2007-03-08

Should You Trust Your Doctor?

I'm reading <i>Protecting America's Health</i>, by Philip J. Hilts.

He tells the story of "arguably the worst drug disaster in American
history, though it received little press coverage." In the late 1970s
and early 1980s, a class of drugs for controlling cardiac arrythmias
became available. The FDA approved these drugs for patients with
serious cardiac arrythmic conditions.

However, around 1981, doctors started using these drugs on patients
with only mild arrythmias. Over the course of the next seven years,
doctors became convinced, based on their experience and casual
observation, that such patients were benefiting from their use of
these drugs.

In 1988, over the protest of doctors, a study was carried out by the
National Heart, Lung, and Blood Institute to assess the efficacy of
the drugs in a more formal way. Doctors felt so strongly about this
that they accused the leader of the experiment of being immoral for
"withholding" the drugs from the control group.

Preliminary numbers in late 1988 showed a significant difference in
the death rates of the experimental and control groups. Although the
experimenters could see that there was a difference, they couldn't
tell which way it went because the study was carried out under a
double blind protocol. The assumption was that the experimental group
(those receiving the drugs) had the lower death rate. More patients
were enrolled in the study.

Imagine their surprise when, in 1989, the blind was broken and the
results examined. The death rate in the experimental group was 3.5
times that in the control group. Throughout the 1980s, doctors had
been poisoning their mild arrythmia patients in the belief that the
drugs lowered death risk. Actually, the study revealed that the drugs
more than tripled the death risk for patients taking them.

When the study was completed, around 400,000 patients were receiving
the drugs in question to address mild arrythmias. Based on this, the
numbers from the study indicated that the drugs were killing about
5000 patients each year. The study only covered some of the drugs.
Other similar drugs used in the same way may have been killing many
more patients.

In the aftermath, doctors said, "We're sorry," flushed their supply
of anti-arrythmia drugs, and put their patients on something else,
right? Well, not quite. That would've involved admitting a mistake.
According to a study by Columbia Univeristy over 81% of cardiologists
did not change existing patients' prescriptions. Rather, they started
prescribing other drugs for <i>new</i> patients.

2007-03-04

What happens when two helpdesks start trading e-mails?

Especially when neither tech is allowed to let a message go
unanswered? Suppose Superlative Help for Information Technologists
got infected by an e-mail virus that sent a Nigerian money laundering
note to Acme Support Services.

"Thank you for contacting Acme Support Services. How can I be of
service today?"

"Everything is hunky-dory here at Superaltive Help for Information
Technologists. What can we do for you?"

"Um... not a thing. You contacted us. What's your technical issue?"

"We don't have any technical issues that we know of. Perhaps I can
interest you in a three year contract for unlimited telephone
support? This week only, it's a low $39.95."

"Thank you for contacting Acme Support Services. If there's anything
further we can do to help you, please don't hesitate to get in touch."

"Thank you for e-mailing us. We're here to help. Keep our contact
information handy. You never know when you might want a friend on the
other end of the phone."

"Look, if you don't have a problem, stop sending me e-mail already! I
can't let a message go unanswered! Every time you send me a message,
I have to write back."

"Why do you keep responding? I can't close this ticket until you
haven't responded for at least 15 minutes. Please don't reply to this
message."

"You're not getting this are you? I CAN'T stop or I'll lose my job."

"Well, I can't stop either."

"Don't write back."

"Shut up."

"Moron."

"Idiot."

... and so forth ...