Show Summary Details

Page of

Printed from Oxford Research Encyclopedias, Economics and Finance. Under the terms of the licence agreement, an individual user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).

date: 28 July 2021

Economic Studies on the Opioid Crisis: Costs, Causes, and Policy Responsesfree

Economic Studies on the Opioid Crisis: Costs, Causes, and Policy Responsesfree

  • Johanna Catherine Maclean, Johanna Catherine MacleanEconomics, Temple University
  • Justine Mallatt, Justine MallattU. S. Bureau of Economic Analysis (BEA)
  • Christopher J. RuhmChristopher J. RuhmFrank Batten School of Leadership and Public Policy, University of Virginia
  •  and Kosali SimonKosali SimonSchool of Public and Environmental Affairs, Indiana University

Summary

The United States has experienced an unprecedented crisis related to the misuse of and addiction to opioids. As of 2018, 128 Americans die each day of an opioid overdose, and total economic costs associated with opioid misuse are estimated to be more than $500 billion annually. The crisis evolved in three phases, starting in the 1990s and continuing through 2010 with a massive increase in use of prescribed opioids associated with lax prescribing regulations and aggressive marketing efforts by the pharmaceutical industry. A second phase included tightening restrictions on prescribed opioids, reformulation of some commonly misused prescription medications, and a shift to heroin consumption over the period 2010 to 2013. Since 2013, the third phase of the crisis has included a movement toward synthetic opioids, especially fentanyl, and a continued tightening of opioid prescribing regulations, along with the growth of both harm reduction and addiction treatment access policies, including a possible 2021 relaxation of buprenorphine prescribing regulations.

Economic research, using innovative frameworks, causal methods, and rich data, has added to our understanding of the causes and consequences of the crisis. This body of research identifies intended and unintended impacts of policies designed to address the crisis. Although there is general agreement that the causes of the crisis include a combination of supply- and demand-side factors, and interactions between them, there is less consensus regarding the relative importance of each. Studies show that regulations can reduce opioid prescribing but may have less impact on root causes of the crisis and, in some cases, have spillover effects resulting in greater use of more harmful substances obtained in illicit markets, where regulation is less possible. There are effective opioid use disorder treatments available, but access, stigma, and cost hurdles have stifled utilization, resulting in a large degree of under-treatment in the United States.

How challenges brought about by the COVID-19 pandemic may intersect with the opioid crisis is unclear. Emerging areas for future research include understanding how societal and health care systems disruptions affect opioid use, as well as which regulations and policies most effectively reduce potentially inappropriate prescription opioid use and illicit opioid sources without unintended negative consequences.

Subjects

  • Health, Education, and Welfare Economics
  • Law and Economics
  • Public Economics and Policy

The Setting for the Opioid Crisis

The opioid crisis in the United States imposes challenges to population health and a range of social and economic outcomes. An argument can be made that the misuse of opioids reflects one of the most substantial public health crises that the United States has faced in modern times. While its origins are reasonably well understood, there are competing theories on why an initial period of opioid misuse developed into a crisis that has persisted for over 2 decades, has led to more deaths than any previous drug epidemic, and shows few signs of abating. Moreover, the policy efforts to address the crisis have achieved only mixed success to date. Beginning in 2020, public health attention in the United States and elsewhere shifted abruptly to the COVID-19 pandemic, but there is no sign that the opioid crisis has become less severe. Preliminary evidence indicates increasing rates of opioid mortality in 2019 (before COVID-19) and 2020, marking an unfortunate reversal in the small decrease in mortality seen between 2017 and 2018. According to the Centers for Disease Control and Prevention (CDC), 12-month counts of drug overdose deaths rose 9.1% from March 2019 to March 2020 or from 67,726 deaths to 73,860 (Ahmad, 2020), and showed a 21.3% increase from June 2019 to June 2020, as of this writing.1

A sizable body of economic research has emerged studying the origins of the opioid crisis, the factors that shaped its transformation from prescription to illicit sources, and the impact of policies and regulations that have been implemented. This review focuses on economic studies located through searches of standard sources and by soliciting feedback from economists working on opioid-related issues. These studies add to large literatures outside of economics that have made, and continue to make, contributions of utmost importance to understanding the opioid crisis.2 Further, this review mainly examines the opioid crisis within the United States. Opioids have imposed substantial costs on other countries, but the United States—to date—has arguably been hardest hit by the opioid crisis. While this article predominantly examines the U.S. experience, future work could more closely review how the crisis has developed in other countries.

Broadly, the opioid crisis can be viewed through an economic lens as resulting from an interaction between supply-side and demand-side factors, and feedback between the two. Supply-side factors provided a setting necessary for a large-scale epidemic to emerge: loose prescribing regulations combined with aggressive marketing of highly addictive substances by pharmaceutical companies. At the same time, demand-side factors related to fundamental changes in the labor market that have stalled economic growth and prosperity for specific subgroups determined who would be most impacted by the developing epidemic. Interactions between supply and demand created feedback loops that facilitated the epidemic’s growth.

Several broad conclusions emerge from the economic literature through the beginning of the 3rd decade of the 21st century. Estimates of economic costs of opioids on society are high, but the estimates of costs vary widely, largely depending on how lost life is factored into the calculations. Labor markets are both affected by the opioid crisis and may play a role in explaining its origins. Most evidence indicates that worse economic times are associated with higher opioid use, although some studies suggest that this relationship may be spurious (resulting from confounding factors) and with reverse causation (where increased opioid use depresses employment and labor force participation) possible. Origins of the opioid crisis are likely to have roots in both supply-side and demand-side factors, but with feedback loops and many different mechanisms at play in the complicated relationship between opioids and economic climate. Regardless, collectively the economic studies to date imply that labor market factors explain at most a small portion of the opioid crisis. Health care providers and the pharmaceutical industry played a critically important role in the origins of the crisis. In particular, aggressive marketing of new prescription opioids with high potential for addiction toward physicians pre-disposed to liberal prescribing of medications provided an ideal setting for the emergence of a crisis. Indeed, the literature suggests that the health care sector played a larger role contributing to the crisis than economic factors. This finding implies a more direct route for policies designed to mitigate the crisis than if its primary origins resulted from the economic climate or changes in social institutions. However, policies that simply “turn off” the supply of prescription opioids, without addressing existing addiction to these drugs, risk failure by driving substitution toward other, difficult to regulate, and riskier substances. Supply-side policies that restrict access to prescription opioids have been aggressively adopted by governments and, on the demand-side, there has substantial movement toward increasing treatment access, through relaxing regulations and insurance expansions. Some economists have also emphasized marijuana legalization as a policy that may have benefits in the form of reducing opioid use.

