A large U.S. healthcare system

Improving healthcare outcomes with big data analytics

Published on 01-Nov-2013

"The health system knew that combining subject matter expertise with rapid data analysis could truly transform the practice of medicine." - —Dr. Richard H. Goldstein, Chief Executive Officer, RGI Informatics

Customer:
A large U.S. healthcare system

Industry:
Healthcare

Deployment country:
United States

Solution:
PureData System for Analytics (powered by Netezza technology), PureSystems, Big Data, Big Data & Analytics, Big Data & Analytics: Operations/Fraud/Threats

Overview

As one major health system began to expand use of RGI’s Healthcare Analytics Solution across the United States, RGI found that the conventional servers and storage technology they previously used were too slow to process the growing amount of data. Using IBM® PureData™ System, powered by Netezza® technology, RGI gained the power and performance to sustain the iterative environment demanded by clinical users. This is helping doctors, nurses and other clinicians at the healthcare system rapidly analyze and identify patterns across millions of rows of data from disparate systems.

Business need:
A large U.S. health system needed to harness healthcare’s “big data” and analyze a complex set of data, including electronic medical records and sensor data.

Solution:
By combining RGI™ Healthcare Analytics Solution® with the IBM® PureData™ System for Analytics, powered by Netezza® technology, the organization has taken the first step in unlocking the potential of big data.

Benefits:
Reduces data load times by an average factor of 8 - 12 times and improves response times for complex queries by 70 - 350 times; enables clinicians to access and analyze healthcare big data to ascertain quality, determine best practice, assess treatment strategies and identify patients at risk.

Case Study

As healthcare organizations embrace electronic health records and other clinical information systems, they have an unprecedented opportunity to learn from their data—but only if they can effectively analyze it.

A solution deployed at one major U.S. health system is making it faster and easier for physicians to uncover vital insights. This powerful platform can rapidly analyze and identify patterns across millions of rows of data from disparate systems.

“The health system knew that combining subject matter expertise with rapid data analysis could truly transform the practice of medicine,” says Dr. Richard H. Goldstein, CEO of RGI Informatics, the healthcare analytics provider that developed the analytics platform.

Unleashing the power of big data

This large U.S. health system needed a way to harness healthcare’s “big data” and realize the value of analyzing a complex set of data: the electronic medical records and sensor data from devices monitoring critical care patients and patients undergoing anesthesia.

Could smart analysis improve patient care? Could it identify trends in outcomes that translate into optimized treatment plans?

In 2004, this healthcare system became the first to implement RGI’s Healthcare Analytics Solution. RGI’s solution allowed the healthcare system to store, analyze, and act on physiologic sensor data at a detailed level, pulling data from across departments and diverse monitoring devices and making meaningful use of electronic medical records (EMRs).

The RGI analytics gathers and analyzes un-modeled and non-standardized healthcare data from disparate applications, including electronic medical records, specialty systems, operating room and intensive care units, clinical information systems, and sensor data from medical devices. Because the RGI software user interface is web based, patient data remains safely on hospital servers. The highly visual interface is easy to use and understand, delivering analytics that allow clinicians to unlock the potential of “big data.”

The health system began to expand implementation of the RGI Healthcare Analytics Solution across the United States. However, as the amount of data grew and additional facilities and data sources came on line, RGI found that conventional servers and storage technology were too slow. Nightly data loads took six to eight hours to complete, and analysis of sensor data could take up to three days.

RGI’s new dynamic user interface required faster database response times to maintain an iterative environment. RGI needed enhanced computational power to optimally deliver and scale its analytics solution to the healthcare system.

IBM PureSystems™ offering earns high marks for performance and ease of use

To identify the best technology to meet the need for faster computing, RGI explored a number of advanced options, including Massively Parallel Processing (MPP) systems, and MapReduce and Apache Hadoop frameworks. According to Dr. Goldstein, an MPP system seemed like the best fit.

RGI then retained Sandia National Laboratories to compare software-based MPP systems and hardware-based MPP systems, including the IBM PureData System for Analytics, powered by Netezza technology. The IBM platform was the clear winner for problems of interest, earning high marks for superior performance and ease of use.

“We like the IBM PureData System for Analytics over other products because of the speed at which it can process clinician queries against ‘dense’ data, including time-series data,” Dr. Goldstein says. “For example, a traditional query for patients with a glucose reading over a certain level may reveal which patients are diabetic, but can’t indicate how well they were treated. With the RGI Healthcare Analytics Solution and the speed of the massively parallel IBM PureData System database, doctors can analyze time-series data fast enough to see how long patient glucose levels were out of range—enabling them to determine best practice and improve the quality of care.”

