Actuaries and the Opioid Epidemic

Actuarial expertise is informing clinical decision-making and health care cost-impact modeling Erin A. Ferries Guy

Photo: istock.com/wildpixel

The United States is experiencing a public health epidemic unlike any we have seen before, with the utilization of a single drug class having a devastating impact on millions of Americans and causing catastrophic medical, societal and economic effects. The statistics associated with the opioid epidemic are staggering. Overdoses are the leading cause of accidental deaths in the United States, which means you are more likely to die from an opioid overdose than you are in a car crash.1 In recent years it has been estimated that up to 130 Americans die of an opioid overdose every day, and in 2017 an additional 11.4 million people reported misusing opioids within the past year.2

Combatting and resolving this national drug crisis has overwhelmed federal, state and local health care resources. However, multifaceted approaches employed in recent years have reached beyond the health care system to other professional communities that have offered new and innovative approaches to curbing prescription drug abuse, demonstrating both success and optimism where it was desperately needed.

Opioid Epidemic: A Brief History

An essential step to identifying solutions to the opioid epidemic is understanding how we got here in the first place. There were several major contributing factors to the rise of the opioid epidemic; however, to summarize, this crisis is the result of a major systematic failure of the entire health care ecosystem. In the 1990s, as an effort to reduce the inadequate treatment of pain, the American Pain Society began pushing a campaign that encouraged providers to treat pain as the ‘fifth vital sign.’3 The pharmaceutical industry quickly saw this as an opportunity to capitalize and began ramping up production of new highly potent opioid formulations, such as OxyContin. Aggressive campaigns for the dispensing of opioids as the “humane treatment option” to address pain also were undertaken by pharmaceutical companies during this time.4 However, the highly addictive nature of these medications was drastically downplayed and covered up; therefore, health care providers began to prescribe them at exceedingly fast rates. Opioid prescribers during the 1990s through mid-2010s were walking a fine line between adequately treating pain and becoming glorified drug dealers. At the height of the epidemic in 2012, the prescribing rate was 81.3 opioid prescriptions dispensed for every 100 persons in the United States.5

To describe the United States as an outlier in terms of opioid prescribing compared to the rest of the world would be a vast understatement. While the United States only makes up 4.3 percent of the world population, its citizens consume more than 80 percent of all opioids in the world, including 95 percent of all hydrocodone prescriptions.6 As perplexing as this statistic is (and, yes, individuals in other countries do break bones and experience pain), the answer is very simple: Other countries recognize the highly addictive nature of these drugs and do not make them available to general patient populations. Rather, they reserve opioid use for patients experiencing extreme pain, such as trauma patients or those with terminal cancer diagnoses.

In the United States, we have developed such a strong dependence on potent and addictive pain medications that it has become common practice for patients to receive opioid prescriptions for toothaches and sprained ankles. These overprescribing patterns are of great concern given that opioids can lead to physical dependence in an extremely short time frame. A patient’s likelihood of utilizing opioids over the long term—defined as continued use greater than one year after starting their first opioid prescription—increases with each additional day supplied for their first opioid prescription.7 This risk starts to increase drastically after just three days of use. After a patient utilizes opioids for five days, which is not uncommon for acute pain conditions, there is a 10 percent chance they will still be utilizing opioids in one year. After 10 days of use, this risk jumps to 20 percent; it then climbs to 35 percent after utilizing a 30-day supply.

There are many other factors that have contributed to the opioid epidemic, including the failure to adequately treat mental health conditions. Research has uncovered that 50 percent to 90 percent of patients with substance-use disorders have a co-occurring mental health disorder, such as anxiety and depression.8,9 The subjective nature of pain creates additional challenges, especially among patients with mental health conditions, as the perception and experience of pain is an incredibly personal experience and often difficult to measure.

Where We Are Today

While the statistics surrounding the opioid epidemic can be dire, thanks to the tireless coordinated efforts from stakeholders across the medical community, meaningful strides have be made in addressing nonmedical opioid use in the past two years. Today we are seeing the lowest opioid prescribing rates in 15 years and experiencing the first decline in overdose deaths in almost three decades.10 However, much of this success needs to be attributed to the work that went on behind the scenes. The contribution of the actuarial community to combating this epidemic cannot be unstated, especially when considering some of the essential skills that are required to address any public health epidemic, skills that are intrinsically abundant in the actuarial community: risk management, problem-solving and risk valuation.

