Drugs and Serious Adverse Events

A “serious adverse event” is drug-speak for something bad that happened when a person was taking a medication. That bad thing can range from a hospitalization all the way up to death.

The U.S. Food and Drug Administration has a voluntary reporting system setup called MedWatch that accepts voluntary reports of medications related to a serious adverse event. I say “related to” rather than “caused” because the MedWatch reporting database makes no claims (and cannot make any claims) that a serious adverse event is directly related to one and only one specific medication.

Deaths in this system can also be reported for people who committed suicide using a particular medication. Although this may not be commonly done, we have no way of knowing because the system doesn’t differentiate between an intentional death related to a medication and an unintentional death.

A study in the Archives of Internal Medicine was recently published that looked at medications related to serious adverse events over a 7-year period (1998-2005) and Furious Seasons has some commentary on the article. The table on page 4, which lists deaths related to a particular medication, has some psychiatric medications on it.

Common drugs such as insulin (used to treat diabetics) and estrogen (used in women for a variety of diseases) rank very highly for being related to a disability or other serious outcomes in the patient. Estrogen therapies alone accounted for over 11,000 disabled Americans in the 7-year period. And yet it remains a valuable treatment option for many women today.

Acetaminophen — you know, plain old Tylenol — accounted for being related to 1,393 deaths in that period. Methadone, the most common treatment for people trying to get off of hard drugs like cocaine or heroin, resulted in a correlation of 1,258 deaths.

But clozapine (Clozaril) is also on that list, related to 3,277 deaths. So is a more common antidepressant medication, Paxil (related to 850 deaths). Two other antipsychotic medications also make the top 15 list.

I don’t find it surprising these medications are on the list. Mental health concerns commonly account for somewhere around 25% of all health concerns, which means that, statistically, we would expect approximately 4 drugs on a top 15 list to be those that treat mental health concerns.

What’s missing from this list? Well, context would be nice.

For instance, we’re given gross totals, but not representative ratios for the population of people actually being prescribed these medications. It’s far more interesting to put out a large number, but less interesting to put that number into context of what is the percentage of these serious adverse events versus all people prescribed and taking the medication. Since the researchers only looked at MedWatch for their data, they didn’t have prescription data to compare these numbers to. So while the numbers say something, they are as much noise as anything else because of the lack of context.

Of course, it’s easy to Monday-morning quarterback, but this research would’ve been more valuable with such percentages given. For instance, it’s far more serious if a drug’s prescriptions result in a 5% serious adverse event reporting rate versus a drug that has only a 0.001% serious adverse event reporting rate. But if tens of millions more people are taking the 0.001% rate drug, that drug’s total number may be significantly higher than the 5% drug.

Drugs like Acetaminophen will always be tops on this sort of list because they are so commonplace and widely used by people everyday. The sheer numbers of people who take painkillers mean that when something bad happens, it’s going to be on a list of serious adverse events.

So without that valuable context, I just don’t know what to make of a list like this. I can give reasons for every one of those 4 drugs making this list (e.g., Paxil was one of the most commonly prescribe antidepressants during this period but without its competitors numbers, it’s hard to say whether Paxil is ordinary in its numbers or out-of-the-ordinary).

So the upshot — we need more research in this area, but research that gives us context and perspective.