Healthcare Dives Into Big Data
Healthcare organizations hope big data and analytics projects can help reduce costs and improve care. Consider these innovative examples.
![](https://eu-images.contentstack.com/v3/assets/blt69509c9116440be8/blteb3ce118bc1d2025/64cb5752164d6345daf4cb1c/BigDataCover.jpg?width=700&auto=webp&quality=80&disable=upscale)
With the mandated adoption of electronic health records (EHRs), many healthcare professionals for the first time got centralized access to patient records. Now they're figuring out how to use all this information. Although the healthcare industry has been slow to delve into big data, that might be about to change. At stake: not only money saved from more efficient use of information, but also new research and treatments -- and that's just the beginning.
For instance, data from wireless, wearable devices such as FitBits is expected to eventually flood providers and insurers; by 2019, spending on wearables-data collection will reach $52 million, according to ABI Research. Another source of health data waiting to be analyzed: social media. Monitoring what people post can help fight insurance fraud and improve customer service.
These are just two ways big data can be used to improve care while cutting costs, experts say.
"We, as a society, need to start creating our own metrics for how healthcare quality is defined. In the sense of looking at costs, we know where there's avoidable cost in healthcare. We just need to get folks the data they need to avoid those pitfalls," said Dr. Anil Jain, senior VP and chief medical officer at Explorys, in an interview. Explorys, which is an innovation spinoff from Cleveland Clinic, is powering Accenture's Predictive Health Intelligence in a collaboration intended to help life sciences companies determine the combination of treatments and services that can lead to better patient, provider, and economic outcomes for diabetics.
Hosted analytics, partnerships and collaborations, and lower-cost internal applications open the door for smaller organizations to use big data, too.
"Earlier, data warehousing and analytics was restricted to larger organizations because it was cost prohibitive. What big data has done has brought it down to smaller orgs. But the biggest challenge with these smaller markets and mid-tier organizations is resources," Manmeet Singh, co-founder and CEO of Dataguise, told us. "Cloud is becoming very prevalent. They're going to store a lot of data in the cloud. They'll outsource a lot of that data to the cloud. Automation of compliance is important."
Having witnessed the impact that big data and analytics have on other markets -- and perhaps on competing healthcare organizations -- healthcare CEOs want to know how their organizations can use these tools. In a PwC study, 95% of healthcare CEOs said they were exploring better ways to harness and manage big data.
Increasingly, CIOs can find similar organizations with pilot or full-blown projects. Forest Laboratories, for example, is collaborating with ConvergeHealth by Deloitte and Intermountain Healthcare on research to benefit patients with respiratory diseases. Using the collaborative, rapid-learning system developed by Intermountain and ConvergeHealth, Forest's researchers use OutcomesMiner analytics software to develop new treatments and therapeutic products and improve patient outcomes.
The move to value-based payments means healthcare providers are taking on more risk, says Jeff Elton, managing director of Life Sciences for Accenture. To manage risk and treat patients most appropriately, providers need data -- accurate data from a range of sources, he tells us.
Expanding use of big data across healthcare organizations should sound some alarms within C-level suites, Singh cautions. "From my perspective, security and compliance should be discussed from the get go. It should be part of their overall strategy."
In the meantime, some healthcare organizations already have plunged into big-data analytics, with impressive results. Click through our slideshow to see some innovative uses of analytics in healthcare.
How are you using big data in healthcare projects? Let us know in the comments section.
Several years ago -- before analytics became a buzzword among healthcare organizations -- UPMC Health Plan wrote its own software application to analyze its vast database comprising information pulled from multiple, disparate sources. This database, which currently holds 6.3 terabytes of data, includes electronic clinical notes, claims data, patient demographics, individuals, self-reported health assessments, pharmacy data, household data, and more.
As the chief analytics officer of UPMC Insurance Services Division and VP of health economics at UPMC Health Plan, Dr. Pamela Peele and her team use the homegrown software to perform analytics and predictive models to forecast patient behavior and provide preventive care. The big-data-based analytics also allows the organization to evaluate itself, creating best-practices and scaling the most efficient and effective programs.
"We use multiple technologies and techniques, including data visualization software that provides a view of highly complex and large datasets revealing underlying and previously unknown patterns and interactions between patients and providers," says Dr. Peele. "We're seeing real progress predicting patient behavior and providing preventative care."