At the apex of the 3rd decade of the 21st century, research is needed to resolve remaining uncertainty on the effectiveness of existing policies; to evaluate policies that are only beginning to be implemented, such as those focusing on harm reduction rather than abstinence, including a possible 2021 removal of the waiver restrictions to prescribing buprenorphine; and to better understand interactions between the opioid crisis and other types of substance use disorder harms, such as the recent and rapid increase in deaths involving stimulants.3

Three Waves of the Opioid Crisis

The ongoing opioid crisis in the United States can be viewed as having occurred in three waves. The first stage reflected massive increases in the use of prescribed opioids and dates from the mid-1990s through 2010. This period was characterized by a loosening of restrictions on the prescribing of opioid painkillers and extensive marketing of them to both health care providers and consumers. The second wave, from 2010 to 2013, was distinguished by extensive growth in heroin use and associated deaths, although problems related to prescribed opioids remained substantial during this period. The third wave beginning in 2013 and continuing through at least 2020, has been characterized by surging deaths and problems related to the use of synthetic opioids, particularly fentanyl and its analogues.

The first wave of the opioid crisis is thought to have begun shortly after the 1996 approval and release of Purdue Pharma’s soon-to-be blockbuster drug OxyContin. Prior to the mid-1990s, opioid prescribing was surrounded by a culture of “opia-phobia,” as opioid painkillers have a long history of misuse in the United States reaching back to the Civil War (Macy, 2018). Throughout most of the 20th century, physicians were reluctant to prescribe opioids for pain management, even to cancer patients and to the terminally ill (Hill Jr., 1993; Paice et al., 1998; Quinones, 2016; Weissman, 1993). In 1996, Purdue Pharma introduced a new generation of opioids with the launch of OxyContin, an oxycodone product with an extended-release mechanism. Purdue Pharma’s marketing approach was remarkable: the firm’s promotional budget was magnitudes larger than those of competitor firms, and far higher than any previous opioid marketing campaign. In addition, the campaign specifically targeted health care providers who already displayed high rates of opioid prescribing despite the prevailing culture of opia-phobia (Alpert et al., 2019).

To further change prescribers’ attitudes, Purdue Pharma and other opioid manufacturers funded the American Pain Society, which lobbied the Joint Commission to prioritize the treatment of pain (McGreal, 2018). The Joint Commission is a nonprofit accreditation organization whose certification is used by the Centers for Medicare and Medicaid Services (CMS) to determine the reimbursement status of medical facilities nationwide. In 2001, the Joint Commission implemented a pain scale, with pain being assessed as the “fifth vital sign,” along with blood pressure, pulse, respiration rate, and temperature. More specifically, health care providers were required to incorporate pain into patient assessments and, where appropriate, encouraged to treat pain symptoms medically, including through the use of prescription opioids. These actions played an important role in the emerging lax culture of opioid prescribing that characterized the first decade of the 2000s. By 2012, 259 million prescriptions for opioids were dispensed and approximately one in four Americans were prescribed an opioid medication each year (Kilby, 2016; Mallatt, 2019). Misuse increased in tandem with prescribing. By 2014, 1.6% of the population aged 12 or older reported using pain relievers nonmedically in the past month (Hedden et al., 2020).

There were an estimated 3,442 deaths involving prescription opioids (not counting synthetic opioids or heroin) in 1999; this number increased to approximately 15,000 overdoses at the beginning of the second wave of the crisis in 2011.4 Similarly, diagnosis rates of substance use disorder (SUD) grew by a factor of 6 between 1999 and 2009 (Paulozzi et al., 2011).5 Opioid prescribing peaked between 2010 and 2012 (Pacula & Powell, 2018; Schuchat et al., 2017).

The crisis then evolved to a second wave where deaths and other adverse consequences associated with heroin use dramatically increased. Alpert et al. (2018) and Evans et al. (2018) show that this transition was, in part, fueled by a reformulation of OxyContin in August of 2010. The reformulated medication was crush-resistant and therefore harder to snort or inject. Additional government policies targeting the supply of opioid prescriptions also led to rising heroin use (Mallatt, 2020b). Heroin overdoses tripled between 2010 and 2013, whereas prescription opioid overdoses not involving heroin plateaued or subsided slightly. The overall effect of these market and policy changes was to increase the total number of fatal opioid overdoses.

There are differences across the U.S. in terms of the source of heroin. This heterogeneity has important implications for policy efforts aimed at curbing use of this substance, and suggests that there are disparities in the type and in the potentially harmfulness of heroin consumed across the country. On the East Coast, heroin has historically been imported in white powder form from South Asia while consumers on the West Coast generally obtained black tar heroin sourced from Mexico (Abouk et al., 2019). Cutting agents and fentanyl are more easily incorporated into powdered heroin in the East and are more dangerous than pure heroin. Thus, the East Coast has suffered disproportionately from a large spike in synthetic opioid overdoses, beginning in 2014. While the source of heroin has remained relatively stable on the West Coast over the various stages of the crisis through the early 21st century, East Coast consumers experienced a change in the source of heroin in 2015 when Mexican cartels became the major suppliers of heroin in this area, effectively pushing South Asian suppliers out of the market (United States Department of Justice, 2018b).

Beginning in approximately 2013, the opioid crisis entered its third wave, characterized by a switch from the misuse of heroin and illicit semi-synthetic opioids to the misuse of drugs containing highly-potent, fully synthetic opioids. The U.S. Drug Enforcement Agency (DEA) began reporting the presence of fentanyl and its analogues in seized heroin in 2013.6 Fentanyl is an extremely potent synthetic opioid that offers several advantages over heroin to suppliers. In particular, fentanyl is relatively cheap to produce, and easier to transport and smuggle since smaller quantities are required. The production of heroin necessitates the relatively expensive, time-intensive, and conspicuous growth of opium poppies. In contrast, fentanyl and its chemical analogues are completely synthesized from ingredients in a lab. Many input components are imported from China to Mexico, where the drug is then synthesized in labs and smuggled across the U.S. border (United States Department of Justice, 2018b). An experienced opioid user may require 1 or 2 grams of heroin per day (depending on purity), but equivalent dosages of fentanyl are measured in micrograms (mcg), which saves distributors the costs and detection risks of transporting many kilograms of heroin through border checkpoints.

Data from the “darknet” (i.e., the non-trackable Internet) suggest that the wholesale price of fentanyl and its analogues is much lower than heroin, with a 100 mcg dose costing about four cents in a wholesale setting. A wholesale kilogram of heroin (typically 50% pure at this stage in the production process) costs between $33,000 and $100,000 (United States Department of Justice, 2018a), while an equivalent 10 grams of fentanyl costs approximately $8,060 to $10,400 online, with prices falling rapidly from 2014 to 2016 (Miller, 2020). Distributors began cutting relatively expensive heroin with much cheaper fentanyl to lower production costs. Anecdotal evidence suggests that early-adopting dealers temporarily captured market share, after which other dealers also began adding fentanyl to their product to remain competitive. According to the DEA the addition of fentanyl to heroin poses “a higher risk of overdose to even the most experienced opioid users.” Fentanyl is also available in diverted prescription form (such as Actiq and Duragesic), but is relatively expensive and thus rarely a source of illicitly traded fentanyl (Lamy et al., 2020).

Fentanyl was involved in fewer than 6,000 overdose deaths in 2014; compared with 9,781 and 19,880 deaths in 2015 and 2016, respectively. By 2017 this drug was associated with 29,131 deaths. The number of fatal fentanyl overdoses rose more slowly, to 32,039 in 2018, but then accelerated to 37,231 in 2019, and continued to rise in the early months of 2020 (Centers for Disease Control and Prevention, 2020b).

