Off-label prescribing, also known as unapproved use, is the physician practice of prescribing a drug or medical device for a purpose different from one of the indications for which the product is approved by the Food and Drug Administration (FDA). The practice is widespread. Although there are no accurate data, estimates run as high as 60% of all drug prescriptions written in the United States in a given year, including a large proportion of chemotherapy and pediatric prescribing.

The practice raises a number of legal and ethical issues. Is off-label prescribing a form of human experimentation, triggering the safeguards established to protect human subjects? When is it appropriate in that it conforms to the standard of care? Could a failure to prescribe off-label leave the physician vulnerable to a malpractice suit? Does the physician have a duty to inform the patient that the product is being prescribed off-label? In prescribing a drug for an unapproved use, does the physician act as a "learned intermediary," thereby relieving the drug manufacturer of liability for resulting patient harm? How does the FDA regulate off-label prescribing? Can a manufacturer promote a product for an off-label use?

When Is Off-Label Prescribing Appropriate?
From a legal and ethical standpoint, off-label use represents a delicate balance between the regulatory objective of protecting patients from unsafe or ineffective drugs and medical devices on the one hand, and, on the other hand, the prerogative of physicians to use their professional judgment in treating patients.

In the world of law, where matters are rarely clear-cut, there are a few certainties about off-label use. First, it does not violate FDA law. The agency itself acknowledges this, its Center for Drug Evaluation and Research, stating, for example: "Neither the FDA nor the Federal government regulate the practice of medicine. Any approved product may be used by a licensed practitioner for uses other than those stated in the product label." In Buckman Co. v. Plaintiffs' Legal Comm., 531 U.S. 341, 350 (2001), the U.S. Supreme Court observed that "'off-label' usage of medical devices (use of a device for some other purpose than that for which it has been approved by the FDA) is an accepted and necessary corollary of the FDA's mission to regulate in this area without directly interfering with the practice of medicine."

Second, it follows from the fact that off-label use is not illegal that using a product off-label does not by itself constitute malpractice (Femrite v. Abbott Northwestern Hospital, 568 N.W.2d 535, 542 (Minn. Ct. App. 1997) (pedicle screws). On the contrary, as a Tennessee appellate court observed in Richardson v. Miller, 44 S.W.3d 1, 13, n.11 (Tenn. Ct. App. 2000): "Because the pace of medical discovery runs ahead of the FDA's regulatory machinery, the off-label use of some drugs is frequently considered to be 'state-of-the-art' treatment. "In some circumstances, an off-label use of a particular drug or device may even define the standard of care."

However, this does not mean that a physician should feel free to use a product off-label in the same way that he or she might employ the product for one of its approved indications. In the first place, as the court in Richardson v. Miller, cited earlier, pointed out, the physician lacks the information on use, dosage, and route of administration that is provided in product labeling for approved indications. Furthermore, the safety and efficacy of the unapproved use may not have been established by adequate and well-controlled clinical trials. Clearly, anecdotal data does not take the place of such investigations, and the fact that the product has been proven to be safe and efficacious for one use does not mean that it is safe and efficacious for any other; if that were the case, there would be no need for label warnings cautioning against use by certain persons or in conjunction with certain other products. Even when an off-label use has been the subject of clinical investigation, the fact that it has not been approved by the FDA may mean that the manufacturer has not sought FDA approval for the indication, in which case there has been no independent agency review of the study findings, or that the agency has rejected the findings.

In short, off-label prescribing bears some inherent liability risk for the physician. In Richardson v. Miller, for example, the court held that the fact that a drug use was off-label could be introduced as evidence that the prescribing physician deviated from the standard of care. The physician in that case employed terbutaline sulphate, administered by infusion pump, to arrest a woman's premature labor, which began when she was 35 weeks pregnant. The woman sued the physician after she suffered a heart attack. The appellate court reversed the trial court's exclusion of the fact that the use of the drug for tocolysis (the delaying or inhibition of labor during the birth process) was not approved by the FDA. (Two additional aspects of the case are worth noting. First, not only was the use of the drug for tocolysis not approved by the FDA; the label for the drug, which is indicated for prevention and relief of brochospasm, contained an explicit warning against tocolysis use. Second, the court made it clear that the fact that the drug was used off-label was not sufficient evidence by itself to prove that the physician was negligent.)

