Human beings are naturally drawn to patterns and routines, especially those that make us comfortable. It is probably related to our distant ancestors who, surrounded by all kinds of threats, came to value predictability as a sign of survival. Of course there are many people out there whose restless nature drives them to break these routines and explore the unknown, but in general there is this tendency, this inertia if you will, to fall back into known territory.
And what is the point of this cheap (and somewhat obvious) anthropological reflection? What does this have to do with healthcare, or anything process mining for that matter? To answer this, we have to refer back to a recent project where we analysed some hospital data describing the discharge process from the ICU, and where we came across some interesting insights. The scope of the analysis spanned the steps of the process from the moment the doctor logs the discharge request to the moment the patient is moved to a room. That is, in order to minimize any medical considerations, we start the process the moment the doctor deems the patient fit for discharge and files the corresponding request.
The first thing that struck us was how long the process took. On average (measured by the median) it took more than 8,5 hours to complete the process, with over 33% of cases taking more than 9 hours. At first we assumed it was because probably there was a bottleneck in room reservations due to saturation. Interestingly, the largest delay in the process didn’t occur in the room reservation phase but earlier in the process, between the submission and the acknowledgement of the request. Additionally, even when the room had been reserved, in 13% of the cases it took more than 5 hours to complete the discharge.
As it is always a good practice to review and share our findings with the people that are actually involved in running the process, we had an eye opening conversation with an experienced doctor. This is where the story of the routines emerges. What we understood from the talk with the doctor was that the discharge process is deeply embedded into the rhythms of hospital life. Since patients are well taken care of in the ICU (except when there is pressure due to high occupancy) there is no real urgency in moving them out. So the process follows the routine patterns of the hospital, determined by things like the timing of doctor rounds, coffee or lunch breaks, shift changes, and in general, the habits of doctors, nurses, and admin personnel. This is the explanation for why although the majority of discharge requests are filed in the morning shift, most of the actual discharges occur in the afternoon shift. It also accounts for why a significant number of patients are not discharged until the following day (presumably because they miss the afternoon discharge window dictated by the routine), and end up spending an extra night at the ICU.
But then, what about efficiency? Well, it seems that (except in certain periods of saturation) this is not a concept that ranks high in the minds of practitioners. The term seems to have acquired a questionable reputation, since it is generally associated with cutthroat cost-cutting, often seen as opposed to the proper care and wellbeing of the patients. Spending an extra night in the ICU is not going to harm a patient, although it will impact his insurance company who will have to pay a premium for that extra night.
Coming from a process re-engineering background, trained to look for ways to improve and streamline processes, our perspective couldn’t be more different. We would argue that efficiency is not really about cost cutting (or not necessarily), it is about reducing waste and getting the most out of the available resources and we are convinced that ultimately, this leads to improved care for the patients.
This conversation with the doctor made us realise that there is a need to explain and promote efficiency within the medical community to help modify some of the underlying patterns so ingrained in day to day practice. The goal is to place the concept high in doctor’s minds, so that it is understood as a key element in delivering better care. And for this Process Mining can be a strong ally, allowing practitioners and administrators to discover the effects of how processes are executed and to have relevant conversations based on shared and objective data.
If you want to understand how the ebbs and flows of hospital life affect the efficiency of your organization, why not give us a call?
Check some of our videos:
- New ➽ ICU discharges: where time slips away
- Fast Track vs Standard Path: Which is Better for Diagnosing Colorectal Cancer
- What causes discards in parenteral nutrition preparation? Key factors revealed!
- How Process Mining Uncovers Medication Administration Issues
The Prompter.io is our open project to share our experience—and that of others—in integrating language models and data-to-text techniques into process intelligence. Don’t miss it!
Also here are some links of interest:
🔍 Can Data Science Strengthen Clinical Risk Assessment?