Background on the Opioid Crisis: Societal Costs

Overall Prevalence and Costs

Data from the National Survey of Drug Use and Health (NSDUH)—the official government source for substance use statistics in the United States—indicate that in 2018, 1.7 million Americans met diagnostic criteria for prescription opioid use disorder (OUD) and over 500,000 for heroin-related OUD (McCance-Katz, 2018). These numbers represent a lower bound on the true prevalence of OUD, as individuals are likely to under-report this condition in survey settings and since the NSDUH excludes groups likely to have disproportionately high rates of OUD (e.g., institutionalized and homeless individuals). Between 1999 and 2018, nearly 450,000 Americans died from an opioid overdose (McCance-Katz, 2020). Opioid misuse and overdoses impose additional costs on the health care system in the form of emergency department visits, direct treatment costs, and expenses associated with neonatal abstinence syndrome (NAS). Nonmedical costs include those related to the criminal justice system, lost worker productivity, and many others. Economists have estimated total costs, using insights from economic cost studies in other contexts. Total societal costs of opioid misuse had been calculated at $55 billion in 2007 and $78 billion in 2013 (Birnbaum et al., 2011; Florence et al., 2016), but these estimates exclude the costs of premature mortality. Taking premature mortality into account, as well as some other costs, yields much higher estimates of $504 billion in 2015 from the Council of Economic Advisers (White House Council of Economic Advisers, 2017) and, more recently, over $1 trillion in 2017 (Florence et al., 2020).

Demographic Trends in Overdose Mortality

The opioid crisis has not affected socioeconomic and demographic groups equally; thus, it is useful to summarize some broad patterns related to age, gender, and race. Overall drug overdose death rates are higher among men than women; about two thirds of opioid overdoses occur among males. However, the gender gap in overdose is smaller for prescription opioid deaths (relative to all and illicit opioids), with men making up only 59% of these opioid fatalities. Non-Hispanic white Americans suffer disproportionately from prescription overdoses, with a death rate of 17.5 deaths per 100,000 population in 2016 (Scholl et al., 2019). American Indians and Alaska Natives have the second highest rate of opioid overdose deaths, but data on these demographic groups are sparse (Rudd et al., 2014; Scholl et al., 2019). Black and Hispanic Americans have traditionally been somewhat less affected by the crisis, perhaps due to under-treatment of pain within these groups, which may have inadvertently protected them from opioid initiation and overuse (Alexander et al., 2018; Frankt & Monkovic, 2019). However, fatal overdose rates are rising among Blacks in the third wave of the epidemic as fentanyl is increasingly concentrated in urban areas, where Black Americans are disproportionately likely to reside. From 2015 to 2017, opioid mortality rates rose especially quickly among Black Americans aged 45 to 64 years in large metro areas (Lippold, 2019). From 2011 to 2016, the age-adjusted rate of overdose deaths involving fentanyl grew the most in the Black population, reaching 140.6% per year, while this rate rose 108.8% and 118.3% annually for non-Hispanic whites and Hispanic Americans, respectively (Spencer et al., 2019).7 Older populations report far less opioid misuse (McCance-Katz, 2018) and lower rates of opioid overdose deaths corroborate these survey responses, whereas individuals aged 18 to 59 suffer relatively higher opioid fatality rates (Centers for Disease Control and Prevention, 2020a).

The risk of opioid overdose is also positively correlated with a myriad of other demographic characteristics, including being disabled, unmarried or widowed, unemployed, uninsured, incarcerated, having low education, being a citizen (in comparison to a noncitizen), renting rather than owning a home, residing in a non-rural area, and having a low income. Residents of South Atlantic States and Mountain States have relatively high rates of overdose.8 This heterogeneity in prevalence rates across population subgroups provides important context for understanding hypotheses related to origins of the crisis and potential impacts of policies.

Labor Market Impacts

In theory, opioids could improve labor market outcomes through better management of chronic pain symptoms and, in turn, enhanced work capacity. For example, Cox-2 inhibitors—which are used to treat chronic pain and thus may have comparable effects to opioids—have been found to decrease sickness days among workers with joint pain (Bütikofer & Skira, 2018). As a specific example, Garthwaite (2012) finds that removal of Vioxx (a particularly effective Cox-2 Inhibitor) from the market in 2004 led to a 0.35 percentage point reduction in overall labor force participation and $19 billion in lost wages in the year following removal.9 However, labor market outcomes could worsen as prescription opioid use increases if addiction or other related problems—for example, dizziness, nausea, and sedation—reduce work capacity. Opioid use can also raise health insurance costs due to direct expenditures on prescriptions for this medication, and other (indirect) increases in health care costs.10 In this section, studies in which the direction of causality operates from the overuse of opioids to labor market outcomes are reviewed.

Most studies to date suggest that increased opioid use leads to worse labor market outcomes. As background, the labor force participation rate for prime working aged men (i.e., 25 to 54 years) has been declining since the 1970s, that is, prior to widespread use of opioids. However, opioids could have potentially exacerbated this trend. For example, in the time period 2014 to 2016, Krueger (2017) uses descriptive analysis to show that half of surveyed prime working aged men who were out of the labor force reported using pain medication on any given day, and about one third took prescription opioids. Using quasi-experimental techniques, Harris et al. (2020) demonstrate that the prevalence of high rates of opioid prescribing decreases labor force participation rates. Further, Deiana and Giua (2018) show that restricting access to opioids through state-level policies has the potential to increase labor force participation, while Beheshti (2019b) establishes that national-level restrictions on hydrocodone prescribing, implemented in 2014, caused improvements in labor market outcomes. Similarly, Aliprantis et al. (2019) find that higher prescribing rates are associated with lower labor force participation for prime working aged men and women, with particularly large reductions for persons without a four-year college degree. Park and Powell (2020) find that the crisis’s transition to illicit opioids led to reduced labor market engagement and to rising rates of disability applications and enrollment. Rietveld and Patel (2020) demonstrate that additional opioid exposure is negatively associated with measures of entrepreneurship and small-business formation. Using firm-worker matched data from Denmark, Laird and Nielsen (2017) find that additional opioid prescribing decreases labor force participation and income. Similarly, Savych et al. (2018) demonstrate that longer-term opioid prescribing increases the duration of temporary disability spells among those receiving workers’ compensation benefits in the United States.

However, other studies contest the assumption that increased opioid exposure leads to adverse outcomes in the labor market. Currie et al. (2018) uncover a positive relationship between prescribing and employment-to-population rates among women, but no association among men. Analyses of survey data on individuals indicate that SUDs are not associated with transitions from full- to part-time work, or with on-the-job problems with coworkers, or financial strain (Baldwin & Marcus, 2014; Maclean et al., 2015). These findings point to a possibility that proper use of opioids for pain treatment may help workers remain on the job or return to work.

