Three ways analytics are improving clinical outcomes
According to Accenture Digital Health Technology Vision 2017, 84 percent of healthcare executives believe artificial intelligence (AI) will revolutionize the way they gain information. And many health organizations are already taking advantage of technologies such as AI and advanced analytics to gain insights that help them improve clinical treatment processes and outcomes. In fact, there’s a broad spectrum of use cases for clinical analytics.
Anticipating patient needs
Some of the first ways that health organizations have applied clinical analytics: looking for gaps in care and better predicting patient needs.
That’s becoming especially important with managed care models where health organizations receive reimbursement based not just on episodic health services but also on factors like length of stay (LOS) and readmission rates. Health systems are taking advantage of analytics to help them correlate staffing with anticipated patient needs and better coordinate care so they can improve patient outcomes and reduce LOS and readmission rates.
[msce_cta layout=”image_center” align=”center” linktype=”blue” linkurl=”https://discover.microsoft.com/practical-applications-for-ai-in-healthcare/?wt_mc_id=AID723268_QSG_SCL_327854#/” linkscreenreadertext=”Download the e-book now” linktext=”Download the e-book: Breaking down AI” ][/msce_cta]
For example, Steward Health Care analyzed multiple types of data—such as CDC, flu, seasonality, and social data—using Microsoft Azure Machine Learning to predict patient volume so they could staff accordingly.
The results have been impressive. The private hospital operator can predict volumes one to two weeks out with 98 percent accuracy. And it reduced the average LOS for patients by one and a half days. In other words, improved nurse scheduling is helping patients get better faster. It has also increased patient satisfaction. All this, plus: Steward Health Care is saving $48 million per year.
Empowering care teams with predictive care guidance
The next level up in clinical analytics is predictive care guidance. A great example comes from Ochsner Health System—where they’ve integrated AI into patient care workflows.
Care teams there get “pre-code” alerts through an Azure-based platform (from our partner Epic) so they can proactively intervene sooner to help prevent emergency situations. The AI tool analyzes thousands of data points to predict which patients face immediate risks.
“It’s like a triage tool,” says Michael Truxillo, Medical Director of the Rapid Response and Resuscitation Team at Ochsner Medical Center, in this article. “A physician may be supervising 16 to 20 patients on a unit and knowing who needs your attention the most is always a challenge. The tool says, ‘Hey, based on lab values, vital signs, and other data, look at this patient now.’”
During a 90-day pilot project with the tool, Ochsner reduced the hospital’s typical number of codes (cardiac or respiratory arrests) by 44 percent. That incredible number demonstrates the impact AI-driven predictive care guidance can have on clinical outcomes.
Accelerating rare disease diagnoses
Yet another example on the clinical analytics continuum is the work we’re doing with Shire and EURORDIS to accelerate the diagnosis of rare disease. Together, we’ve formed The Global Commission to End the Diagnostic Odyssey for Children with a Rare Disease. As part of the commission’s efforts, phenotypic data (the physical presentation of a person) and genomic data are analyzed to gain insights that could help physicians identify and diagnose patients with a rare disease more quickly.
On average, it takes five years before a rare disease patient—of which approximately half are children—receives the correct diagnosis. Harnessing the power of AI-driven clinical analytics, the alliance aims to shorten the multi-year journey that patients and families endure before receiving a rare disease diagnosis. And that’s one of the most important issues affecting the health, longevity, and well-being for those patients and families.
Those are just a few examples of how AI and advanced analytics can transform healthcare and improve clinical outcomes.
Together with our partners, we’re dedicated to learning and growing alongside our customers and helping them achieve the quadruple aim through clinical analytics and other cloud-based health solutions. We’re also committed to helping them meet their security needs and safeguard the privacy of PHI. And our customers have peace of mind when innovating with us thanks to our Shared Innovation Principles that provide clarity around co-creating technology. We value our customers and partners’ expertise and don’t seek to own it. Rather, we help them monetize their technology assets.
However your health organization wants to use—or advance your use of—clinical analytics, you can learn how to take advantage of AI tools and see more real-world use cases in the e-book: Breaking down AI: 10 real applications in healthcare.
