Development and Evaluation of a Measurement of Emergency Room Overcrowding

by Marc Afilalo | Jul 01, 2001

Key Implications for Decision Makers

Problems of emergency room overcrowding are often aggravated by extended patient stays. In addition to the organizational factors, some socio-demographic and clinical patient characteristics can have a decisive influence on their length of stay.

The client load tool produces more accurate client profiles, better planning and more effective resource allocation based on clearly defined needs. Load corresponds to the predicted length of stay. This tool therefore can more accurately predict length of stay for specific client groups, based on fluctuations in demand.

The client load tool can help:

  • Predict the load of clients coming to the emergency department, using a number of characteristics, especially patient age, reason for visit, presence or absence of endocrine problems and clients' perception of the seriousness of their condition.
  • Assess the impact of a changing clientele on the healthcare system.
  • Compare the emergency departments of various hospitals in terms of their client load and their performance.
  • Assess the variation in type of emergency department clientele in a given hospital for various periods.
  • Guide managers and decision makers in the distribution of allocated public funds, based on emergency department clientele in hospitals.

Executive Summary

Background

The Ministry of Health and Social Services has introduced several initiatives in recent years to improve conditions in emergency departments in Quebec. In the past two years, this has resulted in a Forum on emergency room conditions in 1999, the publication of an Emergency Department Management Handbook, the creation of the National Emergency Department Co-ordinating Centre (CCNU) in 2000, and of a committee of experts in charge of this centre.

The client load project was developed in the context of the committee's objectives by striving to design an assessment tool for administrators to measure the load of clients coming to emergency departments. This measurement could help managers determine resource allocation based on clientele profile, assess the benefits of measures implemented by the ministry, and conduct comparisons within and between institutions.

A management tool

Client load is generally measured through cost of treatment and is highly influenced by organizational factors specific to the institution. The proposed tool focuses on a different aspect - the length of each visit to the emergency department. This load is measured solely on the basis of characteristics specific to the patient.

The client load tool could also round out the work accomplished by the transitional measures introduced by the Bédard Commission, based on a cost analysis of services, not the actual clientele. When integrated with the tools developed by the Canadian Emergency Department Information System (CEDIS) and National Ambulatory Care Reporting System (NACRS), the client load tool could help enhance the capacity for assessing the patient clientele.

The client load tool could help manage human resources more effectively, as well as the layout of facilities, number of gurneys, or any other resource. The proposed tool would also allow administrators to measure indirectly whether existing resources are adequate by comparing the predicted load with the observed load in an emergency department. It could also serve as an indicator of quality control or to justify adjustments based on the client groups served.

Approach

Based on close to 3,000 patients, the research factors in the standards set by the tactical intervention group for the triage process in emergency departments, which must be completed within an average of five minutes for each patient.

The research project was carried out in four phases:

  • Development of a conceptual model describing the factors that influence length of stay in an emergency department.
  • Consultations with focus groups to draw up a list of individual patient characteristics that have a potential effect on length of stay.
  • Data gathering based on a list of characteristics and a prospective sample of patients from various emergency departments.
  • Development of a model to predict length of stay.

The results obtained with the client load tool describe the clientele rather than the hospital. They serve to estimate length of stay for patients coming to the emergency department by using characteristics specific to a patient upon arrival in emergency triage. The first series of variables establishes the probability of inclusion in a heavy population. The second predicts the length of stay.

Clientele profiles

  • Variables for predicting length of stay:
  • Age
  • Reason for visit
  • Code assigned in triage
  • Patient perceptions of the seriousness of their condition
  • Number of admissions to hospital over the past three years
  • Whether the person works
  • Presence or absence of endocrine problems
  • Presence or absence of memory problems
  • The results of an assessment of daily activities

Further research

The research conducted constitutes the first stage in development of a tool to estimate client load effectively and objectively.

The subsequent stages consist of:

  • Validating the tool in various settings (for example, overcrowded and uncrowded hospitals); and
  • Verifying whether the tool can be used to discriminate between various population types (for example, elderly patients, cardiac patients, etc.).

Much work remains to be done to analyse contextual or organizational factors in emergency departments. That analysis can now be completed by a measurement system that can assess emergency department functionality in relation to organizational variables.