| |
Doesn't Compute
August 1, 2005 -- One of the places where I have worked as a
volunteer was a post-op and orthopedic floor high up in a tall tower
in a hospital that shall remain nameless.
In this hospital (and, I suspect, in most) the manual, paper-based
procedure for reviewing and renewing medication orders is stunningly
labor-intensive and error-prone. This is a task for which computer
implementation would be ideally suited. It is astonishing to me that
this hasn't been done.
Here's how it works today: In the MAR (Medication Administration
Record) portion of a patient's chart there are two sheets:
-
PRN (drugs administered on an as-needed basis) and One Time
Medication Record
- Routine Medication Record.
Each has seven rows (one row for each medication that has been
ordered) and five columns (one for each of five days). In each cell,
the nurse records the administration of a med on a date. (For meds
that are administered several times per day, the cells can accommodate
several entries.)
On each fifth day of each patient's hospital stay, the nursing
station secretary must see that the form is full and must complete a
new form, copying from the old one all the following information:
- Patient allergies and contra-indications
-
All the drugs, their routes, dosages, schedules, and formulations.
The probability of error is high, so the work must be checked by a
nurse.
The only possible reason I can see for lack of automation of this
procedure is to guarantee that it periodically gets a pair of eyes
to review the medication orders. But are the secretary's eyes
qualified to perform this review? Is five days the right review
period or just the number of columns that fit on the page? Is the
labor cost justified? Have errors ever been caught this way? If
indeed there does appear to be something dubious about a drug order,
is there a mechanism for having it reviewed (and perhaps modified)
by a doctor, especially the doc who made the order in the first
place? I suspect the answers to these questions are mostly no.
|