Effects on Family Life

There is also evidence of the opioid crisis imposing costs in the sphere of family structure and child well-being. The number of children living with an adult with an opioid use disorder increased 30%, and those with an adult using heroin rose 200%, between 2002 and 2017 (Bullinger & Wing, 2019). Buckles et al. (2020) find that as a result of the opioid crisis, the fraction of children living away from a parent and in a household headed by a grandparent increased. Gihleb et al. (2020) document rising rates of child removals and neonatal abstinence syndrome (NAS) over the course of the opioid crisis. From 2000 to 2015, the number of foster care cases related to drug misuse increased 66%. The rate of NAS, where infants experience opioid withdrawal symptoms due to in-utero exposure to opioids, has grown by 500% since 1999. The U.S. spends $1.5 billion on treating NAS-related complications each year as of 2021. Evans et al. (2018) and Evans et al. (2020) illustrate rising rates of child maltreatment cases due to increased opioid misuse.

Effects on Crime

The opioid crisis seems to be less associated with violent and property crime, in the popular press, than the heroin crisis of the 1970s, the crack and cocaine crisis of the 1980s and early 1990s (Fryer et al., 2005; Pollack & Reuter, 2014), or the methamphetamine crisis that occurred in the 2000s (Dobkin & Nicosia, 2009; Plüddemann et al., 2010). Nevertheless, there is evidence linking both violent and property crimes with opioid prescriptions (some of which are diverted to other users) and heroin use. In line with this hypothesis, Bondurant et al. (2018) and Wen et al. (2017) find that expanding access to SUD treatment during the opioid crisis decreases both violent and financially-motivated crimes.11 Dave et al. (2018) document that reducing the supply of prescription opioids decreases violent crime.

Origins of the Crisis

Societal Causes

Famously documented in Case and Deaton (2015), “deaths of despair”—that is, deaths from drug overdoses, suicides, and chronic liver disease—drove a surprising reversal in the decline of midlife mortality among non-Hispanic white Americans in the early 21st century. White mortality patterns began to reverse (from a declining trend to an increasing trend) around the year 2000. This pattern is unique to the United States, as other Organization for Economic Cooperation and Development (OECD) countries continued to experience midlife mortality rate reductions during the same time period. Drug deaths are by far the most important contributor relative to suicides and liver disease, accounting for at least three quarters of the overall effect. Among white men aged 25 to 55 years without a college education, the death rate from drugs, suicides, and chronic liver disease was 125 to 150 deaths per 100,000 population in 2017, a 250% increase over 1992 rates (Leonhardt & Thompson, 2020).

In the late 20th and early 21st centuries, the United States experienced wage stagnation among less educated workers due to automation, import competition, increased market concentration, outsourcing, productivity clusters in abstract tasks, and weakening unions, which, in combination, benefited workers in cities and the highly educated, while posing a disadvantage to those in less connected geographies and the less educated. The United States has also seen declines in the labor force participation rate, increases in mental illness, disability, and chronic pain, as well as falling family and community engagement. Some scholars contend that these changing economic factors have negatively impacted specific groups of society, leaving them without hope of regaining economic stability and their position in society. In response, members of these groups have turned to substances as a means to self-medicate their pain. Deaton argued in his 2020 testimony to Congress: “[Deaths of despair] respond more to prolonged economic conditions than to short-term fluctuations, and especially to the social dysfunctions, such as loss of meaning in the interconnected worlds of work and family life, that come with prolonged economic distress” (Deaton, 2020). Research supports the hypothesis that economic factors have an impact on opioid overdoses. Declines in local manufacturing employment depress wages and employment, and increase opioid overdose deaths (Charles et al., 2019). Trade liberalization with China has caused higher rates of suicide and fatal drug overdoses, especially among white males (Pierce & Schott, 2020). Similarly, increased exposure to automation in the labor market has reduce economic opportunities. (Venkataramani & O’Brien, 2020) show that this exposure led to increased drug overdose mortality, with particularly large effects for men ages 30 to 54 years residing in U.S. counties with high shares of manufacturing jobs. Opioid overdoses are also impacted by short-term labor market shocks such as a rise in the unemployment rate and plant closures: a worsening economic climate appears to increase overdoses (Hollingsworth et al., 2017; Venkataramani et al., 2020). Maclean et al. (2020) show that heroin-related admissions to SUD treatment decline during economic downturns, but the authors find no evidence that prescription opioid-related treatment admissions vary with changes in the business cycle. Similarly, counties that experience growth in industries that employ white males experience fewer opioid overdoses, implying protective effects of industry-specific growth for them (Betz & Jones, 2018).

Dow et al. (2020) find that increases in the federal minimum wage and increases in the Earned Income Tax Credit (EITC) decrease rates of suicide (but not fatal drug overdose) among Americans without a college education, implying that economic assistance may assuage some forms of despair. Although multiple studies indicate a role for economic indicators in determining adverse opioid outcomes, the magnitudes of the predicted effects are small relative to the overall increases in opioid prescribing and overdoses, and are therefore unlikely to be key determinants of the crisis.

Few studies examine how changes in culture or social cohesion affect opioid outcomes, which implies a substantial gap in the literature. In large part, the paucity of research on these conceptually important factors may reflect the difficulty in obtaining detailed and well-measured data on them. Identification using standard econometric methods may be challenging. In particular, since measurements of both economic and cultural well-being have deteriorated for some groups since the early 1970s, separating the causal effects of each may be difficult. In addition, underlying forces may influence both economic and cultural factors, further raising the difficulty of isolating causal effects from spurious correlations. Finally, there is plausible endogeneity in the relationship of worsening economic conditions, declining family stability, and decreasing civic engagement for whites without a college degree. These phenomena have no single clear cause, but econometric identification requires that researchers use believably exogenous shocks to well-measured variables. Research identifying the effect of economic change using small temporary economic shocks (typically by including time fixed effects that account for macro trends) and the resulting local average treatment effects may not be capturing the pervasive influence of these more opaque generational trends.12

While several studies, previously mentioned, suggest a causal relationship between economic conditions and opioid overdoses, other research indicates that economic factors have only limited effects on drug, suicide, and chronic liver disease deaths. Although counties that experience relative economic decline also had higher growth in drug mortality, Ruhm (2019a) finds that the relationship is greatly mitigated (sometimes to zero) by controlling for confounding factors and for allowing for selection on unobservables.

In other words, economic conditions are correlated with population patterns of behavior, and short-run and medium-run economic woes alone do not explain the bulk of the variation in opioid overdose across the United States Similarly, Currie et al. (2018) argue that there is little or mixed connection between counties’ employment-to-population ratio and opioid prescriptions. In particular, the authors find a weak link among women and no relationship among men. Currie and Schwandt (2020) further illustrate that labor market opportunities of the 2nd and 3rd decades of the 21st century do not explain a substantial part of the opioid crisis, and urge attention toward policies aimed at addressing the opioid epidemic itself.

Other research finds results that run contrary to the hypothesis that poor economic outcomes cause more opioid misuse or other negative health outcomes. Metcalf and Wang (2019) show that mining employment is actually positively correlated with opioid misuse—that is, counties realizing a decrease in this type of employment experience lower rates of opioid overdose than those that are also reliant on mining but do not experience the negative economic shock. Evidence from Denmark shows that industry export growth increases on-the-job injury and elevates stress levels and rates of depression, heart attacks, and strokes (Hummels et al., 2016); this finding conforms with other evidence of health benefits during economic downturns (Ruhm, 2000). Furthermore, workers in more injury-prone industries that experience employment growth receive more opioid prescriptions as their industry expands (Musse, 2020). Therefore, even industries that gain from trade in the form of export or overall economic growth more generally may see increased rates of opioid use through the channel of rising rates of injury or illness incurred while working.

