Healthy Skepticism Library item: 16289
Warning: This library includes all items relevant to health product marketing that we are aware of regardless of quality. Often we do not agree with all or part of the contents.
Publication type: news
Bankhead C
Doctors Often Unaware of Drugs' Approval Status
MedPage Today 2009 Aug 21
http://www.medpagetoday.com/ProductAlert/Prescriptions/15639
Full text:
Physicians frequently prescribe medications for off-label uses, mistakenly believing there’s FDA approval for those indications, according to a survey of clinicians.
More than 40% of respondents believed one or more drugs had FDA approval for indications that have uncertain or no supporting evidence, said G. Caleb Alexander, MD, of the University of Chicago, and colleagues.
Overall, the survey participants knew the FDA approval-status for about half of 14 drug-indication pairs, such as gabapentin (Neurontin) for diabetic neuropathy (off label, some supporting evidence), they reported online in Pharmacoepidemiology and Drug Safety.
Physicians’ knowledge of the approval status for drug-indication pairs increased modestly when limited to drugs the physicians had prescribed within the previous 12 months.
“These results indicate an urgent need for effective methods of disseminating information to physicians about the level of evidence supporting off-label drug uses, with specific attention to common off-label uses known to be ineffective or to carry unacceptable risk of harm,” the authors concluded.
Off-label prescribing is common in clinical practice and can be a source of innovation. In some settings, off-label prescribing may represent the standard of care, the authors noted.
However, studies have shown that off-label prescribing frequently occurs when there is uncertain or no evidence supporting the efficacy of a drug for an indication.
FDA approval of a drug for a specific indication reflects a threshold of supporting evidence, the authors continued. On the other hand, lack of approval for specific indications may occur for reasons that do not relate to supporting evidence, such as the safety of a drug for an unapproved use.
To examine physicians’ knowledge of evidence supporting common drug-indication pairs, the authors conducted a random-sample national survey of primary care physicians and psychiatrists identified from the AMA Masterfile.
Physicians selected for the study received a three-page, 63-item questionnaire based on 14 common drug-indication pairs. The pairs varied with respect to FDA-approval status and level of supporting evidence. Respondents indicated whether each pair had FDA approval.
Psychiatrists and primary care physicians were given lists of drug-indication pairs that differed except for six pairs: valproic acid for bipolar disorder, mania; escitalopram (Lexapro) for panic disorder; gabapentin for diabetic neuropathy; trazodone for insomnia; venlafaxine (Effexor) for adjustment disorder; and quetiapine (Seroquel) for dementia with agitation.
The authors received 457 usable surveys, 54% from psychiatrists and 46% from primary care physicians. On average, primary care physicians had prescribed 10 of the drugs at least once in the previous year for any indication, and psychiatrists had prescribed 11.
Respondents knew the correct FDA-approval status of 55% (mean) to 57% (median) of the drug-indication pairs.
When the pairs were limited to drugs the respondents had prescribed in the past year, the mean accuracy increased to 60% and the median to 63%.
Overall, psychiatrists demonstrated better knowledge of FDA approval status (66% mean, 71% median) than did primary care physicians (42% mean, 38% median).
The authors found a significant association between a physician’s belief that an indication had FDA approval and greater evidence supporting efficacy for the indication (P<0.001).
However, 41% of physicians believed that at least one drug-indication pair with uncertain or no supporting evidence had FDA approval, such as quetiapine for dementia with agitation.
The researchers acknowledged limitations of their study, including the limited number of drug-indication pairs used, the relatively low response rate, and the single method they used to determine the level of evidence for each pair.
They also said it was possible that they were unable to assess factors other than level of evidence that might lead a physician to believe a drug was approved for an indication.
