Three Ways to Make the Most of Predictive Analytics

Share on facebook
Share on twitter
Share on linkedin
Share on email

By Callie Patel, Director, Innovation Consulting, Healthbox

Digital transformation is transitioning from a nice-to-have to a mission-critical element of any healthcare provider’s strategy. The HIMSS Digital Health Indicator (DHI) objectively measures a provider organization’s progress toward a digital health ecosystem that enables consumers to manage their health and wellness using digital tools—supported by connectivity with clinicians and provider teams—in a secure and private environment whenever and wherever care is needed. The DHI is the only comprehensive tool to assess and advise health system leaders on how to achieve outcomes using digital health, not just to use technologies or achieve other more specific measures. This blog series dives into each of the four dimensions of the DHI: governance and workforce, interoperability, predictive analytics, and person-enabled healthcare.

Right now, more than 12 billion devices are connected to the internet—and that number is expected to increase to 75 billion in the near future. With this influx of technology comes a surplus of data. To make the most of that data, it must be turned into actionable information. Leveraging predictive analytics in healthcare workflows can make that possible.  

Predictive analytics tracks and traces activities across the journey of care for individual patients, allowing providers to identify outcomes that work best for every individual—and the conditions under which those best outcomes are achieved.

Predictive analytics is incredibly important for informing decision-making in healthcare at the clinical, operational and financial level. HIMSS defines the three subdimensions of predictive analytics as predictive, personalized and operational. They should be used to shape evidence-based care pathways, real-time care decisions, capacity and facility planning, supply chain and logistics planning, workforce productivity and scheduling, equipment downtime and more. Leveraging large and varied data sets from the entire continuum of care is necessary to harness the real power of predictive analytics.

To prepare, successfully deploy and make the most of predictive analytics, organizations should have at least three priorities front and center.

Prioritize population health outcomes

Predictive analytics for healthcare at the population level seeks to improve the health outcomes of a group of individuals by combining historic and current subsets of patient data to forecast outcomes and trends to identify what yields positive outcomes and where costs can be optimized. Focusing on these robust analytics enables systems to learn and define the care delivery strategies that achieve best outcomes (highest quality) and the conditions under which those outcomes are achieved for every population.

Ochsner Health System (OHS), a system across the greater Gulf South Region, has been leveraging predictive analytics for several years. In 2017, OHS implemented an evidence-based immunization assessment as a required component of initial documentation for every admission. The assessment evaluates the history of a patient, specific to previous immunizations, and assesses if the patient is a candidate to receive tetanus, pneumococcal or influenza vaccination. The assessment is tracked, and if the answers indicate a vaccine is needed, a vaccine order is automatically generated. Staff adherence is tracked, and warnings are given for non-adherence. After implementing the assessment algorithm, the OHS influenza administration rate went from 88% to 98.5%. This is a simple yet impactful demonstration of care delivery shifts that can occur with the help of predictive analytics.

Many healthcare organizations have developed new prediction models and tools to better track and forecast COVID-19 cases. Dallas-based Parkland Center for Clinical Innovation created a COVID-19 Vulnerability Index to measure communities’ vulnerability to COVID-19 by tracking and analyzing socioeconomic, clinical, mobility and demographic risk factors. The model determined social deprivations—such as inadequate access to food, medicine, employment and transportation—as the largest contributor to higher COVID-19 mortality rates among Black and Latinx communities. Local healthcare providers are using the index as a tool to better tailor their COVID-19 response to the neighborhoods that need it most, deploy more testing and education in at-risk areas, and plan culturally sensitive initiatives to address infection disparities in Black and Latinx communities that have been disproportionally impacted by the pandemic.

Prioritize individual health outcomes

Predictive analytics for healthcare at the individual level uses historic population data and the individual patient’s real-time data to make a personalized care plan and predictions for the individual. In order to gain each patient’s real-time information, we must be able to ingest data from multiple sources—such as wearables, at-home sensors, mobile devices and more.

This priority put on individual health outcomes should also include proactively identifying individual patient risk at the point of care (e.g., patient deterioration in the ICU, allergy alert, medication error, recalled product, expired product). This precision in care delivery is enabled by automated tracking and traceability of outcomes for every individual.

In a healthcare ecosystem that continues to increase focus on patients as consumers of their own care, predictive analytics improves a provider’s information set and capabilities—giving them more than just expertise and experience to guide them in the decision-making process. This enables them to share decision-making with their patients through real and transparent conversations regarding individual health and wellness goals.

NorthShore University HealthSystem in the Chicagoland area has been leveraging predictive analytics for several years. Teams there created the Clinical Analytics Prediction Engine (CAPE), which assigns each patient a single risk score tied to multiple predictive models. By bringing together individual predictive models, CAPE creates a single platform and builds a framework of coordinated clinical interventions. For example, the system embeds specific, real-time actions in the EHR for physicians, nurses and case managers to execute, such as checking vitals more frequently and educating patients about specific goals of their treatment plans.

Predict impact of operational decisions proactively

Many health systems have accelerated their use of predictive analytics during the COVID-19 pandemic to:

-Protect employees

-Predict patient surges and/or severity

-Manage supply chain (e.g., monitor ventilator inventory)

-Inform other strategic, financial and operational considerations

Without the edge that predictive analytics provides, many health systems would not have been able to stabilize their operations in the impressive fashion that they have.

It is important to remember why the old adage “people, process, tools” has lasted. If predictive analytics solutions are implemented with only technology in mind, organizations won’t realize their true power. If predictive analytics solutions are making predictions at points in the workflow that don’t enable intervention—and/or are making predictions in systems that are different from where the interventions need to take place—we run into the issue of predictions with no action. For example, if during a patient visit, a provider is notified of a social influence that may impact that person’s health and care plan (e.g., food insecurity), yet doesn’t have a way to make a consult order to address that gap in real time, then the prediction becomes nearly useless.

As a result, organizations must simultaneously address the staffing (people) and infrastructure/resources (process) that may also need to be changed with the implementation of robust predictive analytics—so that when something is predicted to happen with high certainty, people will have the resources, know-how and ability to address it. Predictive analytics is the ideal tool to anticipate and manage challenges that inhibit better outcomes.

Ready, set, action!

So what aligns all three of these priorities? Action.

Now more than ever, health systems must find novel ways to improve the health and wellness of the communities and individuals they serve. Unlocking the potential of predictive analytics in healthcare will enable this lofty aim. Whether it is in driving new person-enabled care pathways, better predicting equipment maintenance needs before they arise, or making scheduling decisions in an agile way, predictive analytics makes transformation possible from today’s reactive health system toward a digital health system focused on the future.

Healthbox, a HIMSS Solution and healthcare advisory firm, drives innovation from the inside and out, helping organizations build internal innovation programs, assess the potential of employee-led projects, and look to the market to find solutions to implement or invest in.

Identify your organization’s strengths and opportunities to inform a comprehensive digital health strategy with the complimentary HIMSS Digital Health Indicator (DHI) Virtual Assessment.

For more information on how to take action with predictive analytics, read about the HIMSS Analytics Adoption Model for Analytics Maturity (AMAM) and how it can serve as a guide to measure and advance your organization’s analytics capabilities.

Newsletter

Subscribe to our monthly newsletter to receive Healthbox insights on healthcare innovation.

  • This field is for validation purposes and should be left unchanged.

Recent Posts

Search

Categories

Archives