ED Overcrowding: The Ontario Approach
Howard Ovens, MD, FCFP(EM)
Ontario is Canada’s most populous province, with approximately 12 million people and 130 emergency
departments (EDs). Canada has a national single-payer universal health care system, but provinces are
responsible for administration. After years of problems and failed attempts to address chronic ED over-
crowding, in April 2008 Ontario embarked on an ambitious program to improve system performance
through targeted investments (initially CAN$500 million over 3 years) and realigned incentives. Support-
ing the program were requirements for hospitals to submit timely data and targets for length of stay
(LOS) and annual improvements; results are publicly reported. The program has been continued this
year. While not all our provincial level targets have been met as yet, major improvements have been
made, especially in access to care and LOS in the ED for patients eventually discharged home. The
greatest improvements were made among the cohort of mainly urban, high-volume EDs that had the
worst performance at baseline. This presentation will highlight some of the controversies and challenges
and key lessons learned. Overall, the Ontario experience suggests ED overcrowding is a soluble
problem, but requires a system-level intervention.
ACADEMIC EMERGENCY MEDICINE 2011; 18:1242–1245 ª 2011 by the Society for Academic
overcrowding: preemergency factors that drive demand
for emergency services, size⁄capacity of the ED and its
efficiency of processing, and flow out from the ED to the
hospital wards.3My own jurisdiction, Ontario, Canada,
is currently in the fourth year of a multiyear ambitious
approach to ED overcrowding across the province.
Ontario is the most populous province in Canada
with over 12 million people (one-third of Canada’s
population). Health care is publicly funded throughout
Canada, but service delivery is the jurisdiction of the
province. In Ontario there are approximately 75 hos-
pitals with EDs whose censuses are over 30,000 vis-
its⁄year, spread over a huge geographic area, and
there are approximately another 60 smaller units in
mergency department (ED) overcrowding is a
widespread problem with many negative poten-
tial effects.1,2Many factors can contribute to
rural centers. ED overcrowding has been a problem
in Ontario for many years4and had been blamed for
some tragic cases involving delays in provision of
care to newly arriving patients. Overcrowding had
also led to high rates of ambulance diversion, and
problems with staffing of units with physicians and
nurses due to high stress and low morale, and was
generally becoming a political problem. In response
to these pressures, the government commissioned a
report on the subject in 2006.5The report reviewed
the problem of overcrowding in general and made
many recommendations, but the key findings of the
• ED crowding is generally due primarily to sick
patients who need care and usually inpatient beds,
not inappropriate ED use by patients with low-acuity
PII ISSN 1069-6563583
ª 2011 by the Society for Academic Emergency Medicine
From the Schwartz⁄Reisman Emergency Centre, Mount Sinai Hospital, and the Department of Family and Community Medicine,
University of Toronto, Toronto, Ontario, Canada.
Received June 17, 2011; revision received July 15, 2011; accepted July 20, 2011.
This article represents a component of the 2011 Academic Emergency Medicine Consensus Conference entitled ‘‘Interventions to
Assure Quality in the Crowded Emergency Department (ED)’’ held in Boston, MA.
Funding for this conference was made possible (in part) by 1R13HS020139-01 from the Agency for Healthcare Research and Quality
(AHRQ). The views expressed in written conference materials or publications and by speakers and moderators do not necessarily
reflect the official policies of the Department of Health and Human Services, nor does mention of trade names, commercial prac-
tices, or organizations imply endorsement by the U.S. Government. This issue of Academic Emergency Medicine is funded by the
Robert Wood Johnson Foundation.
Dr. Ovens acts as an advisor on health policy to the Government of Ontario. He has no other conflicts of interest or disclosures to
Supervising Editor: James Miner, MD.
Address for correspondence and reprints: Howard Ovens MD, FCFP(EM); e-mail: email@example.com.
problems (they require few resources). Thus reducing
ED demand is a small part of the solution.
• ED crowding is most commonly due to the boarding
ods—reducing or eliminating this should greatly
improve ED efficiency.
• The problem is system-based, and related to lack of
incentives, or even perverse incentives for hospital
administrators, as well as problems in the wider
health care system. It is beyond the capacity of indi-
vidual hospitals to solve alone—it requires a system
The government responded with a major investment
of effort and funds to try and solve the problem defini-
tively. They appointed an expert panel to provide advice
on their strategy (I have been a member of the expert
panel since inception and am currently the co-chair).