Given the risk of liability for using a product for an unapproved purpose, physicians should do so only when they are convinced that the unapproved status of the use is outweighed by the potential benefit to the patient. Often this will be obvious - for example, when there are no other available products, such as is the case with many drugs prescribed for children, which until recently were rarely tested in that population.

Is Off-Label Use "Human Experimentation" Such That It Triggers Human Subjects Protections?
In some cases, off-label use might seem to be human experimentation: The physician has no hard evidence that the product is safe and efficacious for the intended use but, based on an educated guess, decides to try it and see. The question is whether this triggers the elaborate safeguards designed to protect human subjects in medical investigations, including the need to obtain the approval of an institutional review board (IRB).

In a 1998 article in the Food, Drug Law Journal, two lawyers who represented manufacturers of pedicle screws in products liability litigation argue that it does not. Citing the 1978 Belmont Report and other classic formulations of medical ethics, Beck and Azari point out that there is a distinction between research and therapy and make it sound like the distinction is clear. According to Beck and Azari, the key is the physician's intent: If the intent is primarily to benefit the patient, the intervention is therapy. If the intent is solely to test an hypothesis and obtain generalizable knowledge, the intervention is an experiment. Yet in employing a product for an unapproved use, the physician often has both objectives in mind: She hopes to benefit her patient, but she also hopes to find out whether the intervention can help similar patients. The distinction becomes even more blurred if the physician has a strong financial interest in the success of the intervention.

Nevertheless, the courts seem to agree with Beck and Azari, at least to the extent of saying that the mere fact that a product is used off-label does not mean that the patient is an experimental subject. In a footnote, the court in Richardson v. Miller stated that "the off-label use of a drug or device by a physician seeking an optimal treatment for his or her patient is not necessarily considered to be research or an investigational or experimental treatment when the use is customarily followed by physicians" (44 S.W.3d at 13, n.9). In the Femrite case cited earlier, the court went so far as to hold that the off-label use of surgical screws was not experimentation even when the use in question was the subject of ongoing clinical investigations sponsored by the manufacturer (568 N.W.2d at 540-541). In that case, the court specifically rejected the plaintiffs' contention that physicians "could [not] consider 'off-label' usage to be a standard of care in some cases and simultaneously investigational in others" (Id. at 541).

How Does FDA Regulate Off-Label Use?
The FDA admits that it lacks the authority to regulate the practice of medicine. Accordingly, the agency does not purport to control the practice of off-label prescribing by physicians. But the FDA has taken action against manufacturers of products used for unapproved purposes. The most notable instance was when the agency in 1997 withdrew fenfluramine and phentermine from the market because of safety hazards associated with their off-label combination use for weight loss. Beck and Azari state that only fenfluramine was responsible for the adverse events, implying the off-label combination use was not the problem, but they ignore the fact that when the two drugs began being used in tandem, prescriptions for fenfluramine jumped from 100,000 prescriptions per year to over 5.1 million, thereby exacerbating the harm to patients.

The focus of FDA regulatory efforts toward off-label use has been on product promotion. Originally, the agency took the view that the manufacturer could not promote a product in any way for an unapproved use. The passage of the Food and Drug Administration Modernization Act (FDAMA) in 1997 relaxed these restrictions somewhat. The law permitted a manufacturer to distribute reports of clinical studies on unapproved uses of a drug or device if the manufacturer:
  1. had filed an application with the FDA to approve the new indication;
  2. the manufacturer itself had sponsored the trials;
  3. the manufacturer had filed with the FDA 60 days in advance the reports it intended to distribute; and
  4. the reports bore a disclaimer that the new use had not been approved by the FDA (21U.S.C. §260aaa).

