Thursday, October 29, 2009

Are Drug Labels Missing Information That Patients and Clinicians Need?


In a perspective in this week’s New England Journal of Medicine, Drs. Lisa Schwartz and Steven Woloshin, faculty at Dartmouth University, persuasively argue that the answer to question posed in the title is YES.

Schwartz and Woloshin note that when the FDA reviews a drug, expert reviewers consider a wealth of data about the benefits and side effects of the drug. They also raise questions, make judgments, and make statements about their degree of certainty that the benefits of the drug outweigh the risks. However, effectiveness and harm data that patients and their providers might want to see often do not make it to the drug label. Opinions of the expert reviewer, including their degree of uncertainty about the drug, would be of great interest but these virtually never make it to the label. This information is contained in FDA documents, but these voluminous documents are not practically accessible.


This article is a quick and fascinating read that has 3 fascinating examples that detail how important information is often absent from drug labels. Here are some selected examples from the article:

1) Zoledronic Acid (Zometa): This is used to treat hypercalcemia in cancer patients. FDA documents note that the mortality was higher in patients taking the 8 mg dose than the 4 mg dose. This was not on the label, nor in the journal article reporting the study.


2) Eszopiclone (Lunesta): this is for chronic insomnia. (You may have seen the inspiring television commercial in which a moth graciously puts frustrated insomniacs to sleep). The FDA review notes that users did not have clinically significant improvement in next day alertness or functioning. This would not be found on the drug label.


3) Romelteon (Rozerem): The FDA documents show that the improvement in time to fall asleep was small. There were no improvement on a number of measures such as awakenings or sleep quality. FDA memos suggest the approval decision was a close call. None of this is apparent from the drug label.

Schwartz and Woloshin have pioneered Prescription Drug Fact Boxes that feature a table including data about the benefits and harms of the drug. The FDA is deciding whether to use these boxes on drug labels. Hopefully, they will decide to use these boxes.

5 comments:

Alex Smith said...

Disturbing. Why doesn't this information make it on the label? I assume the pharmaceutical industry somehow supresses it. I don't know enough about how they influence the FDA. Does anyone?

Eric Widera said...

What is disturbing is the lunesta moth - I've had dreams about that thing attacking me. No wonder why patients didn't have improvement in next day alterness.

Here is a link to the example drug box on lunesta. No moth, just good infomation.

Mike Steinman said...

Schwartz and Woloshin's article only scratches the surface of the problems with medication labeling. Frankly, the package insert (which almost no one reads, anyways) is the least of the problems with how drug information is provided to patients and physicians. We have food labels that provide clear, templated information about what one is eating. Is it too much to ask to have the same for drugs, such as a simple fact sheet with key issues for consumers (e.g. major side effects for which to be vigilant, standardized dosing and frequency instructions, and so forth)?

Regarding Alex Smith's comment, drug labels are created in a process of negotiation between FDA and the drug manufacturer. There would clearly be less conflict of interest if the FDA would create the label without having to negotiate with industry about every word. Part of this would require greater funding of FDA to provide the human resources to take on such tasks. Unfortunately, previous attempts to use industry money to support FDA activities through the Prescription Drug User Fee Act (PDUFA) have resulted in their own conflicts of interest, whereby the FDA is beholden to industry for its funding. Thus, a better system needs to be devised to get the FDA the resources it needs to protect the public health while avoiding the conflicts of interest that have resulted in questionable decisions.

Namita Kansal said...

Fascinating to learn! thanks for sharing and it is rather scary to not have all the needed info on the labels... and the Industry-FDA wars/compromise are a revelation to me.

Dan Matlock said...

Thanks for highlighting this topic. It is always appalling to hear these things and I too have been attacked by the Lunest moth (hilarious!).

Another challenge is that the filed of decision sciences/medical communication hasn't fully agreed on the best way to communicate risk to patients. I think we are confident that absolute risks are better than relative risks but after that, it becomes more difficult and the research is mixed. How do you communicate rare but dangerous side effects in an unbiased way (one way or the other)? Also, Valerie Reyna's "fuzzy trace theory" of decision making argues that people make decisions based on the "gist" of the information rather than the "verbatim" numbers. Should these labels contain only the gist of the info? Who decides?

Obviously these drug companies need to be more honest but even honesty is hard in risk communication. Great topic.