Quantifying Risk, Reliability and Uncertainty

As we strive to be a “high reliability” organization, we need to be able to predict and prepare for any eventuality. Quantifying risk and uncertainty is therefore essential to improving the safety and quality of the services that we provide.

Working with colleagues in our Emergency Department (ED), we estimated the average census in the ED by hour of day, for each day of the week based on historical data (see chart below). On any given Monday, for example, there is a 50% chance that the ED will reach capacity by 2:00 PM. What do we do on those days when we are at capacity? Do we have enough staff to accommodate higher demand?  This simple percentile distribution provides useful information for staffing and scheduling decisions, and provides baseline data for improvement initiatives related to patient flow.



The ED meets weekly on patient progression challenges – it helps them understand problems that occurred in the previous week, and consequently prepare for the coming week. We support them with data and statistical analysis as well as occasional mathematical modeling of patient flow challenges. Of note, their continuous improvement work has led to a significant reduction in “boarding times” – the time a disposition decision has been made until the patient has departed to an inpatient unit – down more than 20% in 2016 compared to the previous year.