Læknablaðið : fylgirit - 15.06.2002, Blaðsíða 19

POSTERS / ICELAND 2002: EMERGENCY MEDICINE BETWEEN CONTINENTS
Table 2. Age Group Sub-analysis (% change over 5 years).
EtOH OD’s Behavioral Suicide* MVA Non-MVA trauma
Age 6-10 0 0 +74 0 +5 +9 (46)
Age 11-15 0 0 . +87 +12 +2 +19 (179)
Age 16-20 +33 +49 +62 +117 +15 +43 (433)
* Suicide data field collected 1998-present
P 06 - ED Systems: Efficiency, Productivity
Emergency Department overcrowding: who is the odd one out?
Sanchez M
Hospital Clinic, Barcelona, Spain
Objective: The term “Emergency Department (ED) over-
crowding” is extensively used but is difficult to define scientifically.
The aim was to define overcrowding from the occupancy rate
(OR), and to establish the differences between overcrowded and
non-overcrowded periods from the OR analysis.
Methods: For 3 consecutive weeks, we recorded at 3-hour intervals
the number of arrivals and the number of patients placed in ED as
well as the reason for their continued stay. These reasons were
divided into 4 categories: A) factors related to ED itself: Al-waiting
for a physician, A2-being visited, A3-waiting for test results, A4-
waiting for outcome; B) factors related to hospital itself: Bl-waiting
for a bed going to be left, but still occupied, B2-waiting to have a
bed (lack of bed at that specific 3-hour interval); C) factors related
to ED interrelations: Cl-waiting for test performed out of ED, C2-
waiting for hospital consultant; D) factors not directly related to
ED or hospital: Dl-waiting for ambulance, D2-waiting for relative,
D3-waiting for social assistant. ED OR was calculated as a rate
between the number of patients placed in it and the number of
boxes. Overcrowding was defined as an OR>100%. Such a figure
could be surpassed because 2 patients per box were put when
needed. Percentage of OR for each reason was calculated as well.
Rcsults: ED was overcrowded (OR>100%) 5 days (24%). Despite
more arrivals (20 vs. 12.8, +56%, p<0.05) within overcrowded
periods, the major increase was registered in patients' continued
stay due to hospital factors (15.6% vs. 39.6%, +154%, p<0.001), and
in those not directly related to ED or hospital (1.8% vs. 6.8%,
+268%, p<0.001). Specifically, the percentage of patients waiting
for a bed going to be left (5.7% vs. 15.6%, +174%, p<0.01), waiting
to have a bed (9.9% vs. 23%, +143%, p<0.01), and waiting for a
relative (0.7% vs. 3.1%, +345%, p<0.05) were responsible for such
differences.
Conclusions: OR is not only a good marker to measure ED over-
crowding, but also allows us to analyze the reason for over-
crowding. ED occupancy rate should be an important hospital
administration tool for changing some hospital behaviors and
limitations. In this study, both the lack of enough hospital beds, and
the fact that inhospital patients are discharged too late at evening
hours lead to ED overcrowding.
P 07 - ED Systems: Efficiency, Productivity
•mpact of computerized order entry on emergency physician
time
Stair T
Origham and Women s Hospital. Boston. MA. United States
Introduction: Computerized order entry has been shown to im-
Prove documentation and reduce errors, but has required increased
physician time. We refined the program with a graphical user inter-
face, templates, and other time-saving features.
Objectives: To measure the effect of computerized order entry on
the distribution of physician time in one emergency department.
Methods: Observational study using a time motion technique and
convenience sample. In June 2000, a new computerized order entry
system was activated in the emergency department of our urban
academic hospital. For six months before and after, research
assistants documented physician activity minute-by-minute during
115 half-hour periods, giving a power of .90 to detect a difference of
10%.
Results: Computerized order entry caused no statistically signifi-
cant differences in physician time spent on any given activity. The
percentage of total time spent writing (or entering) orders in-
creased from 5% to 7%. Time spent with patients (including his-
tory, physical exam, and procedures) increased from 28% to 30%,
and time with staff increased from 37% to 40%. Time spent online
(information retrieval) fell from 12% to 7%.
Conclusions: Properly implemented, computerized order entry
caused minimal impact on physician time.
P 08 - ED Systems: Efficiency, Productivity
Access block predicts increased use of inpatient resources
Richardson DB
The Canberra Hospital, Australia
Objectives: “Access Block” (AB) refers to the situation where ED
patients requiring inpatient care are unable to gain access to
appropriate hospital beds for prolonged periods. It is recognised as
a cause of ED dysfunction and has been shown to be associated
with increased ED resource use and increases in waiting time, and
in the numbers of patients who do not wait to be seen. This study
aimed to examine the relationship between AB and inpatient
resource use measured as length of stay (LOS).
Methods: Retrospective descriptive study of all ED presentations
admitted to an inpatient bed in a tertiary hospital in 1999. Data
from the ED information system were merged with data from the
inpatient information system to calculate the total ED time (TEDT
- from arrival to departure ED). Cases were defined as
experiencing AB if their TEDT was more than 8 hours, and LOS
was calculated as the number of calendar days from departure from
ED to departure from hospital, or 1 for patients admitted and
discharged on the same calendar day. These are standard
Australasian definitions. LOS was truncated at 10 days. The null
hypothesis was that there is no difference in LOS between cases
that did and did not experience access block.
Results: Some 11906 admissions were included, with 7.7% expe-
riencing AB. Mean LOS in the first 10 days were 4.09 days without
AB and 4.87 days with AB (P<10E-10 by t-test). Subgroup analysis
showed that this effect was greater in the less urgent triage cate-
gories, but had little relationship with the day of the week, the
season, or the patient age. The effect was much greater in patients
who departed from the ED outside normal working hours (0800-
1600).
Conclusions: Access Block in this hospital is associated with
significantly increased use of inpatient resources over the next 10
days. This effect is independent of patient acuity and seasonal
Læknablaðið/Fylgirit 45 2002/88 19