Paul H. Axelsen, MD
October 27, 2021
Paul H. Axelsen, MD
Almost all physicians write prescriptions, and each prescription requires a physician to assess the risks and benefits of the drug for their patient. If an adverse drug reaction occurs, physicians may be called on to defend their risk-benefit assessment in court.
The assessment of risk is complicated when there is a black box warning that describes potentially serious and life-threatening adverse reactions associated with a drug. Some of our most commonly prescribed drugs have black box warnings, and drugs that were initially approved by the FDA without black box warnings may have them added years later.
One serious problem with black box warnings is that there are no reliable mechanisms for making sure that physicians are aware of them. The warnings are typically not seen by physicians as printed product labels, just as physicians often don’t see the pills and capsules that they prescribe. Pharmacists who receive packaged drugs from manufacturers may be the only ones to see an actual printed black box warning, but even those pharmacists have little reason to read each label and note changes when handling many bulk packages.
This problem is aggravated by misperceptions that many physicians have about black box warnings and the increasingly intense scrutiny given to them by mass media and the courts. Lawyers can use black box warnings to make a drug look dangerous, even when it’s not, and to make physicians look reckless when prescribing it. Therefore, it is important for physicians to understand what black box warnings are, what they are not, the problems they cause, and how to minimize these problems.
What Is a ‘Boxed Warning’?
The marketing and sale of drugs in the United States requires approval by the Food and Drug Administration (FDA). Approval requires manufacturers to prepare a document containing “Full Prescribing Information” for the drug and to include a printed copy in every package of the drug that is sold. This document is commonly called a “package insert,” but the FDA designates this document as the manufacturer’s product “label.”
In 1979, the FDA began requiring some labels to include “boxed warnings” within thick, black rectangular borders which have come to be known as black box warnings. Boxed warnings are usually placed at the beginning of a label. They may be added to the label of a previously approved drug already on the market or included in the product label when first approved and marketed.
The requirement for a boxed warning most often arises when a signal appears during review of postmarketing surveillance data suggesting a possible and plausible association between a drug and an adverse reaction. Warnings may also be initiated in response to petitions from public interest groups, or upon the discovery of serious toxicity in animals. Regardless of their origin, the intent of a boxed warning is to highlight information that may have important therapeutic consequences and warrants heightened awareness among physicians.
What a Boxed Warning Is Not
A boxed warning is not “issued” by the FDA; it is merely required by the FDA. Specific wording or a template may be suggested by the FDA, but product labels and boxed warnings are written and issued by the manufacturer. This distinction may seem minor, but extensive litigation has occurred over whether manufacturers have met their duty to warn consumers about possible risks when using their products, and this duty cannot be shifted to the FDA.
A boxed warning may not be added to a product label at the option of a manufacturer. In order to preserve the impact of their presence, the FDA allows a boxed warning only if it requires the warning. It should be noted that some medical information sources (eg, PDR.net) may include a “BOXED WARNING” in their drug monographs, but monographs not written by a manufacturer are not regulated by the FDA, and the text of their boxed warning does not always correspond to the boxed warning that was approved by the FDA.
A boxed warning is not an indication that revocation of FDA approval is being considered or that it is likely to be revoked. FDA approval is subject to ongoing review and may be revoked at any time, without a prior boxed warning.
A boxed warning is not the highest level of warning. The FDA may require a manufacturer to send out a “Dear Health Care Provider” (DHCP) letter when an even higher or more urgent level of warning is deemed necessary. DHCP letters are usually accompanied by revisions of the product label, but most label revisions — and even most boxed warnings — are not accompanied by DHCP letters.
A boxed warning is not a statement about causation. Most warnings describe an “association” between a drug and an adverse effect, or “increased risk,” or instances of a particular adverse effect that “have been reported” in persons taking a drug. The words in a boxed warning are carefully chosen and require careful reading; in most cases they refrain from stating that a drug actually causes an adverse effect. The postmarketing surveillance data on which most warnings are based generally cannot provide the kind of evidence required to establish causation, and an association may be nothing more than an uncommon manifestation of the disorder for which the drug has been prescribed.
A boxed warning is not a statement about the probability of an adverse reaction occurring. The requirement for a boxed warning correlates better to the new recognition of a possible association than to the probability of an association. For example, penicillin has long been known to cause fatal anaphylaxis in 1/100,000 first-time administrations, but it does not have a boxed warning. The adverse consequences described in boxed warnings are often far less frequent — so much so that most physicians will never see them.