To summarize, the opioid crisis can be viewed through an economic lens as resulting from an interaction between supply-side and demand-side factors, and feedback between the two. At the most fundamental level, supply-side factors related to the increased production and marketing of opioids, combined with changes in prescribing patterns and policies (e.g., the Joint Commission listing pain as the fifth vital sign) were the proximate cause. These factors explain why the crisis began, fairly suddenly, in the late 1990s, after the approval of OxyContin and pharmaceutical industry efforts to normalize the prescribing of opioids. Conversely, demand-side factors such as skill-biased technological change and import competition are unable to account for the timing of the increase in opioid prescriptions and the resulting adverse consequences. However, demand-side factors play an important role (albeit abetted by the pharmaceutical industry’s targeted marketing efforts to health care providers most likely to prescribe opioids on the part of suppliers) in determining who was most affected. Moreover, the mutually reinforcing relationships between economic conditions and opioid use may create a feedback loop of economic and personal despair. In this regard, the crisis can usefully be viewed as a cyclical, reinforcing model of despair and self-medication through opioid misuse that was acted upon by outside forces, including public policy, the changing prescribing culture, and possibly other factors that have not yet been examined in the literature.

Contributions of Health Care Institutions

Particularly important are the effects of additional exposure to opioids and to physicians who prescribe opioids more aggressively. Finkelstein et al. (2019) study those who move between U.S. counties to assess the effect of the local prescribing environment on opioid use. The authors show that moving from a county that has a relatively low rate of top-prescribing physicians to one with a relatively high rate is associated with additional overdoses for migrating individuals. The results imply that 30% of the variation in opioid deaths across counties is explained by place-specific effects, namely physician prescribing behavior. Additionally, Schnell and Currie (2018) find that physician education plays a role in opioid prescribing styles. In particular, those physicians with medical degrees from low-ranked schools prescribe at higher rates than otherwise similar physicians. The finding by Schnell (2018) that physicians differ dramatically in their response to the reformulation of OxyContin further serves to underline the role played by physician prescribing in the crisis. Opioid marketing to doctors also has an impact on how different states experience the opioid crisis. Alpert et al. (2019) and Nguyen et al. (2019) show that pharmaceutical company marketing efforts increased prescribing by physicians. More generally, Lin et al. (2020) demonstrate that geographic area variations in the intensity of medical care provided are positively associated with opioid death rates. Barnett et al. (2017) demonstrate that plausibly exogenous physician assignment in the emergency department affects the probability of long-term opioid prescriptions; Medicare beneficiaries (who had not received an opioid prescription in the 6 months prior to the visit) who are attended to by emergency department physicians with higher opioid prescribing patterns are significantly more likely to receive long-term opioid prescriptions than their similar counterparts who are assigned an attending physician with more typical prescribing patterns. Low-cost interventions to inform practitioners of their prescribing patterns have had limited success in curbing behavior (Sacarny et al., 2016).

Policy Responses to Prescription Opioid Misuse

Governments at all levels have undertaken a range of policy approaches in attempts to curb the opioid crisis. The nature of these policies has changed over time, corresponding to some extent with the types of opioids targeted, and as the character of the crisis has shifted from prescription to illicit opioids. While some localities have departed from the general trend, early policies tended to focus on curtailing the supply of prescription opioids, or raising the financial or time costs of accessing these substances. Later policies have typically emphasized demand-side factors, such as the ability of drug-seeking individuals to obtain prescriptions from multiple health care providers, and harm reduction efforts, such as naloxone access laws.13 Which type of policy response is more likely to be more effective is unclear ex ante. There are also potentially synergies between policies, suggesting gains to implementing multiple complementary efforts.

A rapidly growing literature in economics and policy evaluates several of these regulations. This research suggests the desirability of supply-side and demand-side policies, as well as the importance of considering relevant local factors that may interact with both types of policies; for example, see Pacula and Powell (2018), and Saloner and Barry (2018) for an excellent discussion on this topic, and Barry and Frank (2019), Mauri et al. (2020), and Schuler et al. (2020) for comprehensive recent reviews of state and federal policy studies related to opioid outcomes. Of note, most supply-side policies to date have focused on reducing opioid prescribing, opening the door to possible substitution toward illicit drugs (such as heroin) that are harder to regulate and, arguably, more harmful. Conversely, demand-side policies such as increasing OUD treatment access, often address outcomes that may impact underlying addiction to opioids.

Prescription Drug Monitoring Programs

One of the earliest policies implemented by states, and currently by far the most common, is the prescription drug monitoring program (PDMP). PDMPs were adopted as early as 1939, in California, and were designed to reduce the misuse of prescription medications generally, not specifically opioids (Holmgren et al., 2020). A PDMP is a centralized database containing patient scheduled14 prescription medications.14 PDMPs are designed to increase information available to health care providers related to patients’ history with medically obtained prescription opioids. By 2017, all states had a PDMP in operation (Holmgren et al., 2020).15 Conceptually, pharmacists enter information into the database when patients are dispensed controlled prescription medications, including opioids. Health care providers then have the ability to access the patient’s historical use of opioids and other controlled substances to assist in their decision whether to prescribe controlled medications. The hope is that health care providers will then identify individuals who are potentially misusing opioids and limit access of the drugs to these individuals, thereby curbing misuse without reducing access to medications for legitimate patients. For example, doctor shopping can be identified and prevented by reviewing patient histories.

For PDMPs to reduce opioid misuse, health care providers must actually use the database by checking the patient’s history before prescribing medications. Several features of the earliest PDMPs may have stifled their effectiveness. All PDMPs require pharmacists to enter controlled substance histories into the database. However, the literature has distinguished between whether the law mandates that health care providers check the database at the time of prescribing or dispensing medications. PDMPs with this requirement are often referred to as “mandatory” PDMPs (vs. “voluntary” PDMPs, which do not have that requirement). Since many health care providers contend that the act of checking or entering information in the database is burdensome, prescribing providers often did not engage with PDMPs, as the initial programs did not compel them to do so. Further, early PDMPs were not electronic, adding to the administrative burden of using the system among health care providers. Given this backdrop, some providers pushed back on PDMP adoption because of the hassle of utilizing them, such as difficulty in obtaining logins, the database not being accessible (i.e., the platform being “down”), and incomplete data (Haffajee et al., 2015; Lin et al., 2017; Young et al., 2017). Beginning in 2007, several states adopted arguably stronger PDMPs, the previously noted “mandatory” PDMPs, which required health care providers to check the database prior to prescribing. Conceptually, such mandatory PDMPs should have more impact than voluntary systems, as health care providers are legally bound to query the system.