For example, the two organizations have modeled the future impact of the flu season on members; figured out which members who require high use of medical services will continue to do so in the future; predicted who will need emergency care; and determined the probability of readmission before it occurs, Peele told InformationWeek. As a result, patient care improved, the healthcare provider manages resources appropriately, and both organizations operate more cost effectively.
In the future, other organizations might be able to benefit: UPMC is very aggressive about marketing the technology systems it develops for in-house use to other healthcare providers and insurers. The company has the philosophy that if UPMC has to develop software to meet its needs, then there's an unmet need in the market that UPMC should fill by commercializing its in-house software.
UPMC Tower
(Source: UPMC)
Since deploying the Mobile Area Routing and Vehicle Location Information System (MARVLIS) to help manage about 90,000 calls annually, Jersey City Medical Emergency Medical Services (EMS) improved response rates to less than six minutes, compared with the national standard of eight minutes and 59 seconds. Fewer minutes save lives: Today, half the patients who suffer a cardiac arrest regain a pulse, AllAnalytics reported. Before MARVLIS, only one in five did so.
The system melds geographic information system technology, wireless communications, and a global positioning system to give the EMS real-time information that allows teams to arrive at their destinations more quickly. The EMS can position teams where they're most likely to be needed, based on real-time analytics. For example, MARVLIS taught Jersey City EMS that more night calls come from residential neighborhoods, and that business areas require more services during the day, enabling the service to position ambulances closer to where they would be needed.
(Source: Jersey City Medical EMS)
In April, ConvergeHealth by Deloitte unveiled a multiyear collaboration with pharmaceutical company Forest Laboratories and Intermountain Healthcare to conduct outcomes research designed to improve the treatment of respiratory diseases.
Forest uses ConvergeHealth's OutcomesMiner software with anonymized data from Intermountain's electronic health records to conduct collaborative research in a real-world setting, Dr. Shailja Dixit, head of health economic outcomes and outcomes research at Forest, tells us. The pharmaceutical company can better understand unmet needs, treatment paradigms and paths, patient population, and other issues it needs to know in order to design better treatment options, she says.
Designing analytics for healthcare and life science customers means privacy and confidentiality come first, says Asif Dhar, executive VP for ConvergeHealth by Deloitte, chief medical informatics officer, and principal at Deloitte Consulting. Deloitte worked closely with Intermountain to co-create tools that would empower researchers to conduct their own analyses, he told InformationWeek.
"We are already seeing results," said Dr. Dixit. "We are hoping this will help give us insights into patient care deeper and early on."
(Source: Forest products)
In a pilot program, Carilion Clinic used IBM natural language processing technology and predictive modeling of data in its Epic electronic medical records to identify individuals at risk for heart disease from a pool of 8,500 patients. The pilot, featuring both structured and unstructured data from doctors' notes, achieved 85% accuracy in detecting heart failure, according to IBM.
This condition, notoriously difficult to diagnose, afflicts more than 5 million adults in the US, according to the Centers for Disease Control. The disease costs the nation more than $5 billion annually, yet more than half the afflicted individuals do not survive five years after diagnosis, the CDC reports.
Carilion Clinic's Dr. Stephen Morgan and RN Cathy Jennings review the Epic EMR.
(Source: IBM)
Michigan Health & Hospital Association's Keystone Center, a non-profit organization that focuses on patient safety initiatives, is using ArborMetrix's RegistryMetrix to capture obstetrics data and reports at about 60 hospitals across Michigan.
The goal: to improve patient safety and reduce adverse events. In particular, using data that RegistryMetrix gathers from participating obstetricians, Keystone Center will fold ArborMetrix's cloud-based clinical intelligence and insights into its best-practices.
(Source: MHA)
Michigan Health & Hospital Association's Keystone Center, a non-profit organization that focuses on patient safety initiatives, is using ArborMetrix's RegistryMetrix to capture obstetrics data and reports at about 60 hospitals across Michigan.
The goal: to improve patient safety and reduce adverse events. In particular, using data that RegistryMetrix gathers from participating obstetricians, Keystone Center will fold ArborMetrix's cloud-based clinical intelligence and insights into its best-practices.
(Source: MHA)
-
About the Author(s)
You May Also Like