A boxed warning does not define the standard of care. The warning is a requirement imposed on the manufacturer, not on the practice of medicine. For legal purposes, the “standard of care” for the practice of medicine is defined state by state and is typically cast in terms such as “what most physicians would do in similar circumstances.” Physicians often prescribe drugs in spite of boxed warnings, just as they often prescribe drugs for “off label” indications, always balancing risk vs benefit.
A boxed warning does not constitute a contraindication to the use of a medication. Some warnings state that a drug is contraindicated in some situations, but product labels have another mandated section for listing contraindications, and most boxed warnings have no corresponding entry in that section.
A boxed warning does not necessarily constitute current information, nor is it always updated when new or contrary information becomes available. Revisions to boxed warnings, and to product labels in general, are made only after detailed review at the FDA, and the process of deciding whether an existing boxed warning continues to be appropriate may divert limited regulatory resources from more urgent priorities. Consequently, revisions to a boxed warning may lag behind the data that justify a revision by months or years. Revisions may never occur if softening or eliminating a boxed warning is deemed to be not worth the cost by a manufacturer.
Boxed Warning Problems for Physicians
There is no reliable mechanism for manufacturers or the FDA to communicate boxed warnings directly to physicians, so it’s not clear how physicians are expected to stay informed about the issuance or revision of boxed warnings. They may first learn about new or revised warnings in the mass media, which is paying ever-increasing attention to press releases from the FDA. However, it can be difficult for the media to accurately convey the subtle and complex nature of a boxed warning in nontechnical terms.
Many physicians subscribe to various medical news alerts and attend continuing medical education (CME) programs, which often do an excellent job of highlighting new warnings, while hospitals, clinics, and pharmacies may broadcast news about boxed warnings in newsletters or other notices. But these notifications are ephemeral and may be missed by physicians who are overwhelmed by email, notices, newsletters, and CME programs.
The warnings that pop up in electronic medical records systems are often so numerous that physicians become trained to ignore them. Printed advertisements in professional journals must include mandated boxed warnings, but their visibility is waning as physicians increasingly read journals online.
Another conundrum is how to inform the public about boxed warnings.
Manufacturers are prohibited from direct-to-consumer advertising of drugs with boxed warnings, although the warnings are easily found on the internet. Some patients expect and welcome detailed information from their physicians, so it’s a good policy to always and repeatedly review this information with them, especially if they are members of an identified risk group. However, that policy may be counterproductive if it dissuades anxious patients from needed therapy despite risk-benefit considerations that strongly favor it. Boxed warnings are well known to have “spillover effects” in which the aspersions cast by a boxed warning for a relatively small subgroup of patients causes use of a drug to decline among all patients.
Compounding this conundrum is that physicians rarely have sufficient information to gauge the magnitude of a risk, given that boxed warnings are often based on information from surveillance systems that cannot accurately quantify the risk or even establish a causal relationship. The text of a boxed warning generally does not provide the information needed for evidence-based clinical practice such as a quantitative estimate of effect, information about source and trustworthiness of the evidence, and guidance on implementation. For these and other reasons, FDA policies about various boxed warnings have been the target of significant criticism.
Medication guides are one mechanism to address the challenge of informing patients about the risks of drugs they are taking. FDA-approved medication guides are available for most drugs dispensed as outpatient prescriptions, they’re written in plain language for the consumer, and they include paraphrased versions of any boxed warning. Ideally, patients review these guides with their physician or pharmacist, but the guides may be lengthy and raise questions that may not be answerable (eg, about incidence rates). Patients may decline to review this information when a drug is prescribed or dispensed, and they may discard printed copies given to them without reading.
What Can Physicians Do to Minimize Black Box Problems?
Physicians should periodically review the product labels for drugs they commonly prescribe, including drugs they’ve prescribed for a long time. Prescription renewal requests can be used as a prompt to check for changes in a patient’s condition or other medications that might place a patient in the target population of a boxed warning. Physicians can subscribe to newsletters that announce and discuss significant product label changes, including alerts directly from the FDA. Physicians may also enlist their office staff to find and review black box warnings for drugs being prescribed, noting which ones should require a conversation with any patient who has been or will be receiving this drug. They may want to make explicit mention in their encounter record that a boxed warning, medication guide, or overall risk-benefit assessment has been discussed.
Summary
The nature of boxed warnings, the means by which they are disseminated, and their role in clinical practice are all in great need of improvement. Until that occurs, boxed warnings offer some, but only very limited, help to patients and physicians who struggle to understand the risks of medications.
Leave a Reply