Good System Design in Healthcare

Project details

  • Date

     May 7, 2019

Good System Design in Healthcare

In our last blog, entitled, Lean: The Hidden Method to Achieving Success in Value-Based Care, we discussed the shift in healthcare reimbursement from pay-for-service to value-based care and how Lean is a method that will allow healthcare organizations to thrive in such an environment. In this article we take a deeper dive into what good system design looks like in healthcare.

The Consequences of Fragmented Healthcare

Consider the following scenario. Your five-year-old daughter has an asthma attack after dinner, and you rush to the closest emergency room in a panic. You arrive to, what many patients have become accustomed to, a crowded waiting room and are made to wait in a line to be assessed and registered. Though your daughter clearly has a medical emergency and her breathing continues to deteriorate, there are patients ahead of her with more severe conditions, and she cannot be processed immediately. So, you sit and wait.

Hours pass before your daughter is placed in a room where a nurse makes an initial assessment. Again, you and your daughter are forced to wait until the doctor is ready. This takes an additional hour. The doctor finally arrives, looking puzzled from his last patient experience, and routinely repeats many of the same questions previously asked. Finally, the doctor reaches a diagnosis, three hours after you checked in, and your daughter is administered proper treatment for her asthma. At this point, the doctor writes a script for an inhaler which you must pick up from the pharmacy when it opens the next day. We can all agree that this patient journey is anything but optimal and, from a parent(s) perspective, creates a tremendous amount of emotional stress. However, it is critical to understand why these scenarios exist from a system view.

In most hospitals across the United States, patients who have received treatment and are ready to go home, a process referred to as being discharged, wait and occupy beds in units. Analyses have shown that most patients are discharged on Thursday afternoons. There are many root causes to this, but most of the time it happens because the required information to discharge a patient is batched towards the end of the day. If patient care could be better coordinated between hospitalists, nurses, case managers, environmental services and transportation, etc. beds would free up faster. When they aren’t, patients from the emergency room who need to be transported to beds are made to wait, and patients, like your five-year-old daughter, are also made to wait. Even worse, some patients who suffer from more severe conditions, are turned away or die1.

So why does this happen? Are the workers bad or lazy? No. In nearly all of our engagements, we find that the teams and individuals we engage with truly want to help the patients that come through their doors and are highly driven to do so. According to W. Edwards Deming, A bad system will beat a good person every time. In fact, Deming stated that 94% of errors are related to poor system design.2 Therefore, it’s not the people that are the issue; it is the design of the different systems in play. So, what does good system design look like?

Good System Design

Good systems don’t just happen; they are thoughtfully designed. A system that is built to deliver superior patient care has the following five key attributes:

  1. Organizational success is clearly defined using a balanced scorecard approach, i.e. IHI Triple Aim
  2. The end-to-end (i.e., value stream and patient journey) is at the center of the design and ties back to organizational success
  3. A production system that incorporates lean principles of JIT (just-in-time) is designed across the value stream
  4. A management system is designed to manage the value stream
  5. PDSA (Plan Do Study Act) is continuously deployed to remove waste across the value stream

Prior to designing a system, a determination of what the system needs to achieve must be articulated. Many healthcare organizations use the IHI triple aim3 which seeks to Improve the patient experience of care (including quality and satisfaction), improve the health of populations, and reduce the per capita cost of health care. To design sustainable systems, healthcare organizations must continuously measure how they perform to these measures. Also, systems should be constantly evolving and improving through the creation and achievement of new and ambitious targets/goals that take into consideration patient expectations and market conditions (e.g., what are your competitors doing).

Once this is understood, it is vital to identify the various patient journeys that exist; typically, by service lines, (e.g., Medicine; Surgery; Women and Children; Oncology) and by complexity (e.g., low risk, medium risk, high risk). Patients who have similar journeys can be grouped into a single value stream family. In other words, improvements to one patient journey can be applied to the other. Once this is done, it is essential to identify which value streams families have the most significant impact on organizational success. For example, if we are looking at improving patient experience based on survey data, reducing readmittance, and reducing the length of stay, we need to understand which value streams are currently impacting these numbers the most. The ideal situation is to focus on a value stream family with the highest volume yet worst performance (the most opportunity!).

At this point, we have a focus and it’s time to use value stream mapping as a tool to:

  1. understand how the value stream currently performs
  2. create an ambitious future state of how the value stream needs to perform to achieve organizational success

A current state value stream map will highlight all the waste in the current state; commonly referred to in Lean as muda or the 8 deadly wastes (Defects, overproduction, waiting, non-utilized talent, transportation, inventory, motion, and extra processing). A future state map will guide you in removing this waste and ensure customer flow (commonly referred to as takt time) can be achieved. With this future state blueprint in hand, processes within the value stream (i.e., triage, room patient, rounding) need to develop standard work (i.e., the best-known way of doing the work now that will evolve and be improved upon with time). Each of the processes standard work must connect. For instance, if doctors, nurses, transportation, and environmental services are to each play a part in patient discharging, each must have standard work that ensures they can act in a coordinated fashion, without the waste that originates from operating in silos, to achieve takt.

Once standard work has been designed to achieve the future state, we need a way to manage towards the blueprint. There are three key elements in a sustainable lean management system:

  1. visual management
  2. accountability
  3. leader standard work

Visual management allows us to see, know and act as a group. In other words, it shows us how we are performing in reference to the established key metrics, and it makes problems in the process visible. Huddles or stand-ups are typically used to gather around the visual management boards and discuss identified problems and appropriate countermeasures. Leader Standard Work should be used by management to ensure the first two happen with discipline. This should happen at the organization level with the organizational success metrics, the value stream level with value stream success metrics that connect to the organization success metrics, and at the process level with process success metrics that connect to the value stream success metrics.

Reoccurring problems, typically captured using Pareto Analysis, should be used to drive PDSA. Root-cause analysis should be done by going to the gemba (the place where the work happens), experiments should be planned, executed, studied, adjusted and/or standardized when achieving the desired outcomes. This iteration will allow teams to refine their standard work and move towards making the future state value stream map a reality. The key here is to fail fast.

As you can see, creating good system design in healthcare is no small feat and goes beyond small scale deployment of Lean tools such as 5S and the creation of isolated tools such as a huddle. Bringing it all together requires a seasoned Lean practitioner, i.e. a Sensei, that understands how to properly install, activate, and sustain complex systems like the one described above. We have seen countless system initiatives fail due to poor system design and implementation. If the proper foundation is not created, major system issues will ensue including issues with patient safety and care, team morale, organizational financial performance, and overall market competitiveness.

If you would like to discuss what a good system design and installation would look like for your organization, please reach out. We would be happy to discuss the obstacles and parallel opportunities that exist for your teams and overall organization.