When Dr. Alverdo saw Richard Edmunds' name on the index card of appointments that the secretary handed him, he thought, "What can I possibly say or do that will make a difference?" Mr. Edmunds, a high school English teacher, was 46 and has a history seasonal allergies. Nevertheless, Mr. Edmunds was in the office, on average, every 6 weeks. His symptoms varied but were always difficult to verify or quantify—pain, discomfort, "just not feeling right."
Three months ago, Mr. Edmunds presented with complaints of headaches. He described the pain as generalized and worse in the afternoon than in the morning. The headache was not accompanied with nausea or visual changes. They never woke him from sleep and were not associated with any particular activity or food. Non-narcotic analgesics were ineffective. Based on the history and physical, which included a normal neurological exam, Dr. Alverdo concluded that the headaches were most likely caused by muscle tension and prescribed a course of anti-inflammatory medications, muscle relaxant, and physical therapy exercises. Mr. Edmunds made 3 visits to the therapist but said the exercises were not helping. He kept asking, "How do you know I don't have a tumor or an aneurysm about to blow, Doc?" Finally, Dr. Alverdo ordered a CT scan, which, he had to admit, he believed was a "shot in the dark." The scan was negative. After that report, Mr. Edmunds complained less about the headaches. Dr. Alverdo hadn't heard about headaches for a couple of months now.
Lately, Mr. Edmunds was having chest pains. Faithful about his annual physical exam, Mr. Edmunds had no previous history of exertional angina or shortness of breath. There was no known heart disease in Mr. Edmunds' immediate family. Dr. Alverdo had ordered an EKG and cardiac stress test, both of which Mr. Edmunds passed with flying colors. Then, thinking the persistent pain might be digestion-related, Dr. Alverdo ordered a barium swallow and upper GI series. Negative.
Looking at the name on the card, Dr. Alverdo thought, what's Edmunds going to want today? At the last visit, Mr. Edmunds reported that he could not sleep because of the crushing pain. The only remaining diagnostics for chest pain were highly invasive. Dr. Alverdo resolved that he would not be talked into ordering an angiogram or anything else that would put Mr. Edmunds at risk. There simply was no indication for it. He imagined how the conversation would go. Mr. Edmunds would dispute everything Dr. Alverdo said. This generally went on for 25 or 30 minutes. "Doctor," Edmunds would say. "Do you want me to be one of those cases you read about in the paper where it says, "he kept telling the doctor he was sick, but no one believed him?"
For years, the typical physician has been plagued by fear of omission. What if she omits a critical test, fails to consider a possible diagnosis or doesn't offer a helpful treatment? Not only would her patient presumably suffer, but she too would face threats—of diminished reputation, lawsuit, and worst of all, self-recrimination. To this fear, clinicians have recently added another source of dread. Subsequent to a recent Institute of Medicine report , physicians and the public have grown increasingly aware of the way in which patients are harmed or killed through errors such as illegible writing, lapses in concentration, and the absence of systematic crosschecks.
But there is a kind of error that is arguably more important and ethically problematic. I will call it the "error of compulsion." Errors of compulsion occur when doctors feel compelled to order tests that they know to be unnecessary or not indicated. These errors tend to evolve from 3 typically overlapping sources: (1) excessive patient activism, (2) excessive physician activism, and (3) fear of recrimination. Such errors are not innocuous, since they frequently lead to needless suffering, needless morbidity and even death. They are ethically problematic because they violate one of medicine's fundamental moral maxims—the rule, often called the "principle of nonmaleficence" ; physicians should not harm patients.
Dr. Alverdo is on the cusp of an error of compulsion, and the source (at least ostensibly) is excessive patient activism. His patient, Mr. Edmunds, has a history of requesting (and getting) diagnostic evaluations that are, at best, "shots in the dark." Now Dr. Alverdo anticipates that he will be able to satisfy Mr. Edmunds only by ordering a dangerous test (coronary angiogram) that has little chance of detecting cardiac pathology. I will argue that if Dr. Alverdo orders the angiogram, he errs by subjecting his patient to an unjustified risk.
Before I make my case, however, I should concede that the angiogram might offer some benefit. First, it is possible, despite the inconclusive nature of Mr. Edmunds' symptoms and the negative workup, that Mr. Edmunds has occult coronary artery disease that would be detected through coronary angiography. Second, as Mr. Edmunds' earlier CT of the head seems to exhibit, a definitively negative test can have therapeutic value.
But a remote possibility of pathology does not justify undertaking a risk-laden procedure such as a coronary angiogram. It is rarely possible in medicine to rule out disease with absolute certainty. Physicians must make their recommendations based on probabilities, and here the probability of a coronary etiology is small. Likewise, the prospect of symptom relief through reassurance is not enough to justify the risks of a coronary angiogram. There are other, safer ways to address Mr. Edmunds' symptoms.
If he orders another unnecessary test, Dr. Alverdo will reinforce a dangerous and maladaptive trend in his clinical relationship with Mr. Edmunds. It would be better for Dr. Alverdo to address Mr. Edmunds' compulsion for reassurance–thus averting a potentially vicious cycle of debilitating worry, followed by excessive workup, followed by new worries. Dr. Alverdo should explain the dangerous implications of this cycle. Though financial considerations also pertain (since it would be exorbitantly expensive to pursue comprehensive testing for every unlikely diagnosis), Dr. Alverdo should focus on what is best for Mr. Edmunds. Somehow, Mr. Edmunds must come to terms with medical uncertainty. To wit, he must understand that it is not possible to explain every symptom and it is dangerous to try.
In his references to newspaper cases where the patient "kept telling the doctor he was sick, but no one believed him," Mr. Edmunds opines that undiagnosed pathology is the "worst case" scenario for patients with chest pain and other symptoms that could be linked to dangerous conditions. The response, for Dr. Alverdo, is to explain that exceptional cases do not make good precedents. The real "worst case" scenario occurs when patients die in the course of unnecessary testing. Would Mr. Edmunds choose a very low probability of finding occult pathology when it brings a higher probability of suffering unnecessary complications? Occasionally, such frank discussions fail to help patients overcome their medical worries, and psychiatric referral is indicated.
Excessive patient activism is a corruption, through excess, of the principle of autonomy—ie, the principle that competent patients ought to have the prerogative to decide for themselves. Though patient autonomy is important, it does not compel physicians to offer dangerous or unhelpful interventions. Often excessive patient activism is indirectly encouraged by physicians. In such cases, the real culprit may be excessive physician activism—an overly developed instance of physicians' commendable inclination to "do something" for suffering patients. When it is not tempered by prudence and caution, this powerful motive begets a dangerous "technological imperative." Despite their activist tendencies, and despite fears of legal liability, physicians are beholden primarily to cultivate and protect the well-being of patients. This objective is not served when physicians cave in to imprudent demands.
- Kohn LT, Corrigan J, Donaldson MS. To Err Is Human: Building a Safer Health System. Washington, DC: National Academy Press; 2000.
Beauchamp Tl, Childress JF. Principles of Biomedical Ethics. 5th ed. New York: Oxford University Press; 2000.