Opioid prediction model gives state life-saving prep time
ipopba/Getty Images
West Virginia health officials can reliably forecast drug overdoses based on data from neighboring areas.
A correlation analysis model helps public health officials in West Virginia predict when and where fentanyl-related overdoses will strike so they can make the necessary preparations for treatment and prevention.
In 2020, West Virginia had the highest drug overdose death rate of 81.4 per 100,00, according to the Centers for Disease Control and Prevention. The state is “the end of the line” for drug smugglers coming from the southern border and the Midwest, Director of the Office of Drug Control Policy Dr. Matthew Christiansen said. “When we see overdose spikes in those areas, we know that they’re coming to West Virginia too, and that gives us a little bit of time to be able to respond.”
The model incorporates historical overdose data and trends from nearby counties in other states such as Maryland to predict potential overdose spikes in West Virginia. Christiansen said the model is an expansion of one previously developed in Berkeley County and is valuable in predicting overdose incidents, especially for the eight counties that account for 60% of all overdoses in the state.
The tool is based on the national Overdose Detection Mapping Application Program platform, which enables data sharing among participating government agencies. West Virginia’s model not only incorporates overdose-related data from surrounding counties, but it also gets data from local and state agencies including the West Virginia Office of Emergency Medical Services. Christiansen said data will be updated every few months, allowing time to implement intervention efforts.
“Overdoses can be predicted pretty reliably with a correlation of around 80% or more, depending on the county based on overdose spikes that occur in outside areas that tend to supply those counties,” Christiansen said. For instance, if there is a jump in overdoses in western Maryland, Christiansen said that trend usually appears in West Virginia 48 to 72 hours later.
A network of regional coordinators receive a notification if the tool identifies an increase in overdoses, and they then send email alerts to community stakeholders such as homeless shelters, hospitals, treatment centers or quick response teams so “that we have resources available in these communities [for] people when they need it,” Christiansen said. Also, the state’s overdose data dashboard — developed by the Office of Emergency Medical Services and the Health Statistics Center — provides insights on the age and sex of drug users as well as the time of week and year overdose incidents occur. According to Christiansen, spikes tend to occur midweek and in early spring and fall.
“One of the biggest drivers of the overdose and addiction crisis that we’re seeing today is the opioids that we saw over two decades starting in the early 2000s,” Christiansen said. Many blue-collar workers had legitimate illnesses or injuries that required opioid treatment, but those individuals became addicted and “we’re still here picking up the pieces.” West Virginia also grapples with considerable unemployment, poverty and poor education, which are factors often linked to increased drug use.
The state’s decades-long entanglement with opioids resulted in two major lawsuits. In August, opioid distributors AmerisourceBergen, Cardinal Health and McKesson reached a $400 million settlement with West Virginia to be distributed among cities and counties to finance services such as rehab or first responder training. Likewise, Walmart and CVS reached a $147 million settlement for oversupplying the state with addictive drugs.
Looking forward, Christiansen said he hopes the model can include more indicators of drug activity. For example, data on the availability of health services, programs and treatment options such as Naloxone is valuable in determining the needs for specific areas tackling drug abuse.
Officials should also measure the work they’re doing on the ground, Christiansen said. Metrics will help ensure “that those interventions are having a direct impact on the lives of the people that we’re trying to help so that we’re not wasting time, we’re not wasting money, we’re only putting our resources towards the things that we know are going to work,”he said.
“We will continue to see ups and downs just like we do with all other chronic diseases, mental health disorders, based on stressors of the time,” Christiansen said, “but we have to keep plugging away.”
NEXT STORY: A Governor Imposes a TikTok Ban for State Government