MITRE investigator: Advanced analytics can help predict, fight opioid abuse
Advanced data analytics may hold the power to help predict likely prescription drug abuse in patients, a needed breakthrough as an opioid epidemic sweeps the country.
The problem is "no longer just heroin and cocaine; it's prescription drugs,” said Jaya Tripathi, principal investigator at the MITRE Corp. “If you look at the trends, despite the overwhelming focus and funds, more needs to be done."
[Poll: What topics will define HIMSS16?]
Her team at MITRE researched public and private stakeholders and interviewed board members and law enforcement officials and others over the course of 18 months. The interviews revealed ways to carry out implementation with a broad approach.
In a Wednesday morning session at HIMSS16, “Data Analytics Takes the Pain out of Pain Pill Management,” Tripathi plans to discuss advanced analytics and the ways it can predict drug abusers. She'll also talk about some Big Data challenges and how to overcome them.
MITRE operates multiple, federally funded research and development centers and a robust research program. Tripathi's prescription drug project began three years ago, while she's been in the analytics field for a decade.
Tripathi said the study demonstrated provider challenges, especially in the emergency room, to differentiate the symptoms of similar patients, when prescribing medications.
"They do a lot of subjective interpretation," Tripathi said. "There are vital disagreements on how to handle it."
"I wanted to solve this problem and look at all current solutions, identify gaps and what we need," Tripathi added, saying a better predictive approach would provide the necessary empirical data.
"Data is everything, but you can do better if you know the right techniques," Tripathi said. "If your data is poor, your models will only go so far."
Tripathi noted that each pharmacy must report every controlled-drug prescription to the state, as much as once every 24 hours. Her project has found access to analytics data of this resource at the point-of-sale can make sure the drugs are making it into the right hands and in the right amount.
[Like Healthcare IT News on Facebook]
"We had a lot of information, like drug interaction — so when you outline that you get a much bigger picture," Tripathi said.
"People have done a lot of work out there and sometimes one technique will inform the other. But it also needs to be usable."
Tripathi’s session, “Data Analytics Takes the Pain out of Pain Management,” is scheduled for March 2, 2016 from 8:30 to 9:30 a.m. at the Sands Expo Convention Center Palazzo D.
This story is part of our ongoing coverage of the HIMSS16 conference. Follow our live blog for real-time updates, and visit Destination HIMSS16 for a full rundown of our reporting from the show. For a selection of some of the best social media posts of the show, visit our Trending at #HIMSS16 hub.