Battling the opioid epidemic with knowledge, commitment and advanced analytics

By George Mathew, MD and Rikin Patel

The opioid epidemic in the United States is causing devastation. It is claiming lives, with a reported 20,145 deaths per year due to synthetic opioids other than Methadone (vs. 64,000 deaths per year from overdoses from all causes) by end of 2016. It is creating stress and strain for families struggling to cope with relatives who have become addicted to pain medication. And it is stretching federal, state and local resources to their limits.

Reporting on the epidemic has become a mainstay of the 24-hour news cycle, online, on the air and in print — and all note that opioid overdoses are now claiming more lives than guns or car accidents.

As a healthcare provider, I can relate to the news on a more personal level: As I was finishing my shift in the emergency room recently, an elderly patient was rushed in. He had overdosed on a cocktail of OxyContin, heroin and Ativan, and had been resuscitated in the field by firefighters with naloxone. Although we deal with substance abuse patients regularly, this case struck me because it demonstrates the spread of addiction. The patient was a retired engineer who had started taking prescription opiates for something extremely common, lower back pain, and over time, had become distant from his family as he graduated to combination drugs.

This is one of the many faces of the opioid epidemic — striking across all age groups, ethnicities, locations, income levels and occupations.

As attention to the crisis continues to intensify, a number of federal efforts are underway to alleviate it, including:

  • In December 2016, the Substance Abuse and Mental Health Services Administration (SAMHSA) announced that up to $970 million in grants would be directed through the State Targeted Response to the Opioid Crisis Grants over the next 2 years, enabled by the 21st Century Cures Act[i].
  • On September 5, 2017, the Centers for Disease Control awarded $28.6 million to help fight opioid abuse[ii].
  • On September 15, the U.S. Department of Health and Human Services announced it was awarding $144.1 million in grants to prevent and treat opioid addiction[iii].
  • On October 26, 2017, President Donald Trump formally declared that the opioid epidemic in the United States was a public health emergency, with additional efforts to follow[iv].
  • On October 30, 2017, Seema Verma, the administrator for CMS, announced support for exploring new payment innovation models to focus on, among other areas, opioids and substance abuse[v].
  • On November 1, 2017, the President’s Commission on Combating Drug Addiction and the Opioid Crisis, after months of research, released recommendations on how best to combat the opioid epidemic[vi].
  • On November 2, 2017, CMS invited Medicaid Directors to submit innovative ideas to work together to combat the opioid crisis[vii][1]

In addition to federal efforts, states are looking at options to help people with opioid addictions and training for providers. There are prescription drug monitoring programs that are used to check whether opioid addicts are receiving too many prescriptions from providers, and physicians are being warned they face license loss or incarceration if they overprescribe. Physicians must also undergo mandatory controlled substance training. Other programs include: medication-assisted treatment — on its own and using telemedicine; training emergency responders and nonmedical personnel to administer naloxone; an increase in opioid addiction rehab programs; and the use of media education kits to spread awareness and try to prevent people from becoming addicted in the first place.

As states grapple with how best to combat this crisis, it has become apparent that targeted decisions may have to be made on a budget. But how to make these decisions?  One option for dealing with the opioid challenge while balancing budget constraints is through the power of big data and analytics.

Powerful analytics tools are now available — big data analytics, machine learning, neural networks, robotics — and these can reveal actionable insights that, until now, would have been in the realm of best guesses. We’ve discussed before how analytics tools such as these can help improve outcomes and control costs (see our paper, “The Rise of Advanced Analytics in Medicaid”). Now we can focus these tools on helping states fight the opioid epidemic on the ground.

How these tools are leveraged and incorporated into broader programs to combat opioid overuse and abuse will be integral to program success. There are three key considerations: 1) combining traditional analytics approaches with new models to test culture, model or style; 2) establishing open pathways to unlock cross-functional data across organizational boundaries or states; 3) collaborative efforts between Medicaid directors to solve the opioid and budgetary challenges. Intricate elements are involved with each of these points.

