Monday, July 20, 2009

“Chilling Effects” of FDA “Black Box” Warnings

As noted in our prior posting [7/12/09], the FDA is currently permitting propoxyphene-containing analgesics to remain on the market in the U.S., but is requiring “black box” warnings regarding overdose risks. Earlier, the FDA had ruled to require stronger warnings of internal bleeding and liver damage with OTC pain relievers (NSAIDs and acetaminophen). However, according to some evidence, boxed warnings can have negative effects on prescribing going beyond what might have been anticipated.

Unexpected Consequences

Clearly, the FDA has a medicolegal duty to warn healthcare providers and the public of special or newfound risks associated with products under its purview. “Black box” warnings, the strongest notice of caution, appear in product labeling and are, of necessity, brief and concise. Conscientious drug prescribers heed such warnings, at the least for liability reasons.

Unfortunately, the FDA does not provide easy access to full disclosure of reasons behind their warnings; explaining, for example, which patients under what circumstance might be at greatest risk for the stated adverse effects. Consequently, there can be a pervasive “chilling effect,” producing changes in prescribing behavior that may have unexpected negative consequences.

Harmful Fallout of Antidepressant Warnings

For example, in 2003 the FDA issued an advisory about risks of suicidal thinking and behavior (suicidality) in pediatric patients taking antidepressants, although a boxed warning was not implemented until 2005. The warning was extended in 2007 to include young adults and all antidepressant drugs [see FDA Announcement]. The warning was issued despite the absence of a single suicide death in clinical trials examining antidepressants. (Antidepressants also are often prescribed for persons of all ages to treat certain pain conditions as well as concurrent depression, which is common as noted in earlier post [7/19/09].)

A recently reported extensive investigation by Libby and colleagues [2009] at the University of Colorado, Denver, found that, starting in 2003, there were sudden decreases in newly diagnosed cases of depression and reductions in prescribing of antidepressant drugs in all age groups in the U.S. The researchers assert that unintended negative effects of the warnings were substantial, persistent, and widespread. Economic costs to society of untreated depression increased considerably. Most likely due to untreated depression, there was a “significant increase in pediatric suicide deaths for the first time in more than a decade.” Among youths, aged 10 to 19, there was an upward trend in suicide rates. (For another, very extensive review, see McCain JA, P&T, 2009[Jul];34[7]:355-367.)

Apparently, the FDA has said it would revisit this policy if there was evidence of unintended consequences [Libby et al. 2009]. There is the question, however, of when the FDA will examine the issues and why they did not consider the possibility of such chilling effects and consequences in advance of their warning.

Methadone Misdirections

As another example, in late 2006 the FDA issued an advisory and revised boxed warning on methadone, including notice of cardiac toxicities (QT prolongation and Torsade de Pointes) with the drug [see FDA Advisory]. However, the warnings did not fully disclose evidence in support of this contention and, in fact, the everyday clinical significance of methadone’s alleged cardiotoxic effects is still quite uncertain [for example, see Cruciani 2008]. There have been no deaths reported in the literature directly and solely attributable to methadone-induced arrhythmia.

Curiously, the FDA failed to include in its 2006 revisions any special alert about the need for reduced dosing when starting opioid-naïve patients on oral methadone analgesia (2.5 to 10 mg every 8-12 hours, 30 mg/day maximum). Prior prescribing instructions allowed methadone induction doses of up to 80 mg/day, which could incur hazardous overdose in some patients. [Also see, blog entry from RW Newman, MD.]

At present, it is unknown how many severe adverse events may have occurred due to unintentional methadone overdose when prescribers were unaware of the dosing changes buried within the revised product instructions. Nor is it known how many patients were denied methadone analgesia entirely due to vague concerns about cardiac arrhythmia expressed in the black box warning.

Warnings Are Not Education

In sum, while the FDA relies on “black box” warnings as a policy-enforcement tool, the nature and extent of evidence driving such decisions is often not revealed. In any event, warnings are not the same as providing healthcare providers and the public with the education they need for fully informed and rational clinical decision making. Hopefully, the FDA will have an appreciation of this as they contemplate warnings for propoxyphene drugs and other opioid analgesics coming under their proposed REMS (Risk Evaluation and Mitigation Strategy) initiatives.

Your comments are welcomed.

Reference [noted but not linked above]: Libby AM, Orton HD, Valuck RJ. Persisting decline in depression treatment after FDA warnings. Arch Gen Psychiatry. 2009;66(6):633-639.