10 surprising uses for analytics in healthcare
From watching opioids to segmenting patient populations, provider organizations are becoming AI savvy.
10 surprising uses for analytics in healthcare
Everyone is in on the analytics game, but not every organization has something to show for its efforts. Many providers are learning the hard way that implementing pricey tools doesn’t pay off unless there’s a clear plan to utilize insights in an impactful way. However, there are some that have cracked the code, and these leading organizations are using analytics to make a difference on a daily basis.
Riding the line
Palmetto Primary Care Physicians worked with Optum Analytics to use artificial intelligence and machine learning predictive models to conduct risk stratification analyses to identity those who “ride the line on pre-diabetes and are on their way to being introduced to diabetes,” says Terry Cunningham, CEO at Palmetto. The result was improved care and savings of $4 million across Palmetto’s accountable care organization contracts.
See the full story here.
Getting rules on the fly
An infection control program at Piedmont Healthcare in Atlanta cut the infection rate by 40 percent in the first year. The project was supported by Exasol, the vendor of an analytics database that operates much like Google, explains Mark Jackson, director of business intelligence. “Enter a query, and an array of servers are searched, and all the findings come back together. Now, I can add a new field or rule for requested analyses on the fly.”
See the full story here.
Predicting sepsis risk
A machine learning algorithm has been developed by University of Pennsylvania Health System to identify hospitalized patients most at risk for severe sepsis or septic shock. The sepsis machine learning algorithm was developed internally by researchers who trained a random forest classifier—an approach to classify a wide range of data—to sort through EHR data such as labs, vitals and demographics for 162,212 patients. The key is “having the electronic health record and the tools we’ve built from data in the EHR tell us something new, rather than just alerting us to something that we already know,” says Craig Umscheid, MD, associate professor of medicine and epidemiology.
See the full story here.
Using AI for eyes
University of Iowa Health Care uses an artificial intelligence-powered diagnostic system that detects diabetic retinopathy, a condition that damages the retina of the eyes and is a major cause of blindness in persons with diabetes. IDx-DR uses artificial intelligence to make a diagnostic assessment without requiring a clinician to interpret the image or results “There is simply no reason for more than 24,000 individuals to lose vision each year from diabetic retinopathy,” says Michael Abramoff, MD, a practicing ophthalmologist and founder of IDx.
Read the full story here.
Employing analytics to fight opioid abuse
Piedmont Athens has partnered with Invistics, a vendor of inventory visibility and analytics software that uses machine learning technology and analytics to detect opioid and drug theft across the hospital. Piedmont Athens feeds a range of data to Invistics, which conducts analytics to detect diversions include hospital information systems, medical record systems, employee time clocks, wholesale purchasing, inventory and dispensing cabinets.
Read the full story here.
Ferreting out pneumonia
Massachusetts General Hospital researchers developed an algorithm to provide automated, real-time monitoring of both ventilator settings and EHR data to prevent ventilator-associated pneumonia. An algorithm was found to be 100 percent accurate in identifying at-risk patients when provided with the necessary data.
Read the full story here.
Customizing patient treatment
Rush University Medical Center is adopting machine learning and analytics technologies to process patient information, including from imaging studies and other sources, with hopes of customizing patient treatment and delivering precision medicine. The Chicago-based academic medical center is using a combination of technology from Cloudera and MetiStream, which are working together on products that providers can use to improve patient outcomes. If successful, the initiative could provide insight into how healthcare organizations could cost-effectively improve genomic research and accelerate the pace at which insights could be used to influence patient care.
Read the full story here.
Finding drug diversion risks
Drug diversion in hospitals is not a new problem—data from 2016 estimates that 15 percent of pharmacists, 10 percent of nurses and 8 percent of physicians are challenged by alcohol or drug dependency. In all, as many as 10 percent of clinicians in the drug delivery loop may be diverting medications. Lakeland Health in West Michigan is using software to prevent the diversion of drugs from its pharmacies by analyzing multiple types of data to ensure hospital staff members are following safety protocols and procedures to lessen the risk of prescribed drugs getting into the wrong hands. “We want to ensure every patient in our system is protected from a potential mishandling of medication, while at the same time all team members are compliant with our policies and procedures,” says Kurt Wyant, manager of pharmacy.
Read the full story here.
Keeping an eye on imaging equipment
UCSF Medical Center in San Francisco is turning to an information services and analytics product from Glassbeam to power its clinical engineering analytics program, using predictive analytics and machine learning to ensure that its expensive imaging devices are optimally maintained. The hospital will work with the Santa Clara, Calif.-based company to use its CLEAN blueprint (Clinical Engineering Analytics) to manage components of its imaging equipment, with plans to expand its use to other modalities, such as ultrasound, cath lab and physiological monitoring equipment.
Read the full story here.
Identifying patient trends
CentraCare Health, an integrated delivery system in Central Minnesota, has implemented an analytics tool that is intended to help the organization make strategy decisions based on patient trends. Data that CentraCare receives from Decisionology—a product from Tea Leaves Health—displays customer segmentation information that goes beyond demographics to show the breakdown of patient customers, such as the care mix and services that patients received, as well as downstream utilization and revenue generation potential of key services and patient cohorts, says Melinda Bemis, director of strategic planning and business development at CentraCare. It also shows patients who no longer get care from the provider, opening opportunities to reach back out to the patients. CentraCare operates six hospitals and 27 clinics as well as nursing homes and senior housing services in central Minnesota.
Read the full story here.