A number of economic studies have examined the impact of PDMPs, both those that do and do not mandate health care professionals to check the database, on opioid misuse. Those examining broad measures of a PDMP that do not distinguish between voluntary and mandatory access programs (“voluntary” PDMPs) generally suggest they have a limited impact. For example, using a sample of disabled non-elderly Medicare beneficiaries, Meara et al. (2016) do not observe changes in various opioid misuse outcomes following adoption of a PDMP. Similarly, Buchmueller and Carey (2018) find that among Medicare enrollees, voluntary PDMPs are not effective in reducing patterns of harmful opioid behavior, whereas mandatory PDMPs are effective. However, there is not full consensus, with some studies of voluntary PDMPs showing effectiveness. For example, Kilby(2016) demonstrates a 12% reduction in opioid-related mortality post–voluntary PDMP, and a 10% decline in prescribing among patients with employer-sponsored insurance. Further, other studies suggest that voluntary PDMPs have larger effects on the Medicaid population (Bao et al., 2016) and on users with previous patterns of behavior indicative of opioid misuse (Mallatt, 2019). The limited impact of many voluntary PDMPs is perhaps not surprising, as database use is often relatively infrequent among health care providers. For example, in states where PDMP use is not mandatory, roughly 14% to 25% of health care providers utilize the system (Alexander et al., 2015).

Several studies from the 2nd and 3rd decades of the 21st century empirically distinguish between voluntary and mandatory access PDMPs, and provide more definitive evidence that the latter programs reduce prescription opioid misuse (Ali et al., 2017; Bao et al., 2018; Buchmueller & Carey, 2018; Deiana & Giua, 2018; Grecu et al., 2019; Kaestner & Ziedan, 2019; Wen et al., 2019; Ziedan & Kaestner, 2020). For example, Grecu et al. (2019) show that, following adoption of a mandatory PDMP, the number of admissions to substance use disorder treatment related to opioid use decreases by 20% to 25%, although reductions only emerge 2 years post-policy. The authors also document an age-gradient in the estimated treatment effects, with admissions rates decreasing by 32%, 17%, and 12% among those aged 18 to 24, 25 to 44, and 45 years or older, respectively.

In addition to changes in direct effects on opioid use, research suggests that mandatory PDMPs lower crime (Dave et al., 2018; Deiana & Giua, 2018), improve children’s birth outcomes and decrease NAS (Gihleb et al., 2020; Ziedan & Kaestner, 2020), and reduce foster care admissions (Gihleb et al., 2019). The assumed mechanism is that decreased opioid use leads to these improvements in children’s outcomes. A full consensus has not yet been reached, with some studies showing opposing findings. For example, Mallatt (2018) finds that PDMP adoption leads to increased heroin-related crime in counties with high rates of opioid use.

There may be a feedback loop induced by PDMPs whereby pharmaceutical companies reduce prescription opioid promotions following a mandatory PDMP, which, in turn, lowers demand for these medications (Nguyen et al., 2019). PDMPs also appear to directly decrease health care provider prescribing of opioids. Using rich health insurance claims data, Sacks et al. (2019) show that opioids dispensed to opioid-naive users decline following adoption of a mandatory PDMP. Since the databases contain relatively little information on new users, these results suggest that PDMPs instead reduce prescribing for other reasons, the most likely being hassle costs of using the system (Bachhuber et al., 2018).16

Using unique data from Kentucky, which adopted a mandatory PDMP, Alpert et al. (2020) similarly suggest that reductions in physician opioid prescribing may be at least partially attributable to the hassle costs associated with using the database. Using claims data, Mallatt (2019) shows that voluntary PDMPs target users displaying signs of opioid misuse.

While studies establish that mandatory PDMPs reduce prescription opioid use and improve some associated outcomes, there could be positive or negative spillover effects on the use of other prescription medications, addictive substances, or non-drug pain therapies. For instance, if prescription opioids and other addictive substances are economic complements, then PDMP-attributable reductions in opioid prescribing should decrease the use of other substances.17 However, if prescription opioids and other drugs instead are instead economic substitutes, then consumers will switch to the substitute drug as new obstacles to obtaining prescription opioids emerge. Cawley and Ruhm (2011) provide a careful discussion of the relationships between substances. Whether substitution improves or worsens public health will be determined by the relative harms of the involved drugs. Several studies provide evidence of such substitution. For example, Grecu et al. (2019) show that opioid-related admissions to SUD treatment programs decline following adoption of a mandatory PDMP, and that there are also decreases in admissions for cocaine and marijuana use, which could be economic complements to opioids. However, Mallatt (2020b) finds that consumers substitute to heroin following the establishment of voluntary PDMPs, with particularly strong evidence in localities with high prior levels of prescription opioid use. This result is concerning if heroin use is more harmful than the consumption of prescription opioids; heroin is likely to be more dangerous because of the method of consumption (e.g., injection rather than swallowing pills) or if users are less able to monitor the potency of the dose they are ingesting.

While there is debate as to whether a PDMP must be mandatory (vs. voluntary) to impact outcomes, some researchers argue that this discussion is misplaced. Instead, these researchers argue that this program feature is not relevant for PDMPs to reduce prescription opioid use. Wang (2020) shows that PDMPs (whether voluntary or mandatory) reduce opioid misuse when they are adopted in combination with a state-level health integration technology policy that promotes the sharing of health records electronically. Kaestner and Ziedan (2019) find that adoption of a mandatory PDMP reduces sales of scheduled prescription drugs. However, when also controlling for whether the PDMP is electronic (defined by the authors as “systems that are not paper-based and allow the prescriber to transmit the prescription information electronically to the state authority”), Kaestner and Ziedan (2019) find that the coefficient estimate on the mandatory PDMP variable becomes statistically indistinguishable from zero. They interpret these results to imply that the salient feature of the PDMP is whether the database is electronic, not whether health care providers are mandated to check it.

Overall, while there are many studies on PDMPs, the literature has not yet reached full consensus on the importance of PDMP implementation and design, leaving scope for future work in this area. There is also opportunity for new research studying how supply-side restrictions affect substitution toward other non-drug medical therapies and and how to ensure that policies do not pose undue burdens on underrepresented minority populations (Substance Abuse and Mental Health Services Administration, 2020).

Other Supply-Side Policies

There are a number of less studied state-level supply-side interventions. One important category is pain management clinics laws (PMCLs), which establish minimum requirements that pain management clinics must meet in order to dispense prescription drugs. Broadly, PMCLs are organized to prevent the emergence or operation of “pill mills”—medical clinics that knowingly and willingly dispense prescription drugs to illegitimate consumers. Pill mills were especially notorious in Florida, which was considered the epicenter of the opioid crisis in the early 2000s. By preventing the ability of nefarious clinics from emerging or allowing authorities to shutter such clinics, the PMCLs reduce both the overall supply of prescription opioids and the extent of diversion to illegitimate users.

PMCLs involve many separate requirements for clinics and doctors’ offices, with variation in the specific stipulations across states. Each set of laws specifies which facilities are classified as pain management clinics, typically citing prescribing patterns or advertising practices that are characteristic of pill mills. The packages of laws then add more requirements aimed at reducing prescribing within these clinics or shutting them down altogether.

