The year 2021 has been declared the deadliest year in history when it comes to deaths from drug overdoses. According to the Wall Street Journal data from the Centers for Disease Control and Prevention shows that overdose deaths rose almost 29 percent in the 12 months ending this past April. Rates of overdose rose in 46 states, and drug overdoses reached a record of an estimated 100,306 individuals. As so much of what has happened since early 2020, COVID-19 is partly to blame. Both isolation and the divergence of resources contributed to complicating opioid addiction prevention and treatment. However, drug-related deaths are predicted to rise no matter how long the pandemic crisis continues, primarily due to the continued prevalence of fentanyl.
In a world where we are living with both a global pandemic and a national endemic of Opioid Use Disorder (OUD), solving this healthcare crisis seems more impenetrable than ever. Fortunately, the solution could lie in artificial intelligence (AI). AI models can be equipped to examine the vast amount of data sets across the various health and social factors in the opioid crisis, discovering patterns and insights. With the signing of the American Rescue Plan Act, allocating nearly $4 billion dollars to the treatment and prevention of substance abuse, a large influx of national funding towards data collection and distribution has made these models possible.
Industry innovators are working quickly to utilize these solutions. Healthcare technology firm HSR.health has developed a novel and patented approach that leverages geospatial tech, advanced AI models, and broad sets of social determinants of health data to stratify patients at risk of opioid addiction, diversion, and having an overdose in the next 12 months. The data can be leveraged by clinicians in treating at-risk individuals and mitigating the overall epidemic.
Providing actionable data is key to using AI-provided clinical insights for public health interventions and individual patient treatment plans. According to the Lancet, “Community-level modelling projections suggest that only those communities with increased capacity for treating (and retaining) people with medications for OUD will see a substantial reduction in overdose mortality.” Geocoded health and social data is what is needed to affect change on a community-level, and public health decision makers can use predictions to allocate resources in these communities, identifying and treating those afflicted with OUD. Allocating resources optimally is extremely important in these communities, and therefore AI modeling can truly expand their ability to target and treat OUD in their own communities – in a way that most equitably allocates resources where they are most needed.
In an ideal world, we could look at the opioid crisis holistically, thus being able to optimize resources to aid response and treatment. This will enable us to identify those at risk of opioid addiction and addiction-related calamity or death, and provide them the support they need, while also allowing those who do truly suffer from debilitating pain and benefit from opioid treatment under physician supervision. Whether that involves traditional substance abuse treatment, medication-assisted treatment, physician-guided (not forced) tapering, addiction counseling, or other approaches- allocating exactly to where they are most needed by patients will improve rates of drug-related mortality in communities.
Now, perhaps we paint a picture of what the world might look like if this ideal case is in place.
The author, Ajay K. Gupta, CISSP, MBA, is CEO of HSR.health.