Content, Consistency, and Quality of Black Box Warnings: Time for a Change.

A black box warning (BBW) is the highest level of warning issued by the U.S. Food and Drug Administration (FDA). It aims to inform the patient and the prescriber of the potential major risks or lifethreatening adverse reactions associated with the use of a medication. The warning is usually placed on top of the drug information label and is highlighted with a black box that distinguishes it from the rest of the text. These warnings are required when there is reasonable evidence of association between the drug and a significant safety concern suggested by either human or animal studies (1). Boxed warnings have been associated with changes in clinical practice. A survey conducted in 2009 to assess the effect of BBWs for long-acting -agonists found that 42% of primary care physicians and 35% of specialists were likely to change their treatment regimen for asthma on the basis of the warning (2). In 2010, Dorsey and colleagues published a timeseries study in which they examined the use of atypical antipsychotics before and after issuance of a BBW. They found a significant decrease in their use in the years after the BBW was issued (3). Despite their potential importance, BBWs have been the subject of controversy, due in part to their opaque connection to the underlying body of evidence. In 2004, Stades and colleagues called for easing of the restrictions on metformin use after demonstrating the low-quality evidence supporting its association with lactic acidosis. They also argued that the metformin BBW predisposed scientists to find a link when none existed (4). In 2006, researchers questioned the BBWs for different contraceptive medications that warned against several surrogate risk factors, including low bone calcium, hyperkalemia, and high levels of ethinyl estradiol (5). The researchers' concern was partly due to what they judged to be outdated or recycled warnings from earlier package inserts and warnings that were not linked to trustworthy evidence. In 2011, Kaunitz and Grimes doubted the BBW against possible fracture risk for depot medroxyprogesterone acetate because of its reliance on a surrogate outcome (bone mineral density) (6). Black box warnings have also been questioned because of the opaque nature of the risk–benefit judgments leading to their formulation. In 2014, dueling articles attacked and defended the BBW of the risk for suicide with antidepressants. The authors were concerned about the safety of depressed patients, who would be threatened either by the medication or by its avoidance (and resulting lack of treatment) (7, 8). A clinician–patient dyad engaged in shared decision making and acting on the BBW requires 3 elements: an estimate of effect that allows tradeoffs; a level of certainty in the evidence; and guidance on how to implement or act on the BBW in a way consistent with the patient's clinical context, values, and preferences.