Preventing overdoses with data
While COVID-19 grips communities and dominates the headlines, other public health crises continue to impact millions. The pandemic may have worsened the suffering: For example, more than 40 states are reporting an increase in opioid-related deaths since the beginning of the COVID-19 outbreak.
Between 1999 and 2018, nearly 450,000 Americans died from an opioid-involved overdose. In 2019, overdose deaths increased 4.6 percent, with more than 50,000 of those attributed to opioids.
This has driven government legislation and regulations to address opioid oversupply, as well as increase access to lifesaving tools and treatments, as government and public health officials continue to focus on ending this epidemic.
With these rudimentary prevention tools already in place, the search for an advanced solution to the unrelenting toll of opioids may be found in data.
Blue Cross and Blue Shield of Illinois, Montana, New Mexico, Oklahoma and Texas (Blue Cross and Blue Shield), pursued an accurate way to predict those most at risk for an opioid-related overdose or related adverse event. The companies began analyzing claims data to determine any patterns or key indicators that may help them in addressing the ongoing crisis.
Through claims data, researchers were able to validate a screening tool that could accurately predict which of its members were at highest risk. The Risk Index for Overdose or Serious Opioid-Induced Respiratory Depression (RIOSORD), which is based on several factors, including a member’s underlying health conditions and their prescription medication characteristics (e.g., opioid formulation, concurrent antidepressant/benzodiazepine), was implemented in a predictive analysis.
The Blue Cross and Blue Shield data scientists validated a 2015 study on opioid overdoses among Veterans Affairs patients that had a 90 percent accuracy rate in identifying members at highest risk and found a comparable success rate when applied to the insurer’s claims data. The initial findings, which looked at commercially insured individuals with a prescription for opioids over a two-year period, laid the groundwork for a Risk Identification and Outreach (RIO) program made up of behavioral health, pharmacy and medical clinicians who work closely with the insurer’s data scientists to identify patterns of risk. RIO clinicians provide targeted outreach and intervention to at-risk members and their providers to reduce opioid adverse events and enhance member safety. The collaboration between the Blue Cross and Blue Shield data scientists and clinicians helps utilize predictive analytics to target a specific medical condition.
Since the program began in 2019, Blue Cross and Blue Shield has identified more than 5,500 members in need of assistance and enrolled over 2,300 individuals in case management programs.
“Predictive analysis is commonly used to improve both care and efficiency,” added Dr. Ben Kurian, Blue Cross and Blue Shield of Illinois, Montana, New Mexico, Oklahoma and Texas RIO Executive Medical Director. “Having the ability to identify at-risk members prior to a crisis will help us better meet the needs of individuals and families we insure.”
In addition, the insurer’s RIO program recently started applying these same principles of data mining, organizing and visualizing in an effort to reduce the number of new members that become persistent opioid users. The Blue Cross and Blue Shield companies hope to utilize predictive analytics and multi-disciplinary collaboration to assist in implementing clinically appropriate and effective interventions for other complex health conditions.
Blue Cross and Blue Shield of Illinois, Montana, New Mexico, Oklahoma and Texas are all licensees of the Blue Cross Blue Shield Association, an association of independent, locally operated Blue Cross and Blue Shield companies.
