HHS Embraces Big Data to Help Battle Opioid Crisis

In tandem with White House efforts, the Health and Human Services Department is tapping data to better understand and subdue the mounting opioid epidemic.

As awareness and attempts to slow overdoses swell, Big Data tools are set to play a large role in better tracking opioid prescriptions and enabling federal, state and local health agencies to more effectively allocate resources to communities in need.

On March 19, the White House revealed its Initiative to Stop Opioid Abuse and Reduce Drug Supply and Demand, which seeks to address the driving forces of the opioid crisis through several means. The White House officially declared the epidemic a national emergency in October in response to rising death rates. Opioid abuse led to more than 63,000 deaths in 2016 — a mounting number that is overwhelming public health agencies, healthcare organizations and states alike.

Now, the White House has set its sights on expanding education and treatment opportunities while reducing prescriptions to cut back on opioid addiction. But this isn’t the first attempt the federal government has made to better track, understand and quell the issue. The Health and Human Services Department hopes to control the epidemic through Big Data, following the lead of several states that are calling on data to better track prescriptions.

SIGN UP: Get more news from the HealthTech newsletter in your inbox every two weeks

HHS Could Call on State Data to Track Opioid Prescriptions

President Trump’s proposed fiscal 2019 budget for HHS would allocate $10 billion in funding to address the crisis. Part of the proposal would “require states to monitor high-risk billing activity to identify and remediate abnormal prescribing and utilization patterns that may indicate abuse in the Medicaid system,” HHS Secretary Alex Azar told members of Congress last week, Health Data Management reports.

Azar added that HHS could piggyback off of data from states’ internal prescription drug monitoring programs. PDMPs seek to spot doctor shopping by establishing a database of federal controlled substances that doctors and pharmacists can use to check patients’ medication histories alongside their use of other drugs.

Certain states, such as Florida and Kentucky, have already seen the results of implementing these programs. Florida saw a 52 percent drop in deaths related to oxycodone overdoses from 2010-2012 after implementing a PDMP, for instance, and Kentucky has also seen a significant drop in controlled substance use, according to the Centers for Disease Control and Prevention. Following the lead of successful states, others, such as Virginia and Missouri are well on their way to establishing their own PDMPs.

“[Prescribers] can see at a glance whether they want to slow down [prescriptions],” David Brown, the director of the Virginia Department of Health Professions, told Williamsburg Yorktown Daily, regarding a new tool that the state is implementing as part of its PDMP. He added that other clinicians are lined up to employ the technology as well. “It gives you a tool to show you if the person is doctor shopping or is already addicted. It will tell you if prescriptions are being dispensed at different pharmacies.”

Meanwhile, providers are moving forward with their own goals and solutions. Utah-based Intermountain Healthcare, for instance, is leveraging its electronic health record system alongside new sources of data and its growing health IT infrastructure to cut prescriptions by 40 percent. EHR data could be the answer for health systems looking to cut prescriptions. A new study published last month in the Journal of General Internal Medicine, for instance, shows that electronic health record data could be leveraged to create predictive models to better forecast which patients might be at risk for opioid abuse.

“Our model accessed EHR data to predict 79 percent of the future [chronic opioid therapy] among hospitalized patients,” the study notes. Chronic opioid therapy is how the authors define chronic opioid use. “Application of such a predictive model within the EHR could identify patients at high risk for future chronic opioid use to allow clinicians to provide early patient education about pain management strategies and, when able, to wean opioids prior to discharge while incorporating alternative therapies for pain into discharge planning.”

HHS Opioid Code-a-Thon Highlights Big Data Winners

Beyond current tools, HHS is also encouraging the production of new technologies that take advantage of Big Data to help quell the rising tide of opioid addiction.

In December, HHS held its first opioid crisis code-a-thon, which saw participation from nine teams and 300 coders. They were given access to government data with the aim to build tools that could help prevent opioid abuse, enable better access to treatment and encourage healthy clinical use.

“When the government holds a code-a-thon, there’s an admission in society that we have hit a roadblock, that we need help, and we go out to the polity to help us,” HHS Chief Technology Officer Bruce Greenstein said at October's Connected Health Conference, MobiHealthNews reports. “Maybe it’s not unique around the world, but it’s certainly rare when a government admits that it’s stuck and needs help from its people. And this is one of the most participatory forms of the relationship between a government and its polity, for maybe one of the most important, pressing questions and problems in our society today.”

The team that took home the $10,000 prize in the prevention track created a data visualization tool that helps users — addicts, public health specialists and others — locate drug takeback programs.

"We looked at CDC data and we found that, for the most part, individuals that abuse opioids are not obtaining them from drug dealers. The majority are obtaining them from their families. Over 70 percent of individuals that abuse opioids obtain them or borrow them or steal them from their families,” said Taylor Corbett, the spokesperson at the team presentations for the winning company, Visionist, MobiHealthNews reports. “Our hypothesis was, individuals are getting these meds because they’re left over from an operation or something like this. What can we do to get those drugs out of circulation?"

The winner in the treatment track, meanwhile, used data to model and predict overdoses so that public health agencies could better track and predict overdoses and better stock and allocate resources.

The move toward Big Data is part of a cultural move by HHS to better use and understand data in overall operations and to deliver better care to the U.S. population.

FotografiaBasica/Getty Images
Mar 20 2018