The world is overwhelmed by uncertainty. It can be paralyzing. Healthcare systems are especially feeling this burden. Twenty COVID-19 patients admitted one night, none the next. Under “normal” circumstances, hospitals, clinics, urgent care facilities, and emergency departments face challenges in preparing for fluctuating patient populations from day to day. During a prolonged and unpredictable crisis, how do these facilities prepare for the surge in terms of the four S’s: staff, supplies, space, and systems?
During these times, healthcare resources are precious. Healthcare personnel are overworked and spread thin. Life-saving and protective equipment like ventilators and face masks are in high demand and difficult to acquire. Beds and rooms (especially negative pressure ones) are scarce. Infrastructure and technology are overextended. If a facility could know the number of COVID patients to expect on a given day, it could help them allocate their resources appropriately, saving money and ultimately, lives.
There is a vast wealth of data and tools out there aimed at helping hospital administrators and other policy and decision-makers monitor this crisis. But, how does one sift through and select the metrics that are going to enable their organization to best prepare for future surges?
One tool that shrewd administrators and decision-makers are leaning on right now is capacity modeling. But, for those of us who have experience with this type of modeling, we know that the quality of the forecast depends highly on the quality of the inputs. I have done this type of modeling for the federal government, specifically HHS. We did modeling for various strategic scenarios, all of them in the context of CBRNE (chemical, biological, radiological, nuclear, and explosive devices). It was the job of my team and other subject matter experts (SMEs) to determine if we had enough staff, supplies, space, and systems to respond to these strategic scenarios from a healthcare perspective. Assumptions had to be made and the logic behind the model fine-tuned with discerning eyes. At the end of this project, we were highly confident that we had come up with a model that considered all the nuances of these CBRNE attacks because of we had the collective expertise.
At Whitespace, we have taken a very similar approach, bearing on considerable expertise from the fields of mathematics, epidemiology, information technology, clinical medicine, intelligence, geospatial intelligence, defense, quality assurance/control, and research. Our team has meticulously documented every decision in our process. We have vetted and verified these decisions and our end products with the appropriate SMEs (academics, researches, etc.). We are highly confident that putting our metrics through such a rigorous process, makes them superior to other metrics and analytics.
Whitespace has developed the only analytics that produce an indicator of future disease transmission and risk: close contact. In the case of trying to predict the impact of surges COVID-19 cases (or other infectious diseases), it makes sense for a transmission model, or a compartmental model (e.g., a SIR model) to inform predictions about hospital capacity. Our analytics produce a unique and game-changing ingredient for these models. They produce the highest quality reporting on the rate of close contact across regions each day and are designed to improve the accuracy and precision of the forecasts of future COVID-19 cases at a hyper-local level. These are invaluable predictions that hospital administrators and other decision-makers can trust.