The Case

A 72-year-old woman with history of opioid abuse was sent to the emergency department (ED) from a methadone clinic because she appeared confused when she came to get her daily dose of methadone. In the ED, the patient complained of pain in her epigastric region and back. She was given IV fluids for tachycardia (heart rate 140 beats per minute [bpm]) and hypotension, with prompt improvement in her blood pressure but not in tachycardia. The initial electrocardiogram revealed sinus tachycardia at 156 bpm and initial laboratory test results revealed a white blood cell count of 19,000/μL, troponins at 0.38 μg/L, and creatine kinase of 270 U/L. A CT scan of her abdomen revealed multiple fractures at thoracic and lumbar spinal region with possible cord compression.

Neurosurgery was consulted and the attending requested an MRI to further evaluate the spinal cord. The patient was given pain medication but remained tachycardic. The MRI, performed about 13 hours after admission, revealed an acute L2 fracture with cord compression. Neurosurgery recommended conservative management until morning. The patient was admitted to the ward, still tachycardic and complaining of pain in the epigastric region. The medical team attributed her pain to the fracture.

The next morning, the patient's troponin levels were in the 20s μg/L and creatine kinase was approximately 900 U/L. However, because of a delay in reporting, by the time the medical team saw the laboratory test results, it was 28 hours into the admission. A stat electrocardiogram revealed ST elevation in the inferior leads, diagnostic of a myocardial infarction. A "Code STEMI" (ST-elevation myocardial infarction) was called, but the clinicians chose to delay taking the patient to the cardiac catheterization laboratory, since they wanted the use of any antiplatelet agents to be cleared by neurosurgery in light of her spinal fracture. A bedside echocardiogram revealed severely reduced left ventricular function with an akinetic inferior wall and an ejection fraction of about 10%.

The patient was eventually stabilized on cardiac medications. At the time of discharge, it was unclear how much of her cardiac function might recover. As the team reflected on the case, they believed that they had not responded aggressively enough to the patient's epigastric pain because of her history of drug use and chronic pain.

The Commentary

by Maria C. Raven, MD, MPH, MSc

This case illustrates the challenges inherent in providing acute medical care for individuals with comorbid behavioral health conditions, especially those who are controversially referred to as frequent fliers. Frequent fliers, or more appropriately, individuals with frequent use of emergency departments (EDs), account for a disproportionate share of health care visits and costs.(1) Many also use other segments of the health care system frequently, have serious medical conditions, and are more likely to be admitted to the hospital than individuals who are not frequent ED users.(2) Individuals with frequent ED visits often have underlying co-occurring mental health and/or substance use diagnoses, yet typically their chief complaint in the ED is not related to these conditions, but rather to nonbehavioral medical problems.(3) Many also have unstable housing and inadequate social support networks that complicate their medical care.

The Challenges of Obtaining an Accurate and Unbiased History

Frequent ED users and those with mental health and substance use diagnoses are vulnerable to diagnostic errors and provider biases that negatively impact care, especially when they seek acute health care from providers who do not know them. Despite the fact that a significant proportion of ED patients have underlying behavioral health disorders (4), emergency providers receive little instruction during their training regarding how to best and most safely care for them. Often, ED patients have no one to advocate for them or corroborate details of their medical, behavioral, and social history, and it can be difficult to obtain a history from a patient with severe mental illness or one who is actively using substances. These challenges can prompt providers to label patients as "poor historians" (5), thus providing an excuse for an insufficient history of present illness. In such instances, providers may rely on documentation from previous visits rather than taking extra time to talk to the patient directly or to reach out to a patient's family or outside providers. The phenomenon of copying and pasting (6) previously documented history into a current note can perpetuate a history that may or may not be accurate or relevant to the current visit. It can also give false merit to underlying suspicions of malingering or drug-seeking behavior that can result in withholding adequate diagnostic testing and treatment.

In this case, the treating physicians conflated previous opioid abuse with the patient's current (appropriate) methadone treatment. Moreover, they assumed that because the patient was on methadone and had a history of opioid abuse that she was "drug seeking," ignoring her persistent tachycardia and lab abnormalities. Research has shown that providers' clinical impressions often do not match objective data regarding drug-seeking behavior (7), and racial and ethnic biases play a role in our opioid prescribing.(8) In other words, we can't fully trust our impressions and should be reluctant to document them without objective evidence.

Diagnostic Overshadowing

In turn, these assumptions about the patient's former opioid use and current methadone use presented a second pitfall for providers: diagnostic overshadowing.(9) First described in 1982 (10), diagnostic overshadowing is related to clinicians' tendency to misattribute symptoms of individuals with learning disabilities to their cognitive deficits, leading to underdiagnosis of other comorbid illness. Diagnostic overshadowing also manifests as the process by which an individual receives insufficient or delayed care because the treating provider incorrectly attributes physical symptoms to an underlying mental illness or substance use disorder. This misattribution can prevent or delay providers from making accurate or complete diagnoses, even if such conclusions would seem obvious in patients without behavioral health conditions.

In the case above, a 72-year-old woman complained of epigastric pain and had abnormal vital signs and an elevated troponin. Acknowledging that this is a complex case because she had an alternative diagnosis (multiple fractures) that was easy to anchor (11) onto, these findings should have prompted further evaluation (e.g., serial electrocardiograms; trending troponin) followed by further diagnostic cardiac testing.(12) However, the patient's former opioid use overshadowed the correct diagnosis of an evolving STEMI in multiple ways. First, it affected the vigilance of the treating providers: despite having a significant, persistent tachycardia, ongoing pain, and an elevated troponin, she was admitted to the floor rather than to a step-down or intensive care unit setting that would allow for more intensive monitoring. Second, it inhibited her providers from recognizing clear clinical evidence (hypotension, tachycardia, elevated troponin, and epigastric pain) of an evolving inferior wall myocardial infarction.

