Ordering tests that aren’t indicated can backfire legally
Many emergency physicians (EPs) believe that ordering a diagnostic test is always legally protective, but this is a dangerous fallacy, according to Robert Suter, DO, MHA, professor of emergency medicine at UT Southwestern Medical Center in Dallas, TX.
“That is absolutely not true. You really have not done yourself any favors whatsoever by ordering a test that is not indicated with the rationalization that it is good risk management,” he says.
Ordering a test indicates that the EP believes there is a reasonable probability that a particular condition exists. “You can’t say that you didn’t suspect a diagnosis that you ordered a test for,” says Suter. “A nasty plaintiff attorney could say, ‘Doctor, did you order that test because you are unethical or because you are stupid?’”
Since most tests do not have 100% sensitivity, the EP might someday be in the position of explaining to a jury why the patient appeared to be discharged solely on the basis of the negative test result. It could appear to a jury that the EP felt that simply ordering a relatively low-sensitivity test was sufficient to exclude the diagnosis.
“That is a huge trap that even good emergency physicians fall into, and they don’t seem to realize it,” says Suter. For EPs wavering over whether or not to order a diagnostic test, Suter says good documentation explaining why it wasn’t done can strengthen the EP’s defense:
That the EP considered more serious causes of the complaint, but based on the history and physical exam, there was insufficient reason to justify testin.
For instance, the EP might document, “Chest pain and other symptoms relayed by the patient are not consistent with myocardial infarction (MI). There are no symptoms of sufficient duration to believe that the patient has an MI or that would indicate diagnostic use of cardiac enzymes in the ED.”
That the EP gave the patient appropriate warnings on reasons to return to the ED.
The EP might chart, “Patient warned to come back for further testing if they develop persistent chest pain or other symptoms.”
In the case of troponins, patients can get a falsely reassuring message if they are told simply that blood tests show they didn’t have a heart attack. “I’ve seen some really tragic cases where a patient gets discharged from the ED, or is even admitted to a chest pain unit for serial troponin examinations and discharged when they are negative, when their chest pain was clearly of anginal duration,” says Suter.
In some cases, the chest pain recurred but the patient didn’t return to the ED because he or she relied on the negative test results, leading to a massive MI.
“That is a real tragedy for the patient, and is a setup for a multimillion dollar lawsuit,” says Suter.
A much better outcome for both the patient and the EP could have been obtained, he adds, by communicating the options and clinical uncertainty at the time of the first visit, then telling the patient to return by emergency medical services if the pain comes back and persists.
Liability Fears May Contribute to Crowding
A 2013 commentary hypothesized that defensive practices may contribute to ED crowding.1 “We seem to spend a lot of time on low-acuity patients where we worry about missing things. In the back of my mind, I always thought this was because we are worried about liability risks,” says Darren P. Mareiniss, MD, JD, the paper’s author. Mareiniss is an emergency medicine faculty physician in the Department of Emergency Medicine at the University of Maryland School of Medicine in Baltimore, MD.
EPs’ liability fears have been linked to increased propensity to extend workups, request low-acuity admission, and order radiologic studies,2,3 and ED crowding has been shown to negatively affect patient care.4 In light of this, EPs should consider how their workup might be inappropriate, suggests Mareiniss. “By contributing to crowding, defensive practices could detract from care in the ED,” he says.
Tests May Give False Assurance
Labs or imaging tests with insufficient diagnostic sensitivity can give both EPs and patients false reassurance. For example, “cardiac troponins are 100% sensitive for MI at 12 hours, and that is a wonderful fact,” says Suter. However, infarction takes 20 minutes of complete myocardial ischemia; lesser episodes do not cause MI, and these periods of ischemia cannot reliably be detected by the troponin test.
Some EPs order troponins on patients who have less than 20 minutes of symptoms, and so fall into the ischemia rather than the infarction spectrum of acute coronary syndrome. “Troponins are only 55% sensitive for ischemia,” notes Suter. “That is okay as long as the patient is going to be admitted.”
However, if a patient with symptoms lasting five to 10 minutes per episode is sent home from the ED after a negative troponin level and then subsequently has an MI, “you have really put yourself in a much worse position by ordering the troponin, than if you hadn’t,” says Suter.
