Off-label prescribing common, often not backed by data

May 9 2006

A study of office-based physicians in the United States suggests that about one-fifth of medications are prescribed to treat conditions for which they are not approved by the U.S. Food and Drug Administration (FDA), and that nearly three-fourths of those uses lack strong scientific support, according to an article in the May 8 issue of Archives of Internal Medicine.

The FDA evaluates new drugs entering the market but does not dictate physicians' prescribing practices, according to background information in the article. Therefore, physicians can legally prescribe approved medications for any diagnosis, even if that particular diagnosis is not cited on the drug's label or application for FDA approval. Such off-label prescribing allows physicians the freedom to innovate in clinical practice, but also raises questions regarding patient safety and health care costs, since off-label uses do not receive the same level of scientific scrutiny as approved uses.

David C. Radley, M.P.H., Dartmouth Medical School, Hanover, N.H., and colleagues examined data from a national survey of office-based physicians. In each quarter of 2001, about 3,500 randomly selected physicians reported on all patient interactions during two consecutive workdays. Information about diagnoses and prescriptions was entered into an index, from which the researchers retrieved information about the 100 most commonly prescribed drugs and 60 additional drugs randomly selected from among other common medications. They compared each prescription to the indications listed on the drug's package insert and categorized them as approved or off-label. Off-label prescriptions were further evaluated and classified as having strong scientific support if evidence from drug trials or clinical settings suggested the medication was effective for the condition it was prescribed to treat, or as having limited to no scientific support if such evidence did not exist.

Eighty-five percent of the 725 million total prescriptions given for the 500 drugs in 2001 were either for FDA-approved indications or off-label uses with strong scientific support. However, an estimated 150 million (21 percent) of the prescriptions were for off-label uses and most (73 percent, or 15 percent of total prescriptions) of those lacked scientific support. Cardiac medications (46 percent), anticonvulsants (46 percent) and medications used to treat asthma (42 percent) were the most likely to be prescribed off-label. Psychiatric drugs were highly likely to be prescribed off-label with limited or no scientific support (96 percent vs. 4 percent strong support), as were allergy medications (89 percent vs. 11 percent strong support).

Some off-label uses followed logically from FDA-approved indications, the authors write, such as prescribing asthma medications for other lung diseases. However, other off-label prescriptions were for conditions dramatically different from those for which the drug was approved to treat, such as the use of an anticonvulsant drug for chronic pain. "Differentiating off-label situations that are clinically reasonable from those that may be of concern is an essential first step," the authors write.

"The ability to prescribe medicines off label brings greater latitude to turn scientific knowledge into innovative clinical practice," they conclude. "Although attention should be given to the situations where evidence-based off-label use is clinically beneficial, policy makers must begin to consider strategies for mandatory postapproval surveillance that focus on curtailing underevaluated off-label practices that jeopardize patient safety or represent economically wasteful prescribing practices."