First, incubating, testing and refining of combined analytics approaches should encompass several key elements:

  • Predictive models can identify opioid substance abusers that are at high risk of overdose or need rehabilitation services using just claims data.
  • Heat-mapping can determine areas of high prescription and overdose rates to localize high opioid traffic areas.
  • Robotics and simulation testing can be used to optimize substance abuse support services, such as to determine the right combination of rehabilitation and rescue based on frequency of overdose and access to resources, as well as to monitor progress and recommend changes in existing substance abuse programs.
  • Outlier analysis — typically used for fraud, waste and abuse — could also be used to discover “pill mills,” or providers who inappropriately prescribe narcotics, by tracking prescription data.
  • Integrating capabilities such as sentiment analysis (opinion-mining or emotion AI) could monitor substance abusers and help them before they are about to abuse opioids again.

Second, areas where data traditionally has not been easily shared among organizations or states — prescription drug monitoring programs (PDMPs), claims, medical records, sources of social determinant of health, etc. — must be more seamlessly shared to achieve better patient support. These include:

  • Provide better access to opioid-relevant data by care teams, government agencies and other parties, so that this information can be meaningfully leveraged and easily consumed.
  • Develop a longitudinal view of a patient to effectively support and drive public health programs to high-risk geographic areas or populations; for example, gauge the impact on the managed care organization/accountable care organization model and their relationship with the state.
  • Make data-driven coaching, education and other tools available to the care teams.
  • Integrate analytics into all member interactions through a shift to real-time analytics, facilitating decision making.

Finally, to successfully integrate new analytics capabilities, stakeholders need to collaborate. Medicaid directors will need to join forces to solve the challenge through:

  • Pooling of data (PDMPs, claims, electronic health records, social determinants, etc.) across states and then use predictive models from larger data sets to identify individuals who are high risk
  • Collaboration across industries to improve the state of capabilities
  • Creating ownership to achieve outcomes — perhaps one of the most difficult challenges in a market focused on a capitation model
  • Agreement on how to measure success by determining the metrics to measure program effectiveness and finding which interventions work best

We are caught in a battle against the opioid epidemic, and multiple programs at the federal, state and local level have been put in place to try to address what appears to be an intransigent problem. This is made more difficult because of budget constraints. However, with the right tools, we as a country have a chance to fight back.

Visit us at the DXC booth (#13) at the National Association of Medicaid Directors Fall 2017 Conference in Arlington, VA, November 6–8.


[i] https://www.samhsa.gov/newsroom/press-announcements/201612141015

[ii] https://www.cdc.gov/media/releases/2017/p0905-opioid-funding.html

[iii] https://www.hhs.gov/about/news/2017/09/15/hhs-commits-144-million-in-additional-funding-for-opioid-crisis.html

[iv] http://edition.cnn.com/2017/10/26/politics/donald-trump-opioid-epidemic/index.html

[v] https://www.cms.gov/Newsroom/MediaReleaseDatabase/Fact-sheets/2017-Fact-Sheet-items/2017-10-30.html

[vi] https://www.whitehouse.gov/sites/whitehouse.gov/files/images/Final_Report_Draft_11-1-2017.pdf

[vii][vii] https://www.medicaid.gov/federal-policy-guidance/downloads/smd17003.pdf


George Mathew, M.D. is the Chief Medical Officer for the North American Healthcare organization for DXC. In this role, he serves as the clinical expert and healthcare thought leader to our healthcare clients in transforming the healthcare marketplace. Dr. Mathew graduated from Boston University School of Medicine and completed his residency in Internal Medicine at Greenwich Hospital/Yale University in Connecticut.

 

Rikin Patel is a DXC Technologist with 25 years of diverse experience in Information Technology.  He serves as the Chief Technologist for DXC’s Americas Healthcare & Life Sciences and is a member of the Office of the CTO. Rikin is responsible for building key client relationships, advising senior leadership on technology trends, and providing thought leadership to effectively grow client and DXC business.

 

 

RELATED LINKS

What state Medicaid organizations need to do to address funding uncertainty

Trackbacks

  1. […] Battling the opioid epidemic with knowledge, commitment and advanced analytics […]

    Like

  2. […] Battling the opioid epidemic with knowledge, commitment and advanced analytics […]

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: