Root causes: A roadmap to action
Click here for printer friendly version
From the August 26, 2004 issue
Problem: After receiving a recent report of a medication
error that reached a 4-year-old child, we were once again reminded
that errors are almost never caused by the failure of single
element in the system. More often, there are multiple underlying
system failures that lead to an error, many of which can be
identified when the involved healthcare providers take the time
to uncover them. Thats exactly what happened in the following
case, which was submitted to ISMP with an expressed desire to
share the event for widespread learning in the healthcare community.
Partial patient identifier. The story begins with
a hand-written order for carbamazepine 400 mg twice daily
for an adult patient with a history of seizures. After receiving
the order, a pharmacist retrieved the patient profile by typing
the last name only, and subsequently entered the medication
into the profile of a 4-year-old child with the same last
name as the adult patient for whom the medication had been
prescribed.
Computer system weaknesses. During order entry, the
pharmacist did not recognize that the dose of carbamazepine
represented an overdose based on the childs weight.
In fact, he failed to notice that the patient was a child,
as age was not in a prominent location on the order entry
screen. As a rule, patients diagnoses and comorbid conditions
were not listed on the pharmacy profiles, so the pharmacist
was unable to match the prescribed medication to the childs
medical condition. Likewise, the pharmacy computer did not
require entry of a weight for pediatric patients. Nevertheless,
even if a weight had been entered, there were no functional
dose alerts in the system.
Nonstandard MAR checks. A pharmacy computer-generated
medication administration record (MAR) was delivered to the
unit later that night, but the night nurse did not notice
that carbamazepine was listed in error. The MAR verification
process was not standardized in the hospital, leading to variation
in how different nurses carried it out, if it was carried
out at all. In fact, there was no official policy stating
that the MAR verification process was required; there was
no written procedure for reference; and the process was not
covered during nursing orientation.
Adult dose/dosage form for a child. The next morning,
a nurse gave the child the first erroneous 400 mg dose of
carbamazepine, as listed on the MAR, after crushing the tablets.
She failed to recognize that the dose was too high for a child.
She also never thought to question why the pharmacy had sent
oral tablets for a 4-year-old child, or to ask whether there
was a more suitable dosage form for a child (the drug is available
in chewable tablets and as a liquid suspension).
Unverified patient history. The nurse who administered
the first dose of carbamazepine assumed that the child was
receiving the medication because he had a history of seizures.
However, the nurse did not verify this by checking the patients
medical record. In fact, the child did not have a history
of seizures. Whats more, the nurse passed this erroneous
information on to the nurses working the next shift who, in
turn, continued to pass it along during the following shift-to-shift
report. Thus, the child received three doses in error, from
three different nurses, all of whom believed the child had
a history of seizures.
Language barrier. The childs parents were present
when one of the erroneous doses was administered to their
child. At that time, the nurse attempted to tell the parents
that the medication was used to control seizures. However,
the parents (and child) had very limited understanding of
English, so they were unable to intervene to correct the erroneous
seizure history.
Poor physician access to MAR. The childs physician
did not detect the error right away during routine rounds.
The nursing MAR was not readily accessible for review, and
there was no electronic or pharmacy computer-generated summary
of prescribed therapy. Thus, he did not notice that his patient
was receiving a medication that he had not prescribed.
The error was finally detected after the child became lethargic
and developed nausea and vomiting. At that point, one of the
nurses suspected a problem with the carbamazepine dose and
contacted the physician, who stated that the medication had
never been prescribed for the patient. At the time of discovery,
the childs carbamazepine level was 18 mcg/mL (normal
therapeutic range is 4-12 mcg/mL). This error delayed the
childs discharge, though he recovered without further
problems.
Safe Practice Recommendation: It may be discouraging
to see how many things go wrong when a medication error reaches
a patient. However, a thorough root-cause analysis clearly
demonstrates that there are also many different ways to avoid
similar errors. Had any one of the following system enhancements
been present, its likely that the error would have been
prevented or corrected before it reached the child, or at
least detected before the child sustained any harm.
--Use two patient identifiers (e.g., full name, identification
number, date of birth) to verify patient identity when entering
orders (although not specifically required by the Joint Commission,
we highly recommend it; see our June 3, 2004, newsletter for
details).
--Ask the pharmacy system vendor to build look-alike
patient name alerts into the order entry system
for activation when more than one patient has the same last
name.
--Employ a computerized prescriber order entry system
that is interfaced with the pharmacy computer system to eliminate
the need for pharmacy order entry.
--Standardize, document, and require implementation of a
MAR verification process whenever new MARs are distributed
(or rewritten). Orient new nurses to the required process.
Also ask nurses to compare the pharmacy label (on dispensed
medications) to the initial MAR entry before the first dose
is administered to ensure that the pharmacists and nurses
interpretation and transcription of a medication order is
correct.
--Require documentation of a verified past medical history
(including comorbid conditions) on order entry screens,
MARs, and other records used during change-of-shift reports.
Establish an effective method for communicating this information
to pharmacy so that nurses and pharmacists can always match
the prescribed drug therapy to a verified medical condition.
--Require recalculation of all doses of pediatric
medications before dispensing the drug (pharmacists) and during
initial order transcription/verification onto the MAR (nurses)
to ensure the dose is appropriate.
--Require an entry of weight in computer systems
for pediatric patients before processing orders; and establish
a communication process that facilitates the timely transfer
of accurate patient weights from nursing to the pharmacy.
-- Build and test maximum and subtherapeutic dose alerts
in the order entry system (based on patient age and weight
when applicable).
--Encourage nurses to investigate the possibility of an
error if drugs for pediatric patients are dispensed in
adult dosage forms.
--Post a daily electronic or computer-generated summary
of prescribed medications on each patients chart
and educate physicians about its value in detecting inaccuracies.
--Provide translators for patient/family teaching
about diagnoses, treatment plans, and newly prescribed medications.
Offer written drug information sheets (8th grade level or
lower) to patients in their primary language. Ask the patient/family
to demonstrate their understanding of written and verbal information
provided.
--Establish a process to thoroughly investigate all missing
medications before asking nurses to resend
an order, and/or before dispensing the medication again. In
this case, carbamazepine had not been dispensed for the intended
patient. Thus, nurses likely called the pharmacy for the missing
medication. When the drug could not be found on the correct
patient profile, had a pharmacist located the initial order
and noticed that it had been processed, perhaps further investigation
would have resulted in earlier discovery of the error.
|