SAS analytics platform used to fight opioid epidemic

While SAS and a host of other analytics software vendors have developed tools to help in the fight against COVID-19, SAS has also developed tools to aid those battling the opioid epidemic.

According to SAS medical director Steve Kearney, who spoke during SAS' virtual user conference on June 16, more than 12,000 people per day were treated in emergency rooms for opioid abuse and opioid overdoses in 2018, and more than 130 people died every day from opioid overdoses with the number potentially as high as 180 people per day.

In response to the epidemic, SAS, founded in 1976 when it was part of North Carolina State University and now based in Cary, N.C., has developed interactive dashboards that, according to Kearney, are being used by both healthcare and law enforcement agencies at all levels to track and predict the spread of the opioid epidemic and make data-driven decisions to combat the epidemic.

And much like those tools vendors have provided to healthcare and government organizations to help them make decisions related to the COVID-19 pandemic, the analytics tools SAS has developed to help healthcare and law enforcement agencies battle the opioid epidemic can assist organizations from the biggest federal agencies to the smallest towns and clinics.

The difference is just the data that is loaded and analyzed.

"Through the use of visualization, network analysis, risk scoring and real-time alerts, this analytic environment ... can empower law enforcement and health officials to get services to the people in real time, and most importantly, save lives," he said.

Left on their own, individual healthcare and law enforcement organizations can be isolated and often don't have enough data to understand the intricacies of the epidemic. They have access to raw totals provided by the Centers for Disease Control and Prevention and other state and federal agencies, along with their own data, but that still leaves them disconnected from other relevant data.

Through the use of visualization, network analysis, risk scoring and real-time alerts, this analytic environment ... can empower law enforcement and health officials to get services to the people in real time, and most importantly, save lives.

It can hamper their ability to identify people who may be at risk of overdose or distributing opioids, and doctors or pharmacies that may be complicit in the spread of opioid abuse by overprescribing or illegally prescribing opioid-based drugs.

SAS, however, given the use of its analytics platform by law enforcement and healthcare organizations fighting the opioid epidemic, is able to pull together data from disparate sources and package it so it can be queried and consumed to drive decision-making.

It can take that disconnected data, and with the help of data management features powered by augmented intelligence and machine learning algorithms, turn unstructured data into interactive dashboards that municipal governments and local, state and federal law enforcement agencies can use to make data-driven decisions.

"The SAS approach promotes interoperability in combining data sources," Kearney said. "It allows for a common data language, and an analytic environment that promotes data-driven decision-making."

Using New Jersey as an example, Sree Vuthaluru, a system engineer at SAS, showed how SAS' analytics platform can be used to understand even the most minute details of the opioid epidemic across the state -- even down to an individual level -- and subsequently drive decisions designed to help people.

One dashboard showed the state as whole with the number of fatal overdoses, drug-related arrests and administrations of naloxone, a drug used to reverse the effects of opioid overdoses. A line chart on the same dashboard, meanwhile, displayed the number of fentanyl-related overdoses in yellow and the number of oxycodone-related overdoses in orange.

By clicking on a given county, that same data can then be broken down to the county level.

Another dashboard showed insights geared toward law enforcement agencies -- drug possession arrests and drug distribution arrests, for example -- on one side and information such as overdoses and naloxone administrations more aimed at healthcare organizations on the other. By clicking on different maps, the dashboard was able to show the data at a statewide or county level.

"The purpose behind these maps is to better understand if the areas where individuals are purchasing drugs are different than those where they're using drugs," Vuthaluru said.

Still another dashboard showed where the number of overdoses outpaces the number of arrests and vice versa by overlaying the number of drug-related arrests and the number of fatal overdoses in a given county per 100,000 people in order to standardize the data.

"This small insight helps policymakers better tailor their approaches for different parts of the state," Vuthaluru. "One of the counties may need more treatment centers and naloxone kits whereas another may need more law enforcement."

But the SAS analytics platform does more than simply show data. It's able to take that data and calculate the risk of overdose from the national level all the way down to the individual so that someone can be targeted for help. Using information such as the number of drug offenses, convictions and total opioid dosage prescribed, organizations can develop risk scores.

"The risk score can be altered to add other metrics or variables, depending on what is most relevant for the specific state, municipality or country," Vuthaluru said. "The model to create these risk scores is created within SAS and can be automated to output results as the data is updated."

Beyond identifying individual drug users and locations most at risk, the analytics tools from SAS can be used to pinpoint instances of doctor shopping -- when a person visits multiple doctors seeking one who will dole out opioid prescriptions -- and those of physicians overprescribing opioids.

"SAS is using its technology to better understand the opioid epidemic, as well as make data-driven decisions and actions through investigations," Vuthaluru said.