Empowering Healthcare with Data Science – Webinar Blog Post

Empowering Healthcare with Data Science – Webinar Blog Post

The processes for delivering healthcare are complex. There are many interconnected aspects—staffing challenges, clinical protocols, patient volume, the pressure to do more with less, and recent changes in legislation—that make applying analytics to healthcare especially impactful and, traditionally difficult. Analytics can help healthcare organizations make sense of all of these variables and make a genuine impact on care delivery. But, given all of these complexities, how does a healthcare organization tap into the true benefit of analyzing data for improved patient care and cost-efficient processes?

The answer lies in bringing healthcare-focused data science expertise together with an analytics platform that provides impressive number-crunching and data visualization power, while remaining accessible to users of every skill level.  This is where Pyramid Analytics and Proskriptive join forces. 

In healthcare, it’s critical to apply the right analytics to every phase of the patient’s journey. By doing this, organizations can see where best to target their efforts and what actions are needed to make each patient’s journey the most successful it can be.

That journey consists of pre-admission, inpatient, and post discharge stages. Understanding each stage, and how to prepare for it (or, in the case of inpatient, prevent it) directly supports the goals of cost-efficiency and optimal health outcomes. For example, in analyzing the wealth of data at their disposal, organizations can predict lengths of stay, inpatient safety risks (such as falls or infection), and even likelihood of readmission. If providers have these insights, they can deliver better, more targeted care, and may even be able prevent events that are high cost and unfavorable to patient health.

Proskriptive delivers three key benefits to their healthcare customers: a wealth of healthcare data from public and private sources, data science technology to connect to analytic tools, and the expert services needed to guide organizations in focusing their resources. Proskriptive also guides organizations on how to implement all of this into their current workflow for optimal success.

Pyramid Analytics’ BI Office analytics platform can make exceptional use of Proskriptive’s curated data and data science technology.  BI Office empowers end users, not just IT, to create data models on their own, so they can bring together what they need to analyze. Those models and all changes made to them are stored on a central server; they are never brought down to a device. This central governance supports compliance with HIPAA and other efforts to secure sensitive data. Once the model is created, analytic views and complex scenarios can be developed and analyzed right in the browser. That analysis can then be securely communicated across the organization through interactive storyboards (aka dashboards) and automated narrative reports. With BI Office, analytics isn’t just put to use by a few analysts. It is available to all individuals that need it—from clinicians to department managers to c-suite executives.

Figure 1: Proskriptive and Pyramid Analytics

Highlighting some of the key functionality that the partnership provides can be shown with an example. In a BI Office storyboard, end users can bring in large sets of data and, in just one click, filter the data to see specifically what’s of interest. For example, a dashboard may have a sizable list of diagnosis types by claims billed. A user can easily show the top 50 percent of diagnoses by claims billed, and then slice by different types of specialties, such as circulatory.  This is great to view historic information, but it isn’t predictive and doesn’t include the specific data science healthcare institutions may need. 

With the addition of Proskriptive’s data science expertise, a hospital facility can dig deeper and go further with their analysis to look at the risk level of specific patients and analyze predictive information about that risk. They can also include Impactability scores, the likelihood of impacting the patient for improved outcomes, to determine where to allocate care resources.  The integration of these data science enhancements within BI Office is seamless, as the analysis is created for users automatically (see Figure 2). Users don’t see the complexities within that analysis, but certainly benefit from the results.  By selecting different combinations of risk level, impactability level, and chronic status, care coordinators could quickly see where to focus their time—which patients are in the most need of assistance and have the highest likelihood of improved outcomes due to that intervention.

Figure 2: Analyzing Effects of Risk and Impactability

The Proskriptive / Pyramid Analytics integration can also make it very easy to view flags when important thresholds have been exceeded, such as risk levels in your current population (see Figure 3). With Pyramid Analytics’ BI Office, users can create data-driven text that dynamically updates as conditions change.  It can signal a warning and even prescribe detailed actions.  Having this type of data science expertise integrated directly into your secure analytics platform, improves decision making and empowers confident actions.


Figure 3: Using Dynamic Text Functionality to Guide Actions around Risk Levels

With the Proskriptive and Pyramid Analytics partnership, the pain points traditionally encountered conducting healthcare analytics can be eliminated. Organizations can count on Proskriptive’s data sources, data enrichment and data science capabilities, and view key insights that can be further analyzed on the manageable, scalable, and secure BI Office platform. These insights can be turned into actions that can help healthcare organizations provide better patient care through cost-efficient means.

To learn more about how Proskriptive and Pyramid Analytics can improve healthcare analytics, please contact:

Elsa MacDonald with Proskriptive at emacdonald@proskriptive.com

Got Something To Say:

Your email address will not be published. Required fields are marked *