The Corpus Callosum

Drug Safety in Perspective

on this blog, I’ve ranted about the risks that our government, and our
corporate citizens, e.g. pharmaceutical companies, expose us to on a
daily basis.  

Perhaps it would be good to put some of those risks in perspective.
 That is, to compare the risks of various medications to
others risks that we take on a routine basis.  In the May/June
issue of the journal, Health Affairs, there is an article on the
subject.  The full thing is behind a pay wall, but we’ll get
to the heart of the matter anyway.

More Dangerous, Your Aspirin Or Your Car? Thinking Rationally About
Drug Risks (And Benefits)

Joshua T. Cohen and Peter J. Neumann
Health Affairs, 26, no. 3 (2007): 636-646
doi: 10.1377/hlthaff.26.3.636
© 2007 by Project HOPE

We compare mortality
risks of several common drugs with risks related to work,
transportation, and recreation. Comparing risks can provide a more
intuitive sense of the magnitude of drug risks than stand-alone
estimates can, to help inform policy discussions. The drug risks we
quantify generally exceed the magnitude of risks for other domains
(although aspirin and cars are similarly “risky” under the definition
of risk used here). Nonetheless, these comparisons underscore a crucial
point: that risks should not be evaluated without considering attendant
benefits. We discuss the need for the Food and Drug Administration to
compare risks and benefits quantitatively, consistently, and explicitly.

The article was picked up in what seems an unlikely place, Channel 8
News, WOOD, out of Grand Rapids, Michigan:

Safety: There’s No Magic Pill, Study Says

By Amanda Gardner, HealthDay Reporter

TUESDAY, May 8 (HealthDay News) — If you think fighting red-hot blazes
in collapsing buildings is risky, think again.

Swallowing one aspirin each day for a year poses the same risk of
death, new research suggests.

In an analysis appearing in the May/June issue of Health Affairs,
researchers compared half a dozen risky drugs to various occupational,
recreational and transportation activities, with sometimes surprising

…”It’s nice to have a rigorous evaluation of not just one drug vs.
the other, but some of the more risky pharmaceuticals compared to
things that are widely done in day-to-day life,” added Dr. A. Mark
Fendrick, co-director of the University of Michigan Center for Value
Based Insurance Design, and professor at the University of Michigan
Medical School and University of Michigan School of Public Health in
Ann Arbor…

I have found in clinical practice that it is very difficult to get
people to assess risk objectively.  Frankly, I think studies
such as this might help a little, but probably not very much.

These were then compared to various non-drug activities,
with these findings:

  • Taking aspirin carries a risk of death similar to driving
    a car or
    working as a firefighter. “Motor vehicle risk I’ve always known to be a
    very big risk,” Cohen said. “The risk associated with aspirin is just
    as big.”
  • The risk of dying from Vioxx or Tysabri is about the same
    as dying in a car, working as a truck driver or rock climbing.
  • The odds of dying from aspirin, clozapine, Tysabri and
    Vioxx are
    about equal to the corresponding risk for driving in a passenger car.
  • Out of all modes of transportation, only motorcycles pose
    a greater
    risk (450 per 100,000 person-years) than all drugs, even Vioxx, which
    had a 76 per 100,000 person-years fatality rate.
  • Aspirin, clozapine, Tysabri and Vioxx all had risks equal
    to or
    greater than the annual mortality risks for firefighters and law
    enforcement workers (about 11 per 100,000 person-years). The most risky
    occupation — tree fallers — was much higher than any drug, at 360 per
    100,000 person-years.
  • Rock climbing (36 deaths per 100,000 person-years) had a
    rate about equivalent to clozapine, while climbing in the Himalayas
    (13,000 per 100,000 person-years) had a higher risk of death than any
    of the drugs studied.
  • And, just as a curiosity, the researchers found that
    can potentially face annual fatality risks that are 40 times as great
    as passenger car travelers, while taxi drivers face three times the
    death risk of firefighters.

Cohen commented that this kind of information might help people make
drug-related risk decisions more clear-headedly.  

Of course, physicians should not try to be in the business of making
risk-related decisions for patients; rather, they should do what they
can to help patients make informed decisions.  It is a big
challenge, to distill a complex subject into an understandable and
brief but reasonably complete discussion, all the while the waiting
room, outside the office, is getting more crowded.

I think is is worthwhile for people to reflect upon their own personal
styles for assessing risk, both in the context of medical
decision-making, and in other contexts.

For example, it seems that some people weigh risks more heavily if the
adverse outcome were to result directly from a choice they made, as
opposed to being a seemingly random event.  This can lead
them, paradoxically, to take a greater risk by forgoing
treatment, rather than taking a smaller risk with
treatment, but feeling more responsible if there is a bad outcome.
 Doctors seem to have a hard time understanding this.
 It is comparable to people voting against their own
interests, in a way.  Hard to understand, but it is
their vote to cast.


  1. #1 Aubrey Blumsohn
    June 2, 2007

    A good piece but misses the point somewhat. The aim of a car is to get from A to B – it does that job at some risk. The risk of driving (A) versus staying at home (B) is close to infinite but that does no make the car a defective product. If a drug say Avandia is supposed to decrease cardiovascular mortality in diabetes, but instead increases it, the product is defective. So it’s more than comparing risks.

    In any event, the problems currently facing this industry are not to do with risk. They have to do with the fact that risk is misrepresented through scientific malprocedure and lack of transparency, so that the risk cannot be evaluated by the prescriber or the user in order to make that informed decision.


New comments have been disabled.