The lessons I have learned are very apt for managing services and it seems we are doomed to fail unless good people in the frontline are able to escape from the never ending treadmill, writes Dr. Peter Lachman.
It is how we use the resources we have to meet the increased demand that will be the indicator of how we succeed. For front line staff there is the need to meet the targets set afar and often that are not possible with the way the system is designed.
Over the past few months, I have been working with clinical teams in Ireland and the UK all aiming to improve care of their patients. The issues that they face are similar and often difficult to resolve and are familiar for all who read this journal, i.e. too many patients, too few resources and lack of processes that are designed to achieve the outcomes. We ask them to improve outcomes with the resources that they have. This means they need to work smarter and learn to understand the variation in their system. What is key to improvement is how we all use data. It’s when you study variation that one finds the opportunities to improve.
What is key to improvement is how we all use data. It’s when you study variation that one finds the opportunities to improve.
In many settings, data is presented as Pie and Bar Charts at a single point in time. In healthcare the use of RAG ratings has become the norm. Those red amber green construct superficially seems to offer the reader an idea of what is going on; i.e. those in Red are failing, those in Green are succeeding and those in Amber are somewhere in between. Unfortunately, this is a unidimensional approach to understanding what is really happening and masks the trends that may indicate improvement or a move to lack of good outcomes. One may be losing ground even if one is above the bar and improving if below.
It is time that we looked at alternatives to this approach as being red does not necessarily mean you are failing – the pass line may be arbitrarily set and an organisation may in fact be improving. We were fortunate to have Lloyd Provost (@lloydprovost http://www.apiweb.org/index.php/associates/lloyd-provost) visit Dublin recently. Lloyd worked closely with the HSE on his visit to offer an alternative approach to data management. This approach is summarised in a recent article in BMJ Quality and Safety.
In the article, Jacob Anhøj, and Anne-Marie Blok Hellesøe explain why this addiction to RAG is detrimental to the understanding of what is really going on. They demonstrate that it does not show trends over time, and fails to demonstrate normal variation. They conclude that “Red, amber, green displays are intended to help managers make quick data-driven decisions according to whether a measure meets its target or not. Red, amber, green is a waste of time and resources and potentially harmful because it hides important information while conveying false messages.”
Red, amber, green is a waste of time and resources and potentially harmful because it hides important information while conveying false messages.
So what is the alternative? In his excellent book The Health Care Data Guide: Learning from Data for Improvement (http://eu.wiley.com/WileyCDA/WileyTitle/productCd-0470902582.html),and in the article, there is strong evidence for using Statistical Process Control (SPC) charts as the basis for data presentation. In this way one can look at trends over time and study variation as being common cause, and something one would expect, as opposed to special cause, which would require investigation. Too often we respond to “improvement” or “deterioration” when they are nothing but what the system can deliver. An example is the data on patients on trolleys in the ED in Ireland, which are best presented in an SPC chart. In most cases there is just normal variation over time as the system is designed to have approximately 400 patients on trolleys every day and there is no way this can be reduced without a radical redesign of the system.
It is time that we looked at a vector of measures which will show how the system is really performing. This will give an idea of the key measures over time and provide us and the public with reliable and meaningful data to consider. Read the article or better still read the book to study the way to use data in healthcare. This will be a defining moment for improving outcomes.