Twelve states had passed legislation targeting pill mills by the beginning of the 3rd decade of the 21st century. These policies appear to be effective in reducing prescription opioid use. Using government data on sales of scheduled prescription medications, Ziedan and Kaestner (2020) document 15% to 50% declines in prescription opioid sales after implementation of a PMCL. Evidence from Florida and Texas suggest these targeted efforts are effective at curbing prescribing and reducing harmful secondary outcomes like overdoses (Chang et al., 2016; Lyapustina et al., 2016; Mallatt, 2020b; Meinhofer, 2016; Rutkow et al., 2015). Meinhofer (2016) shows that the substantial crackdown in Florida caused the number of active pain clinic licenses to fall from 988 in 2010 to 407 in 2012. Additionally, quantities of prescribed opioids decreased by 59%, opioid-related admissions to drug treatment facilities increased by 33%, and overdose rates declined.

Florida was also unique in that the DEA arrested many offending prescribers during that state’s crackdown on pain clinics, whereas pill mill laws in other states did not include a substantial law enforcement component.

Chang et al. (2016) found that, prior to the pill mill legislation, the top 4% of opioid prescribers in Florida were responsible for 67% of total opioid prescriptions in the state. After the pill mill legislation and law enforcement efforts were implemented, the high risk providers saw fewer patients and prescribed fewer opioids. Lyapustina et al. (2016) indicate that the Texas pill mill law reduced opioid prescribing by between 8% and 24%. Using nationwide data on business establishment counts, Mallatt (2020a) shows that the laws cause 10% to 15% of establishments within a certain class of specialty clinics (including pain clinics) to exit, equivalent to 23 fewer pain clinics for each treated state. These studies reveal that pill mill laws operate at both the extensive margin, driving pain clinics to exit treated counties and states, as well as the intensive margin, as top prescribers prescribe fewer opioids and see fewer patients.

PMCL adoption appears to have positive spillovers to labor market and crime outcomes. Kaestner and Ziedan (2019) find that employment and earnings improve modestly post-law, while Deiana and Giua (2018) demonstrate that labor market participation rates increase and unemployment rates decline. Further, Mallatt (2020b) shows that PMCLs are negatively associated with crime rates related to heroin possession and dealing.

Many states have recently passed laws limiting the length of initial prescriptions for opioids (typically to 7 days). In considering the impact of these policies, Sacks et al. (2019) unexpectedly show that such policies increase the overall amount of opioids prescribed to new users. This contrary result is driven by the reduction in the length of prescriptions that is more than offset by increases in the frequency of short prescriptions. Determining whether the net effect is harmful or beneficial depends on the relative risks of growth at the extensive margin (frequency of prescriptions) versus reductions at the intensive margin (length of prescriptions).

An important policy that substantially altered the supply of prescription opioids within the United States was the August 2010 reformulation of OxyContin by Purdue Pharma. From its introduction to the market in 1996 through early August 2010, OxyContin, an extended-release version of oxycodone that was often prescribed in high doses, was one of the most commonly misused prescription opioids (Cicero et al., 2005). One problem with the original formulation was that consumers often crushed or dissolved the pills and then inhaled or injected the drug in a more intoxicating form, thereby circumventing the extended release mechanism occurring with oral ingestion. Under pressure from the Food and Drug Administration, Purdue Pharma released a reformulated version of OxyContin that was more difficult to crush or dissolve. The company also quickly (within days) discontinued the original version, thereby abruptly shutting off access to the previous, easy-to-abuse formulation of OxyContin. There were high hopes for the potential of misuse-deterrent reformulations to reduce the injection, snorting, crushing, or chewing of prescription opioids (White et al., 2009).

However, the OxyContin reformulation had unintended spillovers into markets for illicit drugs. While the exogenous and sudden supply shock markedly reduced the use of this opioid, there was substantial and rapid substitution to heroin by consumers. Alpert et al. (2018) conclude that areas with high underlying rates of OxyContin misuse realized large increases in heroin deaths after the reformulation, and that the reformulation explains up to 80% of the rise in fatal heroin overdoses after 2010. Similarly Evans et al. (2018) find that each foregone prescription overdose death prevented by the OxyContin reformulation was offset by an additional death from heroin overdose. Further, because heroin is commonly injected and consumers often share needles, this drug-to-drug substitution led to increased transmission of hepatitis B and C (Beheshti, 2019a; Powell et al., 2019). Evans et al. (2020) find additional negative spillovers taking the form of increased child removals in areas with worse opioid outcomes, and Park and Powell (2020) find negative spillovers to labor force participation, finding increases in disability claiming.

The longer-term effects of these policies are not yet well understood. For instance, new initiation into medications for opioid use disorder might decline such that, in steady state, there would be fewer individuals who misuse opioids under the reformulation than there would have been in its absence. However, research by Powell and Pacula (2020) suggests the opposite outcome. In particular, the authors uncovered evidence of more deleterious effects in the long-run because the reformulation spurred development of illicit drug markets.

States and the federal government have also used the Controlled Substance Act (CSA) as a tool for addressing the opioid crisis.18 Two changes in CSA, targeting rival products and introduced separately, have allowed economists to study whether there are competitive spillovers to prescription products when one but not another product is regulated. In August 2014, the U.S. federal government added tramadol, the second most popular opioid medication at the time, to the CSA (entering this medication at Schedule V, which involves restrictions on refills). Twelve states implemented the identical policy prior to federal action, providing an opportunity to compare effectiveness of the same opioid policy at state versus federal levels. Seven weeks after tramadol’s scheduling, the leading opioid form on the market, hydrocodone combination products, was moved from Schedule III to the more restricted Schedule II (where no refills are allowed). Gupta et al. (2020) find that the tightening of these prescribing restrictions decreased their use, but also caused some increases in prescriptions of close competitors. As a result, there was no statistically detectable short-run reduction in total number of opioid prescriptions.

Enforcement of Illicit Drug Prohibitions

The crisis has transitioned towards illicit drugs such as heroin and fentanyl, thus a discussion of the literature on law enforcement crackdowns during past illicit drug crises is potentially useful. A 2014 review by Pollack and Reuter (2014) summarizes many studies of the effect of such enforcement efforts on drug prices; they do not find solid evidence that raising the risk of arrest or the increasing the length of drug sentences affects street prices. Cunningham and Finlay (2016), and Dobkin and Nicosia (2009) examine the effects of government efforts to make the precursors of methamphetamine less available. Both studies find only temporary effects on price, purity, and harmful outcomes, each outcome returning to its pre-intervention level within a few months to two years. Reuter (1986) similarly finds that cracking down on the importation and high level distribution of cocaine and marijuana in the 1980s had little effect on street-level drug prices.

