Healthcare systems and designers are using computational analysis to solve challenging medical planning and design issues.
Big data and predictive analytics are not just for companies like Amazon. For instance, the healthcare industry has been mining data to better predict trends in their patient populations and analyze potential growth in specialty services for years. However, the use of predictive analytics and computational thinking is relatively new for healthcare medical planning and design. Using computational thinking addresses complex issues regarding operational efficiencies, right-sizing departments or even an entire replacement hospital and frontloading the planning process with the optimal business outcomes to arrive at the right solutions sooner.
Computational thinking is a problem-solving methodology that’s broken into four components: Decomposition, pattern recognition, pattern abstraction, and algorithm design.
Decomposition is all about breaking down a big problem into a set of smaller components. Once broken down, patterns begin to emerge. Through abstraction and modeling, the data can be visualized, and possibilities tested to find the optimal answer.
How does that apply to the design process?
Traditionally, design starts with user input and programming, in which qualitative data informs quantitative data. Using computational thinking, it is possible to look at that data through a neutral lens. For example, nurses and caregivers often spend time traveling from the nurse station to the patient room and subsequent support spaces, taking time away from the patient’s bedside. In many cases, nurses and caregivers create workarounds to avoid taking multiple trips, causing undue stress from multi-tasking. This issue is directly impacted by the physical environment.
When redesigning an inpatient unit for Campbell County Medical Center in Gillette, Wyoming, HGA researched nurse and caregiver travel within several units, in addition to conducting user group meetings and programming analysis. Through extensive time-and-motion studies on-site, the design team collected data on the actual distance and frequency of each trip a nurse made during a shift. This was then modeled using computational design tools, resulting in a clear visualization of travel patterns that told a much different story about the current state.
The floor plans below reflect the patterns that began to emerge.
This information gave the design team and nursing staff an opportunity to better understand the impact the current physical layout was having on their efficiencies. The data then provided the foundation for the team to explore several options for the new design, including centralized vs. decentralized charting stations, and single large support spaces vs. multiple smaller support rooms. Ultimately, the team decided to de-centralize the charting stations based upon the research and computational thinking.
The potential results in this case is threefold: nurses and caregivers will spend less time traveling and spend more time with the patient; the physical demands of the nurses and caregivers will be diminished; and more efficient coverage will offer potential to staff the floor differently, saving valuable FTEs.
Predictive analysis is influencing the design of a large variety of healthcare environments, specifically to right-size a department or facility. Using data from an outpatient clinic housing a variety of specialty services, HGA built a dashboard to analyze and visualize a variety of drivers to better inform the design solution. In the dashboard (top of page), the team can “pull” several levers for evaluation, reflected against baseline data to study the impact of each, such as department gross square foot (DGSF)/Driver spaces, visits per year, average cycle time, target utilization of an exam room, and additional service lines. The dashboard can then employ several filters for these drivers, such as projected growth by service line, location, department, and the demand on physical space down to the provider.
Breaking down the complexities of the healthcare environment into recognizable patterns gives planners and designers more information precisely when they need it—during the design and planning process—to solve a healthcare system’s biggest challenges. To put it simply, predictive analytics is about driving the business. And that is exciting. Move over Amazon.
Aaron Kapphahn is a Medical Planner at HGA, where he specializes in digital and computational design tools.