The Harm of Missing Mimics

The further harm of thrombolysis in stroke mimics involves diagnostic inertia and anchoring bias, in which the true diagnosis is delayed or missed. In a report from Finland describing outcomes of patients assessed during a neurologic emergency department’s drive for door-to-needle times of 20 minutes, 15 percent of their admissions were initially misdiagnosed.⁶ Of 150 patients with initial misdiagnoses, 70 suffered delays in appropriate treatment, several of whom were specifically harmed by the delay.

Financial harms also result from the treatment of stroke mimics. On a population level, the cost-effectiveness case for thrombolysis in acute stroke is based on the original stroke trials, using outcomes observed in participants with disabling strokes. However, as increasing percentages of patients treated with thrombolysis are stroke mimics, the cost-effectiveness disappears. It is estimated that, at current alteplase vial reimbursement of $8,700, a mimic-treatment rate of roughly 10 percent makes thrombolysis no longer cost-effective on a population level.⁷ On top of excess treatment of stroke mimics, up to half of patients being treated at some stroke centers have “minor stroke” (National Institutes of Health Stroke Scale <6) or lack disabling neurological deficits.⁸ Between treating stroke mimics, minor stroke, and nondisabling stroke, the current approach to thrombolysis has completely eroded population-level cost-effectiveness.

Finally, on the individual level, hospitalization following thrombolysis with an acute stroke diagnosis usually generates higher charges than stroke mimic diagnoses.⁹ These acute stroke charges expose individual patients to potentially greater financial harm, compounded by costs from any long-term medical therapies initiated after stroke. There are further potential harms in the U.S. health care marketplace, as individual insurance coverage becomes potentially more expensive or unobtainable with a serious preexisting medical condition such as ischemic stroke.

There may yet be a debate whether there even remains a debate over the use of thrombolysis for acute stroke. There ought to be no debate, however—our patients deserve better than us burying our heads in the sand regarding the collateral harm of its overuse.

The opinions expressed herein are solely those of Dr. Radecki and do not necessarily reflect those of his employer or academic affiliates.

Bias 101

Ascertainment bias: type of sampling bias in which the way data are collected is more likely to include some members of a population than others Confirmation bias: tendency to search for, interpret, and recall information in a way that supports one’s beliefs or values.