Uses of Error
www.thelancet.com Vol 365 June 18, 2005 2149
At about 1 o’clock in the morning on only my third night
on call as a senior house officer I was asked to review two
patients with high serum concentrations of potassium.
One of them had a renal cell carcinoma, but the other had
simply been given too many potassium supplements. The
nurses were busy, so I volunteered to set up dextrose
insulin infusions. The nurses thanked me for my help.
About 1 h later I received a phone call—one of the two
patients had become quite clammy after his treatment for
hyperkalaemia. When I reached the ward, I noted that the
patient with a history of renal cell carcinoma was
unresponsive. A nurse checked his blood sugar (finger
prick), which was dangerously low. After reviving the
patient, I reread the label on the insulin vial and realised
that I had drawn and added to both patient’s bags
100 units (1 mL) rather than the correct 10 units (0·1 mL)
of insulin. The second patient soon became unresponsive
too and, like the first, needed a concentrated glucose
infusion. In short, instead of cutting down the workload
of the nurses I increased it, since both men subsequently
required continuous monitoring of their blood sugar for
the rest of the night.
7 years on, I still cannot forget that incident, which
might have had catastrophic consequences and could
easily have been avoided through simple cross checking
before administration of drugs.
I am a better doctor now
It had already been more than 6 years since I left clinical
medicine to become a “paper doctor”, working in health
economics and business consultancy. Early that
morning, I was out on the first few hundred metres of a
long run, in a quiet area of Berlin. I saw a man in a track
suit stumbling towards me. He suddenly stopped, tried
to gain control by holding onto a lamp post, and gradually
sank down to the ground. I approached him, and
addressed him. He must have been around his early
forties, lean and well built. I suspected that he was
suffering from either alcohol consumption, a psychiatric
disease, or an acute vascular event. It was early in the
morning and I could not smell anything, so alcohol did
not seem too likely but then again it was Berlin, where
many things happen. He did not look at all unhealthy,
but nothing could be ruled out. Having worked for some
time in psychiatry I was a bit inclined towards a
psychiatric disorder, but there was nothing in the man’s
behaviour pointing towards it. He did not answer my
greeting but made a gesture indicating that everything
was fine and I should go on my way. Again, I addressed
him, asking whether he needed some help. By that time,
he had almost got himself standing again, gesturing once
more that he was fine and there was no need for help. I
continued with my run but could not stop thinking about
him. I finally decided that he must have been on a
drinking spree the previous night and he probably felt
embarrassed because of his inability to walk home.
When coming back some time later I saw a police car
on the spot and I immediately knew what had happened.
The man was dead, and I had been the last one to see
him. The forensic examination showed that a massive
myocardial infarction was the reason for his death.
All of a sudden, everything fitted together. His face
had been ash-grey, he had been sweating profusely and
the look in his eyes was haunting. I had never seen
anybody with such panic-ridden eyes, silently screaming
for help. And his gestures? I learned afterwards from a
friend of his that the dead man had always been a very
dominant person and therefore it was not surprising
that he had acted as if he could sort himself out. His face
and his eyes told otherwise.
One of the things that medical training obliges is the
use of all one’s senses. I had forgotten the necessity of
intuition when approaching another human being.
People make so many miniscule signs that the brain will
filter away before they reach the conscious mind—and
sometimes wrongly so. While it may not be feasible for
doctors who do not practise medicine to keep up with
the latest clinical developments, there is no excuse for
failing to approach another person with empathy and an
open mind