Initial investment was roughly CAN$500 million over
3 years. The program began officially April 1, 2008. The
approach has been summarized by one of its architects,
Dr. Alan Hudson, as ‘‘data, money, and shame’’ and
built on experience reducing waiting lists for scheduled
surgical and imaging procedures in the province. The
key elements were:
• Ordering hospitals to report a common data set to
the government monthly (we now have province-
wide data on a host of indicators within 3 weeks of
month end). These are ‘‘the data.’’
• Setting targets for length of stay (LOS) of ED patients
based on triage acuity and disposition—the target for
patients being admitted or complex patients (based
on arrival acuity) being discharged home is 8 hours
from arrival to departure to unit or home and for
noncomplex (low-acuity) patients is 4 hours from
arrival to discharge. The overall goal for the program
initially was to have 90% of patients seen within their
target by end of March 2011.
• Hospitals receive additional funds each year contin-
gent on meeting hospital-specific improvements in
achieving targets. This is called pay for performance:
• Hospital performance is
Other features include an expert panel to advise gov-
ernment on program planning, an intensive coaching
program for hospitals to provide staff with information
on best practices and to introduce them to ‘‘Lean’’ man-
agement techniques, and a major effort to improve hos-
pital outflow by reducing the number of patients in
hospital wards waiting for care in nonacute facilities
such as nursing homes, rehabilitation centers, etc.
(‘‘ALC’’ or alternate level of care patients).
After 3 years the results are mixed: the ED report-
ing system is a great success and provides a wealth of
timely data for local and provincial use. Ambulance
offload delays and diversions have largely been solved
(see Figure 1). Overall, 85% of all patients and over
90% of discharged patients meet their LOS targets
(see Figure 2). Some hospitals have achieved dramatic
improvements in flow; however, a few hospitals have
not made significant improvements yet. There has
been only marginal sustained improvement in flow of
admitted patients at the provincial level (see Figure 3),
and the effort to create capacity by reducing inpa-
tients waiting for alternate levels of care has had
some local success stories but only marginal success
overall. However, if we look at trends prior to 2008
and changes in population and hospital volumes since,
Mount Sinai St. Joseph’s St. Michael’s Sunnybrook Toronto East
90% TOC Time (pre) 90% TOC Time (post)
Figure 1. Ambulance offload times for central Toronto. TOC = transfer of care; pre = before ambulance offload nurse program
April 2008; post = March 2011; UHN = University Health Network (a multi-site hospital corporation operating ERs on two sites,
Toronto General [General] and Toronto Western [Western]). All times are 90th percentiles. Note decreased variability as well as
improved performance. Result is 63.3 mean unit hours saved⁄day. Data courtesy of Toronto EMS.
ACADEMIC EMERGENCY MEDICINE•December 2011, Vol. 18, No. 12•www.aemj.org
we would be in a much worse position on admitted
flow and ALC levels in Ontario today without the
At the unit level, I believe the changes have been sig-
nificant and worthwhile and have positively affected the
patient experience. At my own facility, Mount Sinai
Hospital, an urban academic center in Toronto, flow is
dramatically improved, morale is better, patient satis-
faction is better, and conflict between triage and ambu-
lance staff has disappeared. Our 90th percentile LOS
for all complex patients combined (all those with 8-hour
targets including admitted patients) has gone from
18.5 hours at program inception to 11 hours currently,
while noncomplex discharged patients with a 4-hour
Combined: Percent completed within 8/4 hours by Cohort
Year 1 sites (N=23)Year 2 sites (N=23) Year 3 sites (N=25)Provincial
Figure 2. Three-year trend by program year: percentage of all patients within target LOS. Total number of hospital sites = 128;
only the 71 sites with annual volume of > 30,000 visits were eligible for pay for results program, sites added to program in three
cohorts from worst performing in Year 1 to best in Year 3. Data courtesy of Access to Care at Cancer Care Ontario.
