Operationalizing Prediction Situation
As healthcare organizations continue to implement new and advanced predictive analytics throughout their systems, identifying potential at-risk patients - sepsis patients, or patients with high potential for readmission, for example - becomes standard procedure; operationalizing the data these analytic tools discover now becomes the focus.
Prediction is a substantial advancement in healthcare operations. However, predicting patient status by itself is not impactful, nor worthwhile, without an associated operational protocol that intervenes based on that prediction. Knowing you may have ten patients at risk for sepsis is a huge leap forward, now how do I ensure these patients remain safe and healthy? Someone must do something, in the appropriate order, in a prescribed time-frame. Is this being done?Background
All too often, mining data and generating 'output' is touted as the end-game to complex issues, such as sepsis identification and prediction. The problem is, once a prediction is generated, someone, or something, still needs to intervene. If the workflow for intervention is the same as the workflow before the implementation of a predictive model, it's unlikely you'll realize the full value of the model.
When complementing the implementation of predictive models, it is imperative to exercise operational changes that include a centralized system to visualize each process step, calling someone to action, or intervening on the clinician's or patient's behalf.
Situational Awareness as a Solution
In the coming years, we will witness a change in how we consume, process, and apply data. We are just now moving into predictive analytics. However, we still face the same problems we had yesterday – how, where, and when to alter workflows to treat a patient best.
This transformation begins with Situational Awareness.
Situational Awareness is the ability to identify, process, and comprehend information to choose the next step or make the next decision. It is the perception, conception, and reaction to environmental factors to influence outcomes. More simply, it knows what is going on around you.
While the concept of SA has been around for a while, we are again on the edge of a new movement where the maturing information economy is leading the way to cognitive age, driven by Situational Awareness and its complementing technologies.
Artificial Intelligence is the complementing technologies. AI identifies the problems, like predictions, and determines the required workflow and protocol changes, and then apply the intermediate actions. Healthcare delivery will change. If the workflow protocols discretely call out the next steps in the patient's pathway, and the relevant patient information is known (such as allergies, contraindications), AI can take action autonomously and quicker than an individual. This autonomy doesn't imply that AI is practicing medicine. Many steps within a pathway require licensed, experienced staff input and oversight. Artificial Intelligence and its sister, Robotic Process Automation (RPA), can replace the mundane, routine, protocol-based tasks. These removals ultimately free up providers of all ranks and types to provide care at the top of their licenses.
Edgility's Outcomes Now® Platform is the country's first healthcare specific "Outcomes-as-a-Service" model where physicians, nurses, pharmacists, and other care providers from any location are provided concurrent quality and outcomes monitoring and support. From core measures to financial metrics and patient itinerary to throughput, Edgility provides a situationally-aware, air-traffic-control like, command and operations center to drive outcomes for health systems.
Edgility's ultimate goal is to enable technologies (current and future), so they are integrated and focused on superior patient experiences and outcomes while maximizing current investments. In the AI and RPA space, Edgility's cognitive intelligence platform allows knowledge workers and other high-value assets to focus on creativity and outcomes that utilize reliable robotic processing to perform the repetitive and support tasks to boost overall productivity.