The American Journal of Managed Care November 2004 - Part 2
Assessing the Accuracy of Computerized Medication Histories
Objective: To determine the accuracy of computerized medication histories.
Study Design: Cross-sectional observational study.
Patients and Methods: The study sample included 493 Department of Veterans Affairs primary care patients aged 65 years or older who were receiving at least 5 prescriptions. A semistructured interview confirmed medication, allergy, and adverse drug reaction (ADR) histories. Accuracy of the computerized medication lists was assessed, including omissions (medications not on the computer record) and commissions (medications on the computer record that were no longer being taken). Allergy and ADR records also were assessed.
Results: Patients were taking a mean of 12.4 medications: 65% prescription, 23% over-the-counter products, and 12% vitamins/herbals. There was complete agreement between the computer medication list and what the patient was taking for only 5.3% of patients. There were 3.1 drug omissions per patient, and 25% of the total number of medications taken by patients were omitted from the electronic medical record. There were 1.3 commissions per patient, and the patients were not taking 12.6% of all active medications on the computer profile. In addition, 23.2% of allergies and 63.9% of ADRs were not in the computerized record.
Conclusions: Very few computerized medication histories were accurate. Inaccurate medication information may compromise patient care and limit the utility of medication databases for research and for assessment of the quality of prescribing and disease management.
(Am J Manag Care. 2004;10(part 2):872-877)
Studies have demonstrated that the medication profile
in outpatient and inpatient medical charts
often is inaccurate.1,2 Due to the lack of reliability
of the medical record as an accurate source of medication
history, many hospitals and clinics have begun
using computerized medication profiles, and many
groups and government agencies advocate computerized
medical records and physician order entry to
reduce the incidence of adverse drug events and medication
errors.3-5 However, little is known about the
accuracy and reliability of computerized medication lists.
In addition, pharmacy benefit management (PBM)
databases are increasingly being utilized in clinical
research. Information from these large databases has
been used to assess compliance and adverse drug events
in several studies.6-8 In addition to these private insurance
databases, the Department of Veterans Affairs
(VA) has a large pharmacy database used for clinical
research. This VA pharmacy database has been utilized
to calculate the Chronic Disease Score (RxRisk-V) to
assess the burden of chronic disease on treated populations,9
assess healthcare utilization within the VA system,10
and evaluate prescribing practices.11 To our
knowledge, no studies have evaluated the validity of
clinical data found in computer medical records.
This study was performed to evaluate the agreement
between information in the VA computerized medication
profile and information obtained through a structured
The study was conducted at the Iowa City, Iowa VA
Medical Center (VAMC) primary care clinics. The Iowa
City VAMC is a 100-bed hospital and a primary teaching
affiliate of the University of Iowa Carver College of
Medicine. Sixty internal medical residents, 10 staff
physicians, 4 physician's assistants, and 3 nurse practitioners
staff the primary care clinics.
The patients in this evaluation were aged 65 years and
older, were enrolled in a primary care clinic at the Iowa
City VAMC, and had active prescriptions for 5 or more
regularly scheduled nontopical medications. Patients with
impaired cognitive function or enrolled in a pharmacist-based
anticoagulation clinic were excluded. Patients gave
informed consent, and the institutional review board at
the University of Iowa Carver College of Medicine and
Iowa City VAMC approved the study protocol.
A clinical pharmacist evaluated patients at their clinic
visit. The evaluation consisted of a focused record
review of the computerized
medical record, a semistructured
patient interview, identification
and classification of
and a history of allergies and
adverse drug reactions (ADRs).
Medication lists compiled by
structured patient interview
were compared with computerized
medication profiles at
the time of the interview to
determine overall agreement.
Patients were instructed to
bring all of their medications
with them for the interview.
Accuracy of the computerized
medication list was
assessed by 3 methods. The
first was the percentage of
patients who had perfect
agreement between the medication
profile gathered via
structured interview (ie, the
actual number and names of the medications taken)
and the number and names of the medications on the
computerized record. The second method assessed
omissions, which were defined as medications that
were not on the computer record, but that currently
were taken by the patient. To determine the overall percentage
of omissions, the denominator was the actual
number of medications taken by the patient. Finally,
commissions were defined as medications that were on
the computer record, but that were not currently taken
by the patient. To determine the overall percentage of
commissions, the denominator was the total number of
medications on the computer record. Omissions and
commissions were reported as both the mean number
of medications per patient and a percentage of the total
number of medications for all patients in aggregate.
