Systems Concepts – Part 2: Best Parts Do Not Make A Best System

This vlog post is the second in a series about the concept of systems. It explores the systems design principle that having all of the best individually performing parts does not nesessarily provide the best performing system. That concept is then applied to the example of systems of care for out-of-hospital cardiac arrest. (Duration – 7:25)

Notes and Resources

  • Originally posted: January 2018


In the last post, we heard Dr. Ackoff explain what a system is in a general way. In this post, we’re going to look at how that concept translates to systems of care for high-risk time-sensitive emergencies. Dr. Ackoff’s example of a system was an automobile. Having all the best parts of cars does not mean you will have the best car. It will not even operate. The reason is that all of these –best- parts do not all fit together.

In a system of care, the same principle applies. The performance is determined not by the sum of the behavior of the parts of a system of care, it’s a product of their interactions. Let’s consider this in context of an out-of-hospital cardiac arrest.

Consider a theoretical community where they have an amazing level of bystander performance in recognizing a cardiac arrest – more than anywhere else!!! That same community also has an extremely high percentage of the population that also knows how to perform hands-only CPR – more than anywhere else! They have AEDs all over town – more of them per capita than anywhere else! They have a 9-1-1 communications center with the best technology and staff – as good or better than anywhere else! They have a fire-rescue department that has all the best equipment and a absolutely top notch staff. They do a great job of caring for patients prior to ambulance transport – they are the very very best at that!! This community has an ambulance service with the very best ambulances, equipment and staff. The crews do the very very best in assessment and treatment of patients that are in their ambulances. The community has an emergency department with the all the best equipment, facilities and staff. The care they provide to patients while in that ED is the very very best.

Now consider an out-of-hospital cardiac arrest in that community if all of those -best parts- do not fit together well. A person walking down Main Street collapses. The part of the system for training bystanders to quickly recognize CPR, call 9-1-1, and starting hands-only CPR is excellent. The call to 9-1-1 is made and the dispatcher helps with coaching the CPR on scene. However, the dispatch ‘part’ of the system is not connected very well to the ‘part’ of the system for bystander AED utilization. There’s a nearby AED, but the locations of the AEDs are not available to the dispatcher, so the opportunity for a very early defibrillation for this patient is missed.

There are recurring problems with assigning the call to the fire-rescue agency with jurisdiction at the location of an incident. Unfortunately, it is the jurisdictional boundary, not the closest available fire-rescue unit that determines which rescue unit gets the call. There are similar issues with the ambulance assignments. Eventually though, the fire rescue unit does arrive on-scene. They do excellent CPR, apply a state-of-the-art automated CPR device with absolute minimal interruption in compressions and also apply the defibrillation pads of a state-of-the-art monitor / defibrillator.

The ambulance crew arrives. The crews from the fire-rescue agency and ambulance service do not get along well with each other. The is a lot of institutional ego and take-over politics going on in the background. They do not train together and these conflicts spill over into diminished teamwork on scene. The patient is going in and out of arrest, so they need to keep the defibrillator patches and automated CPR device available if needed during transport. They use different monitor/defibrillators and the patches are not cross-compatible. The automated CPR device used by the fire rescue agency is not the same as used by the ambulance crew. They each have the best equipment and when working separately do a great job, but when they have to work together, things do not go smoothly. Eventually, the CPR equipment and defibrillator monitors are switched out to get the ambulance services equipment in place. The patient is placed into the ambulance, and they get a sustained return of pulses. They get a 12 lead. It clearly shows a STEMI. But, the ambulance crews and the ED staff do not get along very well either. They gave a report while en route, but did not give it much priority and hence, the report was not given until they were just around the corner – even though there were plenty of opportunities to give an early notification during the 14 minute transport without compromising patient care.

As a result, the ED had no time prepare. They did not have time to a get the bed opened up in the code room of the ED, so the crew and ED staff did their best in a different room. The ED assessed the patient, got another 12 lead, and drew blood to check cardiac enzymes. But, the cardiology department does not trust EMS or the ED to activate the cath lab. The interventional cardiologist eventually comes in to examine the patient. The interventionalist is an amazing operator and has all the very best equipment and skills to match. Eventually, the cath is performed and a stent was placed. There were no recurring instances of re-arrest. The patient was discharged alive with only a mild neurological deficit. Discharge orders for medications, specialty and primary care follow-up, and cardiac and neuro rehab were all ordered, but with incomplete compliance.

This patient was cared for in community with all of the best individual parts – but they do not all fit together very well. The outcomes for this patient, and many others are not nearly as good as they could be. The take away here is recognizing that the difference between the top and lower performing systems of care is primarily a result of how they make the pieces fit together – not how well each of the pieces work individually.

How well those pieces fit together is largely a function of policies, politics, and communications. These are things that can be controlled and optimized with minimal expense. But that does not make the challenges of fitting the pieces together easy to do. Great equipment can help. Training can help. Great technology can help. But all the pieces have to fit together well. We all need to think about these challenges from a systems perspective and not just about how to make our individual stovepipes better. Now, what if we can have the best parts –and- make them fit together extremely well. Is that the ultimate? Maybe, or maybe not. Let’s look at that in the next post.