Interventions to Reduce Diagnostic Errors

The case above highlights the fact that diagnostic errors contribute to the excess mortality experienced by individuals with mental health and substance use diagnoses.(13) Multiple categories of interventions can reduce diagnostic errors; three types may be effective to improve care for this patient population.(14) Educational interventions that augment teaching in health professional training about how to optimize care for individuals affected by mental illness and substance use disorders are needed, with an emphasis on this population's increased risk of comorbid medical illness and mortality. In addition, specific techniques such as motivational interviewing can be introduced as a part of curricula to improve patient–provider communication. Case conferences to review actual examples of diagnostic overshadowing can help to alert providers of this bias. Technology-based systems interventions, such as the use of text or pager alerts to communicate critical laboratory values or vital signs, may eliminate the reliance on busy clinicians to repeatedly check for laboratory abnormalities and reduce the chance they are overlooked. If communicated to an entire treating team, they may also stimulate some team members to rethink and expand a diagnostic workup. Finally, providers' technique must be improved so that their approach to cases like the one above is standardized: abnormal vital signs and laboratory values must be acknowledged and addressed. We must rely on our own history taking and verify assumptions about drug-seeking or other subjective behaviors with objective evidence, such as data from prescription data monitoring programs.(15) Also, we can learn to incorporate expertise from other team members, such as ED or hospital-based social workers. If a patient cannot provide a complete history, we should be concerned that something is clinically amiss, rather than automatically labeling a patient as a "poor historian." The onus is on the provider, not the patient, to obtain as full and accurate history as possible.

Take-Home Points

• Health care providers, especially acute care providers in the emergency and inpatient settings, require more training to provide high-quality care for frequent ED users, many of whom have underlying mental health and substance use conditions.
• Frequent ED users have a high burden of medical disease and are frequently admitted to the hospital.
• Diagnostic overshadowing is an important yet under-recognized patient safety issue. To overcome it, providers must:
  • Recognize that it exists and can adversely impact their ability to provide high-quality care.
  • Increase efforts to obtain an accurate history when caring for frequent ED users and those with behavioral health diagnoses and to recognize the need for objective data when documenting.
  • Use interventions that can reduce diagnostic errors and enhance patient safety.

Maria C. Raven, MD, MPH, MSc Associate Professor of Emergency Medicine School of Medicine University of California, San Francisco San Francisco, CA

References

1. LaCalle E, Rabin E. Frequent users of emergency departments: the myths, the data, and the policy implications. Ann Emerg Med. 2010;56:42-48. [go to PubMed]

2. Sommers AS, Boukus ER, Carrier E. Dispelling myths about emergency department use: majority of Medicaid visits are for urgent or more serious symptoms. Res Brief. 2012;(23):1-10. [go to PubMed]

3. Billings J, Raven MC. Dispelling an urban legend: frequent emergency department users have substantial burden of disease. Health Aff (Millwood). 2013;32:2099-2108. [go to PubMed]

4. Coffey R, Houchens R, Chu BC, et al. Emergency Department Use for Mental and Substance Use Disorders. Rockville, MD: Agency for Healthcare Research and Quality; 2010. [Available at]

5. Gillman MW. The patient as historian. N Engl J Med. 1992;326:1785. [go to PubMed]

6. Siegler EL, Adelman R. Copy and paste: a remediable hazard of electronic health records. Am J Med. 2009;122:495-496. [go to PubMed]

7. Weiner SG, Griggs CA, Mitchell PM, et al. Clinician impression versus prescription drug monitoring program criteria in the assessment of drug-seeking behavior in the emergency department. Ann Emerg Med. 2013;62:281-289. [go to PubMed]

8. Singhal A, Tien YY, Hsia RY. Racial-ethnic disparities in opioid prescriptions at emergency department visits for conditions commonly associated with prescription drug abuse. PLoS One. 2016;11:e0159224. [go to PubMed]

9. Shefer G, Henderson C, Howard LM, Murray J, Thornicroft G. Diagnostic overshadowing and other challenges involved in the diagnostic process of patients with mental illness who present in emergency departments with physical symptoms—a qualitative study. PLoS One. 2014;9:e111682. [go to PubMed]

10. Jones S, Howard L, Thornicroft G. 'Diagnostic overshadowing': worse physical health care for people with mental illness. Acta Psychiatr Scand. 2008;118:169-171. [go to PubMed]

11. Etchells E. Anchoring bias with critical implications [Spotlight]. AHRQ WebM&M [serial online]. June 2015. [Available]

12. Jneid H, Anderson JL, Wright RS, et al. 2012 ACCF/AHA focused update of the guideline for the management of patients with unstable angina/non-ST-elevation myocardial infarction (updating the 2007 guideline and replacing the 2011 focused update): a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol. 2012;60:645-681. [go to PubMed]

13. Harris EC, Barraclough B. Excess mortality of mental disorder. Br J Psychiatry. 1998;173:11-53. [go to PubMed]

14. McDonald KM, Matesic B, Contopoulos-Ioannidis DG, et al. Patient safety strategies targeted at diagnostic errors: a systematic review. Ann Intern Med. 2013;158(5 Pt 2)381-389. [go to PubMed]

15. Paulozzi LJ, Strickler GK, Kreiner PW, Koris CM; Centers for Disease Control and Prevention (CDC). Controlled substance prescribing patterns—prescription behavior surveillance system, eight states, 2013. MMWR Surveill Summ. 2015;64:1-14. [go to PubMed]