The effect of copayment status on commission and
omission rates also was examined. Veterans who receive
medications dispensed by the VA may be required
to make a copayment for their medications. The typical
copayment is $7.00 for a 30-day supply of a medication.
Allergy and ADR agreement between computerized
profiles and structured patient interview was assessed.
An allergy was defined as a known sensitivity or hypersensitivity
to a drug, and an ADR was defined as any noxious,
unintended, and undesired effect of a drug after
doses used in humans for prophylaxis, diagnosis, or therapy.
The structured interview was compared with the
computerized allergy/ADR information to assess perfect
agreement of the allergy/ADR information. Omissions
were reported as both the number of patients with an
allergy or ADR not included on the medication profile,
and the number of allergy and ADR omissions for all
patients in aggregate. Commissions included the number
of allergies and ADRs found on the computer medical
record, but denied by the patient.
Proportions, means, and standard deviations were
reported where appropriate. Differences between
copayment status were compared with a Student's t
test. All analyses were conducted using SAS version 8.1
for Windows (SAS Institute. Cary, NC).
A total of 493 patients were evaluated. Their mean age
was 74.3 years, 97.8% were male, and 70.0% made a
copayment for their medications (Table 1). Patients had
a mean of 10.7 medications on their computer medication
profile and were taking a mean of 12.4 medications.
Of all medications, 65% were prescription, 23% were
over-the-counter (OTC) products, and 12% were vitamins/herbals.
The percentage of patients with complete
agreement between their computerized medication profile
and what they were actually taking was 5.3%.
There was a mean of 1.3 commissions per patient;
12.6% of all medications on the computer list were not
being taken by patients (Table 1). There was a mean of 3.1
omissions per patient; 25.0% of all medications the patients
were taking were not included on the computerized
medication profile. Our results indicate that very few
patients had complete agreement between the structured
medication history and computerized medication lists.
We also evaluated the effect of copayment status on
the accuracy of the medication profile (Table 2). In our
sample, 70% of the patients were required to make a
copayment for their medications. Patients with a copayment
had a mean of 9.6 medications listed on their computerized
medication profile, compared with 13.0 in the
group without a copayment (P < .01). The copayment
group had a mean of 1.3 commissions per patient, and
the group without a copayment had a mean of 1.6 commissions
per patient (P = .06). That is, in the copayment
group, 13.0% of medications were commissions, compared
with 12.0% in the group that did not have to make
a copayment. Copayment status was not significantly
associated with number of commissions.
To evaluate the omissions, the denominator becomes
the total number of medications the patient was actually
taking. The copayment group was actually taking a mean
of 11.8 medications, compared with 13.9 for the group
without a copayment (P < .01). The copayment group had
3.4 omissions per patient, and the group without a copayment
had 2.4 (P < .01) That is, 28.7% of the medications
taken by patients with a copayment were not included on
the computerized medication record, compared with
only 17.5% for patients who did not have a copayment.
Patients with a copayment
had a significantly
higher number of
omissions on their computerized
Table 3 lists the
omissions by mutually
(13.3%), and gastrointestinal
represent the drug
classes most frequently
included on the
that the patients were
no longer taking
(commissions). Vitamins/minerals (26%),
and gastrointestinal agents (11.5%) were the classes of
medications most frequently omitted from the computerized
medication list (omissions).
Table 4 lists the top 10 commissions and omissions
by individual drug name. Aspirin (5.0%), docusate
(3.5%), and albuterol (3.1%) were the 3 agents most likely
to be found on the computerized list, but which
patients were no longer taking (commissions). Sixty-six
percent of all commissions were prescription medications.
The individual drugs most frequently omitted on
the computerized profile were aspirin (10.4%), multivitamins
(8.2%), and acetaminophen (6.7%). Thirty-four
percent of all omissions were prescription medications.
Two thirds of patients (n = 318) had complete agreement
between the computer record and patient report
for allergies and ADRs (Table 5). Thirty-eight patients
(7.7%) reported at least 1 allergy that was not on the
computerized medication record. Of the 215 allergies
reported by these patients (some patients had more
than 1 allergy), 50 allergy omissions (23.2%) were not in
the computerized medication record. There were only 2
(0.9%) allergy commissions, in which the patient denied
an allergy that was in the computer medical record.
A total of 140 patients (28.3%) reported at least 1
ADR that was not on the computer medication record.
Of the 360 total ADRs confirmed by patients, 230
(63.9%) were not included in the computerized medication
record. There were 6 ADR commissions (1.6%), in
which the patient denied an ADR that was recorded in
the computer medical record.