After two lawsuits were filed by the Washington Legal Foundation and some of these provisions were declared to be an unconstitutional infringement of commercial speech [Washington Legal Foundation v. Henney, 56 F. Supp.2d 81 (D.D.C. 1999); Washington Legal Foundation v. Friedman, 36 F. Supp.2d 16 (D.D.C. 1999); Washington Legal Foundation v. Friedman, 13 F. Supp.2d 51 (D.D.C. 1998)], the agency revised its interpretation of the law: A manufacturer who complies with the provisions of FDAMA obtains a "safe harbor," meaning that the agency will not take regulatory action against it and will not regard the reports it disseminates as evidence that it violated the law. If a manufacturer does not comply with FDAMA, the agency will not treat its failure to comply as a separate illegal act. But the agency's position does not specify the circumstances in which it will take action against a manufacturer for disseminating information about off-label uses, so the limits on off-label promotion remain unclear.

Should Off-Label Promotion Be Allowed
Should manufacturers be permitted to promote products for unapproved uses? A good summary of the arguments pro and con appears in an article by O'Reilly and Amal4 See if they change your mind:

Pro-Off-Label Promotion
  • Off-label promotion makes more information available to physicians, enabling them to make better treatment recommendations for patients.
  • Off-label promotion allows manufacturers to avoid or postpone the costs of obtaining FDA approval so that they can make products available more quickly and invest more in research and development.
  • Off-label promotion especially benefits patients with orphan diseases, who often must rely on off-label uses for treatment.

Anti-Off-Label Promotion
  • Off-label promotion undercuts the FDA's ability to ensure safety and efficacy.
  • Off-label promotion removes incentives for manufacturers to conduct studies on safety and efficacy.
  • Off-label promotion encourages manufacturers to seek FDA approval only for the narrowest, most-easy-to-support indications.

Instead of curtailing manufacturers' ability to promote their products for off-label uses, a different approach would be to require them to conduct safety and efficacy testing for major off-label uses of their products. But how would manufacturers ascertain whether their products are being prescribed for approved or unapproved uses? With the exception of drugs prescribed for physician-assisted suicide in Oregon, there is no requirement that physicians indicate the intended use of a prescription on the prescription itself.

Some information is obtainable from adverse event reports that physicians file with manufacturers and the FDA. Drug compendia compile off-label uses based on information from expert consultants. Additional data could be obtained from insurers, or from confidential surveys of physicians or patients. If the FDA could get this information, Congress could empower it to require manufacturers to conduct safety and efficacy trials on products whose off-label sales exceeded a certain threshold. Noncompliance could be punished by civil penalties that would strip the manufacturers of their profit from the unapproved uses.

Is a Physician Required to Inform the Patient that a Drug is Being Prescribed Off-Label?
The doctrine of informed consent obliges physicians to provide patients with material information about proposed treatments, alternatives, and the potential risks and expected benefits of each. Is the off-label status of a potential use of a product material information?

Different states have adopted different standards for determining whether information is material. The prevailing standard is whether a reasonable physician would provide the information, with the jury (or judge, if the case is being heard without a jury) required to decide the matter based on conflicting testimony by the parties' medical experts. An increasing number of states have adopted a "reasonable patient" standard, where the test is whether the jury (or judge) conclude, without the benefit of expert testimony (since there is no such thing as "an expert reasonable patient"), that a reasonable patient would have regarded the information as important. Finally, a few states have adopted an "actual patient standard," which is similar to the reasonable patient standard but gives more deference to the plaintiff's claim that the information should have been provided.

As noted earlier, off-label status does say something about potential risks and benefits, or more precisely, about how much we know about them, since it indicates that either the use in question has not been the subject of adequate and well-controlled clinical investigations, or that the FDA has not reviewed such studies. Certainly under the reasonable and actual patient standards, and perhaps even under the reasonable physician standard, this might be information that would be important for the patient to have when deciding whether or not to pursue a particular course of treatment.

