AI, Analytics, and the Opioid Epidemic
Opioids are ravaging the United States. The National Institute on Drug Abuse estimates that every day over 130 Americans die after overdosing on this class of drugs. The problem is complex and it will take a multi-faceted approach to ameliorate the underlying issues. The healthcare sector is on the front lines of this public health crisis and must take the lead by managing prescribing patterns, promoting medication assisted treatment (MAT) protocols, taking steps to destigmatize addiction, and including complementary treatments in benefit design.
Data is the New Oil for Health Plans
Within the healthcare ecosystem, health plans are uniquely positioned to address the opioid epidemic. They have rich data about members, as well as the resources to coordinate care and manage providers. Health plans’ member data is what can truly drive successful and cost-effective interventions to prevent and treat addiction.
A few years ago, British mathematician Clive Humby was credited with saying “Data is the new oil” and this metaphor seems particularly apt in the realm of opioids and total population management. Valuable patient insights reside within health plan data, but that data must be first be refined and analyzed before the value can be realized.
Risk analytics are the key to identifying which patients are currently in significant danger of becoming dependent on opioids. Artificial intelligence (AI) can turbocharge patient risk information further by generating predictive risk analytics which anticipate patients who may be in danger in the future of opioid abuse, based on different characteristics.
Academics at the University of Toronto observed in the new book “Prediction Machines” that using AI to generate analytics reduces the cost of prediction. As a result, the power of prediction can be applied to many different types of problems, including the opioid epidemic. HMS’ Elli risk intelligence engine provides predictive population insights and highlights health plan members that are a high priority for intervention. Health plans can see which members are taking Controlled Substance prescriptions by class, who is receiving Medication Assisted Treatments (MATs), and who has experienced drug overdoses by drug class.
Five Ways to Identify, Assess, & Address Opioid Misuse Using Analytics
As health plans join the war on opioids, they can take five steps to identify, assess, and address opioid misuse in their populations. Each of these leverages the power of analytics:
- Utilize Risk Analytics and Total Population Management. Predictive risk analytics highlight which members are rising in risk. With that information, health plans can engage in care management interventions. Intervention results can be fed back into the system to improve analyses related to future risks.
- Look at the Prescriber Side of the Opioid Equation. The value of health plan data isn’t limited to patients. Analytics can also shine a light on providers who prescribe the highest volumes of controlled pharmaceuticals, as well as the network pharmacies that fill the most prescriptions for opioids. This data may reveal distinct patterns within groups or even irregularities within providers’ prescribing patterns.
- Analyze Data Beyond Claims and Authorizations. Thanks to technologies like machine learning, generating predictive analytics is faster and cheaper than ever before. This means that health plans needn’t limit their analytics to claims and authorization information. Predictive analytics can be based on a deep repository of diverse data that includes self-reported information, survey responses, and other social determinants of health. With HMS, health plans can route responses from engagement channels, CRM systems, and care management systems into the risk intelligence and analytics engine. This leads to richer insights and more effective patient interventions.
- Intervene Immediately. Once a health plan has insight into members who are at risk for opioid misuse, care managers, providers, and care teams can take action. Once again, analytics can play a role in speeding the response. HMS’s Total Population Management platform generates risk scores that quantify which patients need intervention now. As members reach critical risk score thresholds, care managers and providers can be alerted proactively – either in real-time or daily.
- Create a Smart System. A coordinated response to the opioid epidemic isn’t possible unless all participants in the healthcare ecosystem have access to patient data. One way to achieve this goal is for health plans to create member profiles with alert capabilities. When key metrics reach certain levels, member services, care management, providers, and clinical care teams can be notified. This type of “smart system” can ensure that members receive the care they need.
With HMS Population Health Management, now it’s possible to rapidly understand your membership and take meaningful action in a timely way. By combining data, technology, and analytics, HMS is ready to help health plans intervene with at-risk patients in the moment. The goal is not only to treat individuals who are already dependent on opioids, but to prevent misuse from occurring in the first place. Join HMS and let’s end the opioid crisis starting today.