If the patient sues for failure to diagnose an MI, the plaintiff attorney will ask the EP why the patient was sent home when cardiac disease was suspected. If the EP responds that he or she didn’t think the patient had cardiac disease, the lawyer will ask why the troponin was ordered.
“At that point, you only have two answers, neither of which is good,” says Suter. The EP can respond that he or she didn’t realize the test wasn’t 100% accurate. “The plaintiff attorney can then ask, ‘Did you miss class when they talked about the physiology of MI and cardiac enzymes, doctor?’ It could be very ugly,” says Suter.
Or the EP can respond, “I decided to make a life or death decision for my patient, the plaintiff, based on a test that I knew was only a little bit more accurate than flipping a coin,” says Suter.
For a patient with chest pain, whom the EP plans to discharge because the symptoms do not appear to be cardiac and are of insufficient duration to be an MI, Suter says that in most cases, the EP is better off not ordering cardiac enzymes. Instead, he says, “justify your disposition based on clinical and historical factors arguing against cardiac disease considerations.”
Similarly, instead of ordering a CT scan for a patient with abdominal pain that appears to be benign, for instance, a better approach might be to discharge the patient with a careful clinical examination and informing the patient that “there is always a very small chance it is something more serious that is too early to diagnose,” says Suter, with clear instructions to return to the ED if the pain persists or worsens.
“That is a far better plan than to do a CT scan and tell them that they don’t have a serious condition, when you should know that CT is not 100% sensitive for many common abdominal conditions, and so there is still a chance of a serious disease being present,” says Suter.
CT scans of the brain are often ordered for patients presenting with weakness, dizziness, or headache without EPs carefully considering the appropriateness and reliability of the test, says Jonathan Edlow, MD, vice-chair and director of quality in the Department of Emergency Medicine at Beth Israel Deaconess Medical Center in Boston.
“EPs often do it because it’s fast and widely available, and the patients are glad they’ve had a test,” he says. “But if a CT is ordered without thinking through the sensitivity of the test, it can get them into trouble.”
For a patient with an acute-onset thunderclap headache, CT is not reliable for acute subarachnoid hemorrhage (SAH) if the patient presents more than six hours after onset, notes Edlow. “CT is a pretty lousy test for an acute ischemic stroke, and it’s a horrible test for acute cerebellar or brainstem stroke,” he adds.
Edlow has reviewed several claims against EPs in which the chart suggested that the EP believed that stroke or SAH was excluded because of a negative CT scan. “The plaintiff’s side will pick it apart and say there was a bad outcome because the emergency physician didn’t do the right test,” he says.
If the EP doesn’t think that the patient’s story suggests a cerebrovascular cause, says Edlow, it’s better to document the reasons than to do a scan that is not helpful to rule out the diagnosis. “Of course, one has to have good reasons — elements of the history and physical examination — that inform one’s actions,” he says.
- Mareiniss DP. Could fear of malpractice contribute to ED crowding? Amer Journ Emerg Med 2013;31(11):1612-1613.
- Katz DA, Williams GC, Brown RL, et al. Emergency physicians’ fear of malpractice in evaluating patients with possible acute cardiac ischemia. Ann Emerg Med 2005; 46:525–533.
- Wong AC, Kowalenko T, Roahen-Harrison, S, et al. A survey of emergency physicians’ fear of malpractice and its association with the decision to order computer tomography scans for children with minor head trauma. Pediatr Emerg Care 2011;27:182-185.
- Berstein, SL, Aronsky, D, Duseja, R, et al. The effects of emergency department crowding on clinically oriented outcomes. Acad Emerg Med 2009;16:1-10.
- Jonathan Edlow, MD, Department of Emergency Medicine, Beth Israel Deaconess Medical Center, Boston, MA. Phone: (617) 754-2329. E-mail: email@example.com.
- Darren P. Mareiniss, MD, JD, Department of Emergency Medicine, University of Maryland School of Medicine, Baltimore, MD. E-mail: firstname.lastname@example.org.
- Robert Suter, DO, MHA, Department of Emergency Medicine, UT Southwestern Medical Center, Dallas, TX. Phone: (214) 648-4838. E-mail: email@example.com.