Stanford, Zaric, Carter: Downstream bottlenecks – the hidden cause of ER congestion

Special to Western News

What happens when a busy 16-lane highway funnels into something half its size? Anyone who drives regularly on the 401 knows. Backups extend far from the point of lane reduction into the 16-lane stretch. No one would suggest adding more lanes to the 16 already in place, as the problem lies in the downstream bottleneck created by the lane reduction.

That, in a nutshell, is the parallel to the problem today with overcrowded emergency rooms once again in the news.

The scenario is familiar: Vast volumes of incoming traffic, patients waiting for hours, and sometimes days before being admitted. For most, a visit to emergency is an inconvenience to be avoided if at all possible. However, for some, the results are tragic, as was the case recently in Montreal.

There are important problems in ER itself, such as patients who forego walk-in clinics and other alternatives, or who don’t have a primary care physician. Diverting some of this “traffic” from the ER “highway” would have notable benefits in reducing congestion levels. Recent events have shown, however, that congestion downstream is the main factor affecting backups in emergency.

Peter Toth, Canadian Association of Emergency Physicians president, said the “root cause” of the problem was that “hospitals are currently running at 100 per cent.” The main problem is widely recognized as patients occupying acute care beds waiting for long-term care (LTC). In fact, it is just the largest piece in a wider problem.

In Ontario, “alternate level of care” (ALC) describes all situations in which a patient occupies the wrong type of bed. The ALC estimate from 2009-10 was 15.7 per cent, up from about 14 per cent in 2008-09. Typical estimates of ALC run up to 40 per cent of all patients in many medical wards.

A chronic shortage of LTC space may lead to patients remaining in hospital wards. The wards can also be blocked due to a lack of access to other downstream facilities such as rehabilitation hospitals, which in turn can be backed up because of a lack of readiness in the home or in LTC, all of which prevent patient discharges.

In Ontario, efforts to alleviate these problems by adding more LTC capacity are hampered by a requirement that new LTC facilities establish 95 per cent occupancy before they open their doors. This inadvertently induces new facilities to accept patients who might otherwise not require LTC for some time. By ensuring such high utilization of these facilities before they even open, new LTC capacity can make very little contribution to alleviating upstream problems of ALC and excessive waiting in emergency departments.

Solving ALC problems is bound to be very challenging. The key philosophy to address it is to keep patient flow moving, by providing a remedy to situations where one patient population backs up others. One recent success has focussed on patients who no longer require an acute care bed but are not medically ready to participate in a course of rehabilitation, or who require additional time to determine the appropriate disposition. The introduction of a Transitional Care Unit, where patients continue to receive restorative care, has proven successful in several environments, including Parkwood Hospital in London, Bridgepoint in Toronto and the Southern Health network in Melbourne, Australia.

In our opinion, for maximum benefit, the talents of industrial engineers and management scientists, with their insights about patient flow, should be called upon more often. This was the case with the establishment of a Medical Assessment Unit (MAU) at Monash Medical Centre in Melbourne, Australia, where a patient flow specialist with expertise in IE/MS techniques was part of the team.

The goal was to identify patients with the “right” characteristics for diversion to the inpatient team in MAU shortly after presentation, in the hope of resolution on a time scale of hours rather than days; those not discharged within 48 hours are then admitted to the hospital. The introduction of a 10-bed General Medicine MAU, along with other initiatives, led to a 16 per cent increase in the proportion of General Medicine patients staying less than 48 hours and a six hour reduction in the average time in ER.

An aging population will cause increasing demand rates for many medical services over the coming decade. Difficult decisions need to be made as resources are limited. The involvement of experts who understand randomness, process flow and congestion will help to ensure that time and resources are not wasted reinventing the wheel.

David Stanford is a professor in Statistical & Actuarial Sciences and queuing theorist at the University of Western Ontario. He is currently splitting his sabbatical between the University of Melbourne in Australia and the Centre for Research in Healthcare Engineering at the University of Toronto.

Greg Zaric is an Associate Professor and Canada Research Chair in Health Care Management Science at The Ivey School of Business, University of Western Ontario.

Mike Carter is a Professor of Mechanical and Industrial Engineering and Director of the Centre for Research in Healthcare Engineering at the University of Toronto.