Countless research and publications socialized by the actuarial community examining opioid risk and predictive models have helped inform and guide providers, health plans, policymakers, researchers and countless other parties affected by the opioid crisis. While this is a time for appreciation and reflection, we are not out of the woods yet, unfortunately. Although important strides are underway, we need to keep in context the fact that opioid prescription dispensing rates are still three times higher than they were 20 years ago.11 Continued vigilance by all stakeholders will be required to reduce and proactively prevent future cases.

There are several areas where the role of the actuary has been integral to success. This article should heighten awareness and serve as a call to action for members of the actuarial community—this is an opportunity to use your unique talents to contribute to combatting one of the most difficult public health challenges to ever face our nation.

The Actuary’s Role

When the number of individuals who misuse, abuse and overdose on opioids skyrocketed into the millions over the past decade, it became easy to remove the human element from the problem—especially when it is something like drug abuse. The general assumption is to label the individuals who misuse and abuse opioids as “stereotypical drug abusers.” However, one of the most shocking aspects of the opioid epidemic is the large proportion of individuals who do not fit the “stereotypical drug abuser” profile.

Opioid abuse is unique from other forms of drug abuse in that abusers span vast patient populations—opioid abuse does not discriminate drastically across genders, education levels, socioeconomic classes and geographic locations. This is partially thought to be due to the stigma that because opioids often originate from doctor’s offices, they are safer than illicit drugs. This deviation from standard assumptions is exactly where actuarial expertise has become critical. Actuarial communities across industries have developed innovative risk prediction methodologies, decision trees, regression models and machine learning to identify risk profiles of patients who may be at a higher risk of developing opioid misuse or an opioid use disorder. Advanced analytics and predictive models have been utilized to analyze patterns from data integrated from multiple sources, including medical and prescription drug claims, electronic medical records and state-based prescription drug monitoring platforms. Connecting these different data sources allows for identifying factors and quantifying risk that may not be completely evident from one data source alone, such as:

  • Doctor shopping behavior (patients obtaining opioid prescriptions from multiple prescribers and filling them at multiple pharmacies)
  • Concurrent use of opioids with other high-risk drugs, such as benzodiazepines (e.g., Valium, Xanax) and/or stimulants (e.g., Adderall)
  • Cumulative dosage of opioids (morphine milligram equivalents (MME))
  • Diagnoses history (depressive/anxiety disorders)
  • Incarceration history

This data has been constructed into new and innovative models that are directly integrated as risk scores and alerts or notifications into prescriber-, pharmacy- and health plan-facing platforms. These risk scores help health care providers establish if the risk of dispensing the prescription outweighs the benefit for the individual patient before prescribing an opioid. The notifications also identify patients demonstrating potentially aberrant opioid-seeking behaviors and prevent patients early on from escalating to opioid use disorder.

The development of this predictive modeling also allows for data-driven decision-making to help determine which specific treatments or targeted interventions would be most appropriate and effective at helping the unique individual. These risk models have been developed actuarially and are employed to identify which interventions would help reduce a patient’s risk of experiencing an overdose event, such as identifying patients utilizing high doses of opioids so they can be provided with overdose rescue medications (e.g., naloxone) to keep on hand, or targeting specific treatments such as pharmaceutical therapies (medication-assisted treatment (MAT)), physical and occupational therapy, behavioral health services, or combinations of multiple therapies. An equally important task is identifying patients who have medical needs and would benefit from being on opioids for acute or short-term use, or those with severe chronic conditions who would greatly benefit from opioid use. These actuarially developed individual risk profiles are not only essential for identifying high-risk patients, but also for identifying providers whose prescription patterns may indicate an issue with inappropriate or overprescribing of opioids and other controlled substances. The development of these risk assessment models helps in conducting targeted provider outreach and education, and in informing prescribers about adhering to clinical practice and care guidelines.

Quantifying the Cost of Epidemic

Quantifying the overarching cost of the opioid epidemic has been one of the most important responsibilities tasked to the actuarial community. Measuring the economic impact has helped socialize the scale of lives impacted, and understanding which communities face the greatest disparities can inform policy so we direct resources where they would be most impactful. While the health care system is estimated to accrue upward of $80 billion in direct costs due to opioid-related events each year, it is not adequate to solely examine the costs incurred by the health care system as the primary economic measure of the opioid epidemic.12 To fully measure the vast economic implications of the opioid epidemic, we must account for factors such as lost productivity, indirect medical costs, criminal justice costs, societal costs and lives lost.

The Society of Actuaries (SOA) estimates the total economic burden caused by the opioid epidemic from 2015 to 2018 was $631 billion, with other estimates including the statistical value of life ranging up to $2.5 trillion during this time frame.13,14 Understanding the contributing factors that have economic impacts in other areas of society, such as excess costs associated with incarceration and limited costs attributed to behavioral health care services, also highlight underlying issues in the rise of the opioid epidemic. Unfortunately, there are no economic models that can also capture the pain and suffering felt by family members of individuals who have died from fatal opioid overdoses. However, quantifying societal costs has helped increase our understanding of the overall impact of the epidemic and allows policymakers to secure funding for research, treatment and educational programs.

