I once had a student in my course on Improving Healthcare Processes tell me that since they were pregnant they would look for a hospital that was not an Academic Medical Center (AMC) because, “Everyone knows that the AMC has to be less efficient.” I didn’t want to debate the point at the time. Healthcare, and particularly childbirth and related care are intensely emotional and personal topics. I also understood that the AMC setting by nature adds steps to any care delivery process, and common sense suggests that adding steps to a process must increase a variety of metrics that are easy to measure such as activity times, and cycle times. (In this context Cycle Time means time from entry to exit from a medical facility.)
Example – A Prototypical Clinic
Take a simple clinic visit as an example. In a proto-typical Private Practice setting, a patient enters the clinic, interacts with staff at the front desk, waits to see the physician, enters an examination room, interacts with the attending physician (Attending), visits the front desk to make the next appointment and then leaves the clinic. When we look at the same event in the AMC we have to add multiple steps. In this setting it is typical to have the patient placed in the Exam room, followed by a review of the case by a Resident or Fellow, then face time with the Resident, a discussion between the Resident and the Attending, a subsequent visit involving the Resident, the Attending, and the patient in the exam room before the visit is completed. Thus, the educational mission of the AMC adds at least 3 steps. Review of the Resident, Face Time with the Resident, and a conversion involving the Resident and the Attending to the standard process flow. Common sense dictates that this will extend the Cycle time for at least 2 reasons. One, steps are being added, and Two the Resident is almost always slower than the attending to complete the same task. In other words, the AMC sacrifices efficiency in the name of Education.
An alternate view
However, this common understanding is woefully incomplete. Our research in this space is presented with full technical detail in Williams, et.al (2012), and is fully explained in our book at, Amazon.com, but here is the thumbnail version of what happens. Adding the Resident to the process does increase the Cycle Time for each patient because it adds steps and a slower server. However, it simultaneously adds an alternate approach to manage patient flows because it allows parallel processing. This refers to the ability for the most expensive resource (the Attending) to focus on one patient while a less expensive resource (the Resident) faces another. This allows the patient to have increased face-time with a Physician, even though there may be reduced face-time with the Attending. This also increases the capacity of the system because double-booking patients becomes feasible if 2 doctors are available to serve patients in 2 rooms at the same time. In addition, an Attending now has options to accelerate a patient’s visit if needed by changing or omitting the Residents interaction with the patient.
This more complex process makes it harder to predict cycle times, capacity, and waiting times simultaneously. In this context we can think of capacity as the maximum arrival rate that can be processed by the system without falling behind schedule. A famous result known as Little’s Law indicates that these three variables are interrelated. The average number of jobs in the system is the product of the average cycle time and the average time a job spends in the system. Adding steps to accommodate an educational mission will increase cycle time. However, it may increase capacity at the same time. Consequently, the average number of jobs in the system may rise or fall. It depends on the relative values of the increase in cycle time traded off against the increase in capacity.
In addition, we also have to think about the patient experience in the modified system. Patients almost always prefer increased face-time with a physician. Having two doctors telling you that you are going to be ok may have a larger intellectual and emotional impact when compared to hearing the same message from one. In other cases, a patient speaks with the Resident and then remembers some point they wanted to raise by the time the Attending comes into the room. Having two contact points, instead of one may lead to the conveyance of more information both ways. This value is hard to measure, but surely is positive.
Take away
The bottom line is that the “Common sense” that the AMC must be less efficient is often wrong because it makes the resulting process both more complex, and more flexible at the same time. As a result its capacity and patient outcomes become more rich questions to be understood. More details can be found in the publications stored on our documents page and a full discussion is included in our book available at Amazon.com.