Admi?ed: Percent completed within 8 hours by Cohort
Year 1 sites (N=23) Year 2 sites (N=23)Year 3 sites (N=25)Provincial
Figure 3. Admitted: percentage completed within 8 hours by cohort. Total number of hospital sites = 128; only the 71 sites with
annual volume of > 30,000 visits were eligible for pay for results program, sites added to program in three cohorts from worst per-
forming in Year 1 to best in Year 3. Data courtesy of Access to Care at Cancer Care Ontario.
Ovens•ED OVERCROWDING: THE ONTARIO APPROACH
target have had their 90th percentile LOS in the same
period go from 6.6 to 4.3 hours. In October 2009, due
to H1N1 influenza, Ontario had a record volume of ED
400,000), but flow performance improvement was main-
tained, demonstrating the sustainability of the improve-
ments in the face of a significant volume surge.
Provincial performance also held steady during a signif-
icant province-wide bed crunch related to the higher
acuity flu season of this past winter.
The surprising thing so far is that the improvements
have been largely due to coordinated efforts to improve
processing within EDs, as flow out of the ED is only
modestly improved. Thus, most of the improvement we
have seen has been due to culture change, some tar-
geted staffing increases, and process improvement
within the ED: ‘‘RAZ’’ units (see below), patient flow
coordinators (coordinate bed supply and demand daily
hospital-wide), ambulance offload nurses (dedicated
staff to assume care of patients arriving by ambulance
in the waiting room if necessary—which has almost
eliminated the ambulance offload delay problem in
urban areas), and better staff scheduling (match peaks
in patient arrivals better).
One of the most successful and widely adopted strat-
egies within EDs has been what we have called ‘‘RAZ’’
for Rapid Assessment Zones. To prevent stretcher grid-
lock, an area in the ED is set aside for an internal wait-
ing room and patients who are able get undressed and
into hospital gowns and wait in a chair. They only use a
stretcher in one of several nearby exam rooms when
the doctor examines them, for nursing procedures, etc.,
and at all other times they wait in the chair. Patient
acceptance has been surprisingly good.
The program has not been without controversy.
Some health care workers have felt they were being
criticized or were being asked to work harder, faster,
or longer and were already doing the best they could.
We have tried to stress that individuals are being asked
to work smarter, not harder, within a better, more effi-
cient system—and that they should never compromise
any individual’s care to meet performance targets. To
support this position, we allow hospitals to elect to
observe up to 5% of their patients ‘‘off the clock’’ by
admitting them to a conforming observation area
(‘‘CDU’’ or clinical decision unit) of their ED under
strict performance criteria. While some have felt this is
a cynical attempt togame
changing patient experience, we monitor compliance
closely and feel it is a worthwhile effort to ensure no
one uses the program and its targets as an excuse for
providing substandard care.
Perhaps the greatest barrier to solving ED over-
crowding is an attitude of nihilism, that nothing can be
done. Efforts at the individual unit and hospital level to
improve flow are to be supported, reported, and cele-
brated. However, ultimately the experience in Ontario
and other jurisdictions4,7,8proves that ED overcrowd-
ing is a system problem requiring a system solution.
2. Richardson DB, Mountain D. Myths versus facts in
emergency department overcrowding and hospital
access block. Med J Aust. 2009; 191:369–74.
3. Juan A, Enjamio E, Moya C, et al. Impact of hospi-
tal management measures to increase efficiency in
the management of beds and reduce emergency
department congestion. Emergencias. 2010; 22:249–
4. Schull MJ, Szalai JP, Schwartz B, Redelmeier DA.
systematic hospital restructuring: trends at twenty
hospitals over ten years. Acad Emerg Med. 2001;
5. Physician Hospital Care
Access to Emergency Care: Addressing System
Issues. Available at: http://www.health.gov.on.ca/
access/improving_access.pdf. Accessed Sep 2, 2011.
6. Ontario Ministry of Health and Long-Term Care.
Ontario Wait Times. Emergency Room Search Sec-
search/er/. Accessed Sep 2, 2011.
7. Olshaker JS, Rathlev NK. Emergency department
overcrowding and ambulance diversion: the impact
and potential solutions of extended boarding of
admitted patients in the emergency department.
J Emerg Med. 2006; 30:351–6.
8. Trzeciak S, Rivers EP. Emergency department over-
crowding in the United States: an emerging threat
to patient safety and public health. Emerg Med J.
ACADEMIC EMERGENCY MEDICINE•December 2011, Vol. 18, No. 12•www.aemj.org