The Path Forward

An important aspect of addressing the opioid crisis is maintaining a delicate approach in tapering opioid prescriptions and not abruptly cutting off individuals who have a physical dependence. Not only are there serious withdrawal side effects that patients may experience when abruptly discontinuing opioid use, but patients also may turn to desperate measures and cheaper alternatives if they experience difficulty obtaining prescription opioids. While the utilization of opioids is beginning to level off, a resurgence of heroin use is rising. Unfortunately, this occurrence may be highly correlated—more than 80 percent of new heroin users started out by misusing prescription medications.15

As the face of the epidemic continues to transform, so does the need for tracking and predictive analytics models that evolve with ever-changing public health needs. As early innovators of data science, it’s imperative that actuaries have a seat at the table to provide insights into the opioid epidemic and public health in general. The opioid epidemic has been a powerful demonstration of the integration of actuarial sciences into clinical decision-making and health care cost-impact modeling.

Erin A. Ferries Guy, Ph.D., MPH, is public policy lead at Humana.

References:

  1. 1. National Safety Council. Preventable Deaths: Odds of Dying. National Safety Council (accessed June 15, 2020).
  2. 2. Bose Jonaki, Sara L. Hedden, Rachel N. Lipari, and Eunice Park-Lee. Key Substance Use and Mental Health Indicators in the United States: Results from the 2017 National Survey on Drug Use and Health. Substance Abuse and Mental Health Services Administration, 2018 (accessed April 30, 2020).
  3. 3. Campbell, James N. 1996. APS 1995 Presidential Address. Pain Forum 5, no. 1:85–88.
  4. 4.  Jones, Mark R., Omar Viswanath, Jacquelin Peck, Alan D. Kaye, Jantinder S. Gill, and Thomas T. Simopoulos. 2018. A Brief History of the Opioid Epidemic and Strategies for Pain Medicine. Pain and Theapy 7, no. 1:13–21.
  5. 5. Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Opioid Overdose: U.S. Prescribing Rate Maps. Centers for Disease Control and Prevention, March 5, 2020 (accessed June 15, 2020).
  6. 6. International Narcotics Control Board. Narcotic Drugs: Estimated World Requirements for 2015, Statistics for 2013. United Nations, 2015 (accessed June 15, 2020).
  7. 7. Shah, Anuj, Corey J. Hayes, and Bradley C. Martin. 2017 Characteristics of Initial Prescription Episodes and Likelihood of Long-Term Opioid Use—United States, 2006–2015. Morbidity Mortality Weekly Report 66, no. 10:265–269.
  8. 8. Ross Stephen, and Eric Peselow. 2012. Co-occurring Psychotic and Addictive Disorders: Neurobiology and Diagnosis. Clinical Neuropharmacology 35, no. 5:235–243.
  9. 9. Bride, Brian E., Samuel A. Macmaster, Siobhan A. Morse, Cayce M. Watson, Sam Choi, and John Seiters. 2016. A Comparison of Opioid and Nonopioid Substance Users in Residential Treatment for Co-Occurring Substance Use and Mental Disorders. Social Work in Public Health 31, no. 7:678–687.
  10. 10. Supra note 5.
  11. 11. Guy, Gery P., Kun Zhang, Michele K. Bohm, et al. 2017. Vital Signs: Changes in Opioid Prescribing in the United States, 2006–2015. Morbidity Mortality Weekly Report 66, no. 26:697–704.
  12. 12. Florence, Curtis S., Chao Zhou, Feijun Luo, and Likang Xu. 2016. The Economic Burden of Prescription Opioid Overdose, Abuse, and Dependence in the United States, 2013. Medical Care 54, no. 10:901–906.
  13. 13. Davenport Stoddard, Alexandra Weaver, and Matt Caverly. Economic Impact of Non-Medical Opioid Use in the United States. Society of Actuaries, October 2019 (accessed June 15, 2020).
  14. 14. The Council of Economic Advisers. The Underestimated Cost of the Opioid Crisis. WhiteHouse.gov, November 2017 (accessed June 15, 2020).
  15. 15. National Institute on Drug Abuse. Prescription Opioids and Heroin Research Report. National Institute on Drug Abuse, June 9, 2020 (accessed June 15, 2020).

Copyright © 2020 by the Society of Actuaries, Schaumburg, Illinois.