“Evidence-based medicine” (EBM) is a mantra mouthed by many physicians and health care policy makers today. The idea sounds good on the surface: that medical treatment be based on sound studies showing a significant statistical benefit (vs. harms due to side-effects) of a particular drug or treatment. When wasteful and ineffective treatments are rooted out of the system, patients will benefit, and treatments will be more cost-efficient. This is, advocates of EBM affirm, a continuation of the tradition of medicine since the scientific revolution of the seventeenth century: good medicine is based on the best science of the day. What could be the problem with that?
Set the clock back to the 1970s. Medicare and Medicaid, even then, were costing taxpayers dearly. Medical costs were rising rapidly, and policy makers sought some means of controlling such costs. Government officials began to collect data from doctors and from hospitals on the drugs and medical procedures used to treat particular diseases and injuries. This included data on the average number of days spent in the hospital, for example, after surgery for acute appendicitis. In this way Diagnosis Related Groups (DRGs) were born.
DRGs were originally meant to be useful data so Medicare and Medicaid could better know how their money was being spent. But in the early 1980s, Medicare and Medicaid used DRGs to limit care to what had been customary. These federal programs would only pay for customary care; if a patient had to stay in the hospital an extra day after surgery, tough. Later, private insurance firms established similar policies, policies that ignored the needs of individual patients. Not everyone is “customary” in the care needed. Someone with a fever after appendicitis surgery might need to stay in the hospital an extra day or two, but insurance companies make it difficult for the patient to be compensated.
Now “evidence-based medicine,” if used the wrong way, can function like DRGs. It can be used to place all treatments and patients undergoing those treatment into a single category that ignores individual differences between patients. This can lead to undertreatment, in the case of a patient that needs more drugs or a treatment that “evidence-based medicine” does not support. Or it could lead to overtreatment if a patient has a condition that would normally be considered risky, but that patient is an exception. Take the rule of thumb that more than six PVCs (“skipped heartbeats” that originate in the ventricles, the lower chambers of the heart) a minute are a danger sign. Often I get far more than six a minute–sometimes fifteen a minute, sometimes ten, sometimes with runs of bigemeny (one normal beat, one PVC, and so on), sometimes trigemeny (two normal beats, PVC, two normal beats, PVC, etc.) But after I was given a stress test it was determined that I was one of the exceptions and that these PVCs are not dangerous for me. But suppose a physician followed the general rule and treated me with beta blockers. These carry their own side effects and risks. The point is that even the best evidence-based conclusions may not fit every individual patient. So prudence remains necessary in applying evidence based medicine to particular patients–medicine is based on science, but medicine itself is a practical art. Hopefully, evidence-based medicine will be used wisely to benefit patients and prevent needless or harmful treatments. Applied solely as a cost-cutting measure or applied legalistically without prudence, it can do patients more harm than good.