Yet the courts by and large have ruled that physicians do not have to reveal that a proposed use is off-label during the informed consent process. A typical case is Alvarez v. Smith, 714 So.2d 652 (Fla. Ct. App. 1998). This case involved pedicle screws, a particularly notorious episode in the history of off-label use in which the manufacturer, having failed to secure FDA approval for the use of its device to help fuse the spine, changed the description to "nested bone plate" and obtained agency approval to use it to secure long and flat bones, such as in the arms and legs. This allowed the product to remain on the market, enabling surgeons to use it off-label in the spine [Femrite v. Abbott Northwestern Hospital, 568 N.W.2d 535 (Minn. Ct. App. 1997)]. Over two thousand patients complained that the screws injured them, leading to massive litigation, including a Supreme Court case, Buckman Co. v. Plaintiffs' Legal Comm., in which the Court ruled that federal law barred plaintiffs' claim that the manufacturer had defrauded the FDA.

In their effort to find a way to recover damages, some of the pedicle screw plaintiffs alleged that, by failing to disclose the off-label status of the screws when used in the spine, their surgeons had proceeded without obtaining their informed consent to the surgery. Alvarez and other cases rejected this contention, often relying on an argument put forth by Beck and Azari in the article cited in the first paragraph of this Cyberounds®: "The bare fact of off-label use of a device or drug carries with it no medical information, either express or implied. While patients might have some assurance that uses actually appearing on labeling are safe and effective, they cannot imply from a label's silence that a particular use recommended by their physician is unsafe, risky, novel, or untried" (Beck and Azari 1998, at 89). This may be true, but off-label status does mean that the use lacks the same assurances of safety and efficacy as an approved indication. Arguably this is something that the patient ought to be told. The point, however, has been lost on the courts.

The unwillingness of the courts to extend the informed consent requirement to include giving patients information about the off-label status of a proposed use may stem from the fact that they often are dealing with cases in which the plaintiffs are scrambling for some basis on which to sue physicians when other approaches have failed. It also may reflect recognition of the fact that, for some reason, physicians often neglect to obtain the patient's informed consent to the use of specific drugs. At least in my experience and that of others I have informally surveyed, physicians rarely explain why they are recommending a specific drug, mention alternatives, or provide information about relative risks and benefits.

How Does Off-Label Use Affect the "Learned Intermediary" Doctrine?
One of the reasons plaintiffs attempt to sue physicians for failing to obtain their informed consent to an off-label use is that they have been frustrated in their attempt to sue the product manufacturer. This can happen because, as in the pedicle screw cases, their claim against the manufacturer is precluded by federal law. It also can result from the learned intermediary doctrine, which says that, in the case of prescription drugs and devices, the physician rather than the manufacturer is responsible for providing product information to patients. In order for the doctrine to apply, however, the physician community must be aware of the information.

In Proctor v. Davis, 682 N.E.2d 1203 (Ill. Ct. App. 1997), the court determined that the doctrine did not insulate the manufacturer from liability. The case involved the accidental injection of a corticosteroid, Depo-Medro®, directly into a patient's eye, destroying the eye. The drug was approved for certain uses, but not for periocular injection. The plaintiff proved that the manufacturer, Upjohn, was aware of dangers from periocular use and not only failed to alert physicians, but actively promoted the drug for periocular use.

Insurance Coverage for Off-Label Use
Since off-label use is often regarded as appropriate or indeed state-of-the-art treatment, an important question is whether health insurance plans that provide coverage for drugs will cover them for off-label uses. Plans often decline to do so on the ground that the use, being unapproved, is "experimental," and therefore excluded by their standard policy exclusion for experimental interventions. A number of states have passed laws requiring insurers to pay for off-label uses of approved drugs.6 However, these laws do not apply to pharmacy benefits provided by employer self-insured health plans. Neither Medicare nor Medicaid specifically exclude coverage of off-label uses, but the recent investigations of Pfizer for promoting off-label uses for an anti-epileptic may trigger closer scrutiny of reimbursement claims for off-label uses, especially when the Medicare prescription drug benefit takes effect in 2006.