equipment maintenance and usage, wrong site, wrong procedure, wrong patient errors, poor training, and mismatched blood typing. Unlike medical complications, these adverse outcomes are completely preventable and have little to do with actual medical practice and more to do with administrative problems. Yet, studies and the media continue to lump preventable systemic/administrative adverse events in with medical complications like infections, bleeding, and medication reactions that will occur even under the safest of situations.
The first step is to separate systemic errors from medical adverse outcomes because the expectations and approaches to each can be quite different. Systemic errors are approached much the same way the FAA approaches aircraft accidents. Based on a root cause analysis, equipment is redesigned and/or new or revised procedures are implemented to try and prevent future errors.
Complications as a result of medical treatment requires an understanding and acceptance that this the trade-off we make in order to aggressively treat illness and that a certain finite level of risk will exist even under the most ideal set of circumstances until and unless new technologies or practices are developed to reduce this risk. With this in mind it is more appropriate to measure rates of complications related to the best known techniques in order to set a benchmark.
But even the use of complications or mortality and morbidity rates can be misleading in that different hospitals tend to treat different subsets of patients who may be sicker or healthier than the general population.
Some suggest that better metrics might be how well care is coordinated, how closely evidence-based medical guidelines are adhered to and how short the wait times are. But no matter how we ultimately measure medical safety, we need to start by understanding that there is a significant difference between system/administrative errors and medical complications.
What is defensive medicine?
Defensive medicine is the deviation from sound medical practice to avoid the threat of malpractice litigation.
According to a study in JAMA, over 90 percent of physicians surveyed admitted to practicing defensive medicine. This can range from “positive” defensive medicine, like ordering unnecessary tests, referring to consultants, or performing unneeded procedures; to “negative” defensive medicine, like avoiding high-risk patients or procedures.
Why practice defensive medicine?
Physicians practice defensive medicine to avoid malpractice litigation. A malpractice lawsuit is the most scarring ordeal that a physician can undergo, both emotionally and financially. There is an expectation that doctors have to be 100% accurate with their diagnoses. A missed diagnosis, whether it’s a 1 in a 100, or 1 in a million case, makes a physician vulnerable to a malpractice lawsuit. However, medicine by nature is an imperfect science, and the expectation of perfection is not realistic, nor possible. There is no test that is 100 percent accurate - an unfortunate outcome can occur even if a doctor practices textbook medicine. A recent study in the NEJM showed that almost 40 percent of malpractice cases were found to be without medical error.
Due to this uncertainty regarding unfortunate outcomes, physicians err on the side of caution and practice defensive medicine. It is much easier to defend the fact that a doctor ordered a test, as opposed to not ordering the test at all.
What are the downsides to defensive medicine?
Defensive medicine is expensive, has no basis in evidence-based study, and exposes the patient to a host of complications. Contrary to popular opinion, more medicine does not equal better medicine.
If a physician is 99 percent sure of a diagnosis, how much more will a patient be willing to pay for that added 0.5 percent certainty? A CT scan costing hundreds of dollars? An MRI costing thousands? Now, there may be some patients who are willing to spend that money for the most thorough workup possible. However, it is not feasible to routinely spend that kind of money to achieve minimal improvements in diagnostic certainty which may not benefit the patient at all.
Defensive medicine is one of the most important drivers in rising health care costs today. There are few reliable studies to back this up. This is because defensive medicine is impossible to quantify. There is a fine, and largely undefined, line separating thorough care and defensive medicine. What one doctor may interpret as a “being cautious”, another may say is defensive. Because defensive medicine cannot be quantified and is so subjective, its impact on the cost of health care has been minimized and under-publicized.
The practice of ordering extra tests is also bereft of evidence. There are no studies suggesting that ordering PSA screening tests saves lives from prostate cancer, or ordering routine abdominal CT scans saves lives from appendicitis. Is relying on the evidence good enough? The answer is no. The standard of care used in medicine cannot be applied to the courts. Standard of care varies from jury to jury.
Take the case of Dr. M. Major clinical guidelines, including the American Cancer Society and the American College of Physicians, suggests that the physician discuss the pros and cons of PSA screening tests with the patient. Since there is no evidence that this test saves lives, and may in fact cause damage by leading to unnecessary prostate biopsies, it is recommended that the decision of whether the test is ordered be shared between the doctor and patient. Dr. M did just that, and documented the discussion appropriately. Unfortunately, the patient later went on to develop late-stage prostate cancer, and sued Dr. M and the hospital for not ordering the PSA test. Citing the clinical guidelines of the ACS and ACP did not help - the hospital was found to be at fault. Again, it is much easier to defend the fact that a physician ordered the test, as opposed to not ordering the test at all.
Many would think that “the more tests, the better”. Nothing could be further from the truth. Tests themselves have their own risks: ranging from radiation exposure from CT scans to serious complications like bleeding and infection from needle biopsies.
Since no test is 100% accurate, unnecessary testing can lead to “false positives”. This is defined as having a positive test result in the absence of disease. False positives lead to progressively more invasive tests, which may eventually lead to a non-dangerous diagnoses or even nothing at all. As the tests become more invasive - like a needle biopsy or cardiac catheterization - the complications become more dangerous. Exposing patients to these unnecessary complications, for the sake of avoiding malpractice litigation, is bad medicine.
What can a patient do to help curb defensive medicine?
A patient should understand why a test is being ordered. Ask questions. How necessary is the test? What diagnosis are you looking for? What are the risks of not doing the test? What are the risks of the test itself? Understand that the goal of perfection in medicine is impossible, and that simply ordering more tests is not necessarily better medicine.
How can defensive medicine be reduced?
Obviously a difficult question, since it is a difficult entity to even quantify. The focus of the question is, “What does the physician want to avoid?” The answer of course, is the ordeal of malpractice litigation. Even if physicians do win the majority of malpractice cases that make it to trial, the mere process of a malpractice suit is tremendously scarring. Remember, the vast number of cases are settled, never reaching a jury. And the solution is not simply, “practice better medicine” or “make less mistakes”. Keep in mind that almost 40 percent of malpractice cases do not involve medical error. Unfortunate outcomes despite textbook medicine are a fact of life - that is a hard truth that one has to accept.
Some have suggested that capping malpractice awards, no-fault insurance, arbitration, or health courts as approaches to curb defensive medicine. Although I suspect that these options will help curb defensive medicine, the probability of such sweeping reforms happening in the near future appear minimal.
A simpler way would be to have clinical, evidence-based, guidelines globally applied to malpractice cases. That way, standard of care would be more consistent, and not vary from jury to jury. It will lessen the impact of “hired gun” experts, who can support whatever standard of care is convenient to the lawyers. Doctors can then focus on practicing evidence-based medicine, confident that the standards they are held to in the community, will be the same in the legal world. Only when that confidence is gained, will defensive medicine start to decline.