Faster processing leads to faster insights

To demonstrate the benefits of this approach, RGI successfully implemented a massively parallel Healthcare Analytics Solution with the IBM PureData System for Analytics, powered by Netezza technology, in five of the healthcare system’s medical centers.

“The IBM PureData System for Analytics is powerful and extremely fast,” says Dr. Goldstein. “Data loads that once took six to eight hours are now done in 30 - 60 minutes and query response times for complex queries have improved by 70 - 350 times compared to conventional platforms.”

The system maintains performance despite multiple simultaneous users, and sustains the iterative environment demanded by clinical users. This means that doctors, nurses and other clinicians can access healthcare big data and use the insights gleaned from their analysis to ascertain quality, determine best practice, assess treatment strategies and develop predictive algorithms to identify patients at risk. As a next step, these algorithms can be deployed in near real time to provide early identification and treatment of serious clinical conditions.

Combining clinician expertise with analytics leads to new insights

The health system’s first use cases for big data analytics have focused on improving inpatient clinical quality metrics, such as reducing the percentage of hospital-acquired infections.

“There are well established prevention protocols to reduce the incidence of hospital-acquired infections,” says Dr. Goldstein. “However, once the RGI analytics was in place, our client learned that the most consistently followed protocols were with very sick patients who had often already developed a complication. This was a major insight for us: hospitals were overlooking patients at risk for hospital-acquired infections who had the most to gain from established protocols—those without complications. Armed with this insight, the hospitals extended the protocols, which, along with other efforts, helped decrease their infection rates and make them the health system leader.”

There’s also an opportunity to use physiological data to predict and to identify patients at risk. There are billions of data points from electronic medical devices and sensors—heart rate monitors, ventilators, blood pressure monitors and so on—captured during a patient’s hospital stay. Big data analytics can reveal patterns hidden in this data, alerting clinicians to worrisome trends.

“We can develop algorithms that identify a patient at risk and alert the physician, which can result in earlier treatment of serious clinical conditions,” says Dr. Goldstein.

Rapid results—increasingly close to the point of care

Specialized clinical information systems for use with critical care patients and during anesthesia collect and store patient assessment information, which is transmitted nightly to RGI and then immediately available for analysis. When this information is used in quality-of-care improvement initiatives, the RGI analytics eliminates the need for manual data abstraction and reporting. The process is fully automated from the time of initial data entry to reporting to systemwide quality improvement databases. Results are available within 24 hours of data entry, compared to three to four months for manual data collection and entry.


Simplifying platform migration
The PureData System for Analytics, powered by Netezza technology, is ready to use right out of the box, making it easy to transition to a massively parallel database. Hardware and software capabilities are designed, integrated and tuned for high-performance analytics. Standard interfaces simplify set-up and administration.
“The IBM system is easy to deploy and is much more efficient for working with big data than comparable products,” Dr. Goldstein says. “Once fully deployed, we expect to achieve a five-to-one reduction in the number of servers we need.”

Legal Information

© Copyright IBM Corporation 2013 & © Copyright RGI Informatics 2013 IBM Corporation Software Group Route 100 Somers, NY 10589 Produced in the United States of America November 2013 IBM, the IBM logo, ibm.com, PureData, and PureSystems are trademarks of International Business Machines Corp., registered in many jurisdictions worldwide. Other product and service names might be trademarks of IBM or other companies. A current list of IBM trademarks is available on the web at “Copyright and trademark information” at ibm.com/legal/copytrade.shtml Netezza is a registered trademark of IBM International Group B.V., an IBM Company. RGI and Healthcare Analytics Solution are trademarks for RGI Informatics. This document is current as of the initial date of publication and may be changed by IBM at any time. Not all offerings are available in every country in which IBM operates. The performance data and client examples cited are presented for illustrative purposes only. Actual performance results may vary depending on specific configurations and operating conditions. It is the user’s responsibility to evaluate and verify the operation of any other products or programs with IBM products and programs. THE INFORMATION IN THIS DOCUMENT IS PROVIDED “AS IS” WITHOUT ANY WARRANTY, EXPRESS OR IMPLIED, INCLUDING WITHOUT ANY WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND ANY WARRANTY OR CONDITION OF NON-INFRINGEMENT. IBM products are warranted according to the terms and conditions of the agreements under which they are provided. RGI Healthcare Analytics Solution is not an IBM product or offering. RGI Healthcare Analytics Solution is sold or licensed, as the case may be, to users under RGI Informatics’ terms and conditions, which are provided with the product or offering. Availability, and any and all warranties, services and support for RGI Healthcare Analytics Solution is the direct responsibility of, and is provided directly to users by, RGI Informatics. The client is responsible for ensuring compliance with laws and regulations applicable to it. IBM does not provide legal advice or represent or warrant that its services or products will ensure that the client is in compliance with any law or regulation. WAC12387-USEN-00