The power of real-time, predictive analytics for improving patient care

The proliferation of EMR systems in hospitals has provided clinicians and administrators with more data and information than ever before. At the same time, managing this deluge of data and extracting value from it is often cited as one of the key challenges facing leaders. And when it comes to using EMR information to track, manage and prevent disease, raw data alone falls short.

"Large quantities of data are only as meaningful as the insights they yield," said Adam Klass, chief technology officer at VigiLanz, during a May 19 webinar sponsored by VigiLanz and Becker's Hospital Review. Efficient data analytics, which derive actionable insights from medical data, are linked to significantly reduced costs, increased business intelligence and improved clinical outcomes, Mr. Klass added.

Each of these imperatives is central to health systems' efforts as they traverse from the world of fee-for-service reimbursement models to those based on value, as clinical care and operational efficiency become more closely tied to reimbursement.

VigiLanz, a healthcare technology and analytics SaaS company based in Minneapolis and Chicago, has developed a clinical and business platform suite with real-time, predictive analytics and modeling capabilities. Its Enterprise Intelligence Resource platform (EIR), a layer that that sits on top of the EMR, creates actionable insights from a wide variety of disparate medical data sources to handle clinical situations as they occur, and is compatible with all EMR providers.

VigiLanz' EIR platform pulls information from EMRs transactional workflow and documentation data in real time and standardizes and normalizes that data as it comes in, 24/7. The EIR platform aggregates historical medical data from a variety of disparate databases across health systems and performs asynchronous surveillance to alert clinicians of possible issues, including exception based notifications. It also uses machine learning to constantly improve specificity, and can be tailored to an organization's own patient population and best-practices.

These real-time analytics capabilities can be deployed to monitor a variety of initiatives, such as antimicrobial stewardship, deep vein thrombosis prevention, glycemic control, fall prevention, readmissions, infection control and sepsis.

While real-time analytics capabilities are valuable in many different elements of healthcare, its use in combating sepsis — for which early intervention is critical to effective treatment — is particularly promising.

In 2009, sepsis was the sixth most common principal diagnosis for hospitalization in the U.S., accounting for 836,000 stays or 2.1 percent of all hospitalizations, according to the most recent national discharge data reported by the Agency for Healthcare Research and Quality.

The need to improve processes that identify patients at risk for developing sepsis and provide the earliest intervention possible is clear. However, the general model for identifying sepsis — defined as systemic inflammatory response syndrome — only indicates the onset of sepsis after the patient may have already developed it. This is clearly not an effective means of prevention and does not position clinicians to intervene as effectively as possible.

Fortunately, advances in tools that perform predictive analytics offer healthcare organizations the opportunity to identify sepsis at the earliest possible sign, and even prevent it from occurring at all.

VigiLanz's analytics methodology, termed Temporalytics, enables clinicians to identify the optimal time to deliver care by its ability to analyze the outcomes of varied response times to clinical issues. Temporalytics uses advanced statistical models and machine learning to analyze data related to the timing of clinicians' actions and compare it to the response targets set in the VigiLanz system. The real-time analysis predicts scenarios where a change in the optimal response time or reduction in variability around an existing best practice can improve patient care, and ultimately improve the hospital's bottom line.  

"This is unique to VigiLanz," said Bart Abban, chief data scientist at VigiLanz. "This capability is made possible by the time stamps we collect."

Using VigiLanz's data-warehouse of real-time and historical data and its Temporalytics methodology, the company builds cohorts of a hospital's entire patient population to identify those most vulnerable to developing sepsis. Based on the time-stamped metadata, VigiLanz can perform retrospective analysis and essentially timeline when patients of a particular population began developing sepsis in the past, and how they reacted to interventions at different stages of the infection's progression. Based on this information, VigiLanz's predictive model identifies patients at high risk of developing sepsis and alerts clinicians to administer interventions at the earliest possible time. 

Unlike the SIRS criteria, which only alerts clinicians that a patient may be septic after they have already begun showing signs, VigiLanz's Temporalytics methodology is both highly sensitive and highly specific. This allows clinicians to focus interventions on those patients who are most likely to develop sepsis, as opposed to treating a wide population, many of which would not actually get the infection. The cost-savings from this model are significant, with the possibility of saving tens of thousands of dollars per case and millions per year.

The need to fight sepsis is clear, from the patient safety standpoint, as well as the cost perspective. Traditional tools consistently fall short, as rates of sepsis have continued to rise. However, with the promise of predictive analytics, hospitals can be empowered to stop the deadly syndrome before it starts.

To view the webinar on YouTube, click here.

To download the presentation, click here

© Copyright ASC COMMUNICATIONS 2017. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.

 


IC Database-3

Top 40 Articles from the Past 6 Months