In contrast, a study of Australia’s 2001 effort to reduce the importation of heroin uncovers a large impact on both purity and price (Moore and Schnepel, 2019). The authors also find a short-term increase in violent and financially-motivated crime, as well as longer-lasting decreases in overdoses and property crime. Australia may present a unique case for the enforcement of illegal heroin shipments due to its island geography, which allows greater success in targeting suppliers. Applying Australia’s approach to the U.S. may therefore not realize the same efficacy, as the land border with Mexico is more porous and there is evidence that fentanyl is often shipped to the U.S. by mail.19

Using web scrapes of darknet drug markets from 2014 to 2016, Miller (2020) documents that wholesale fentanyl is priced 90% below equivalent doses of heroin on the black market. Moreover, as fentanyl became more prevalent over this period, its price fell. In 2014, the international law enforcement initiative “Operation Onymous” shut down darknet drug markets and resulted in the arrests of darknet market administrators, sellers, and customers. However, this intervention caused only a small and temporary price increase in fentanyl, which was overwhelmed by the general downward trend during the same time period. Miller (2020) shows that while Chinese efforts to limit the illegal manufacture and export of various fentanyl analogues did flatten the downward time trend in fentanyl prices, the resulting prices remained strikingly low at the wholesale level. On the other hand, Mulligan (2020) argues that reduced law enforcement efforts after 2013, due to the “Holder memo,” played an important role in the emergence of illicit fentanyl.20

Demand-Side Policies

A key set of policies that are crucially important within the context of the opioid crisis relates to treatment of substance use disorders (SUDs), including OUD-associated care. Nuances of the U.S. SUD treatment system in terms of insurance, providers, and patients may pose barriers to a successful treatment-based policy response. Interested readers are referred to Saloner et al. (2018) and Maclean and Saloner (2018) for a review of the U.S. SUD treatment delivery system. This section summarizes key aspects salient to the opioid crisis.

Historically, much of SUD treatment has been largely isolated from the general health care delivery system. Scholars have raised hypotheses for the reasons that lie behind this segregation including: a stigmatized view of SUDs as moral failings rather than medical conditions, limited treatment options and understanding of addiction, concerns among insurers and public payers that the demand for SUD treatment is highly elastic and that covering such services will lead to prohibitively high health care costs, and informal treatment options that are designed to operate outside a structured health care system (e.g., Narcotics Anonymous). These reasons, and others, are likely not independent and instead interact to produce the U.S. SUD treatment delivery system in its current form.

Until the 2nd decade of the 21st century, there was little role for insurance (public or private) payments for SUD treatment. Insurance has tended not to cover SUD treatment or, if included, services have been subject to higher quantitative (e.g., cost-sharing) and non-quantitative (e.g., annual and lifetime use limitations) restrictions. Instead, providers relied on grants and contracts from federal, state, and local governments to finance treatment. For example, the SAMHSA Prevention and Treatment Block Grant Program has been a major source of funding over time. Patients have used their own funds to finance treatment, and many facilities offer charity care or payment assistance (e.g., the use of “sliding fee scales” is common). Fragmented payment sources (e.g., grants generally provide blocks of funding for specific objectives) have constrained provider finances and prevented investments in capacity building and technology adoption (Buck, 2011). State and federal policy changes in the early 21st century have expanded coverage, both in terms of the services and the number of people covered. These policies have generally led to increased service use and a greater reliance on insurance as a source of payment.

There are a range of treatment options available for patients, although the willingness of some SUD providers to accept insurance remains limited despite the previously noted polices, which likely impedes access to such services for some patients (Wen et al., 2019). Treatment often begins with detoxification, a process that often involves medication, through which the body rids itself of substances. Detoxification is generally not viewed as treatment per se, but rather as a precursor to care. Patients can receive care in office-based settings from psychiatrists, psychologists, social workers, and so forth. Such treatment may include the use of medications and/or psychotherapy. Treatment is also available through stand-alone SUD treatment facilities that offer outpatient services extending from less intensive care, such as hourly counseling services, to more intensive treatment that can encompass partial hospitalization where the patient receives care that lasts multiple hours per day. Residential treatment involves 30- to 90-day stays where the patient resides in the facility full-time. Finally, patients can receive care in general hospitals or those with specific SUD units, or psychiatric hospitals that fully specialize in treating patients with acute SUDs and other behavioral health conditions. In the 2nd and 3rd decades of the 21st century, there have been concerted efforts—supported by advocates, governments, patients, payers, and providers—to integrate SUD treatment within general care. Integrated care models can include placing SUD treatment within primary care and/or emphasizing treatment of broader health needs in specialized SUD treatment. See McClellan et al. (2020) for a discussion and an example.

Measuring and ensuring care quality have been challenges within the SUD treatment delivery system. For instance, to date there are no consensus quality metrics within the system although experts note that providing such information to patients is crucial to improve outcomes (Mark et al., 2018). In particular, scholars note that the SUD treatment delivery system lags behind other sectors in terms of the use of modern technology, many providers do not offer the continuum of care recommended by addiction medicine professional organizations, and there are frequent treatment shortages (Buck, 2011; Mark et al., 2018).

Policies, such as insurance expansions, that promote access to treatment are important, as they allow those with SUDs to reduce their use of opioids and associated harms rather than simply curtailing the supply of specific drugs that can be replaced with other substances through economic substitution. Several federal policies since 2014 offer potential to enhance treatment access to those with SUDs and improve overall treatment quality. The Comprehensive Addiction and Recovery Act (CARA), 21st Century Cures Act of 2016, and Substance Use-Disorder Prevention that Promotes Opioid Recovery and Treatment (SUPPORT) for Patients and Communities Act of 2018 are all designed, in varying degrees, to improve SUD treatment and patient outcomes. For example, the CARA Act authorizes over $181 million each year to address the opioid epidemic in particular, with the specific goal of increasing prevention programs and the availability of treatment. Provisions of the SUPPORT Act are designed to standardize the delivery of addiction medicine. Similarly, core provisions of the Affordable Care Act (ACA), which took effect in 2014, along with the Mental Health Parity and Addiction Equity Act (MHPAEA) of 2008 expanded coverage to millions of Americans, and reduced insurance coverage differentials between general medical and SUD treatment.

Policies that impact medications related to the treatment of chronic condition symptoms (for example, chronic pain) may offer an indirect pathway to reduce patient reliance on opioids and, in turn, mitigate associated harms. As a specific example, prescription opioids and marijuana may be substitutes along at least some dimensions, with marijuana sometimes serving as an alternative treatment for chronic pain. Several studies suggest that patients substitute marijuana for opioids following the adoption of a state medical marijuana law (MML). In particular, these studies use health insurance claims data and show that prescriptions for opioids decline post-law (Bradford & Bradford, 2017, 2018; Bradford et al., 2018; McMichael et al., 2020). Recreational marijuana laws (RMLs), the first of which was adopted in 2012, may have a similar impact on the utilization of prescription opoids (Wen & Hockenberry, 2018). Moreover, following an MML adoption, reported chronic pain declines among older adults (Nicholas & Maclean, 2019). Similarly, both health-related work absences (Ullman, 2017) and workers’ compensation benefit receipt (chronic pain is a common ailment among those receiving this benefit) (Ghimire & Maclean, 2020) decline post-MML, suggesting that the use of marijuana may sometimes be effective in reducing chronic pain and other work-impeding ailments. Jayawardhana and Fernandez (2021) show that opioid-related hospitalizations decline following an MML adoption, although the authors note that some policy features may mute the protective effect.

Dispensaries, venues in which consumers can legally purchase marijuana, appear to be particularly important in the relationship between marijuana and opioids. Powell et al. (2018) show that the opening of legal medical marijuana dispensaries reduces opioid-related admissions to SUD treatment facilities by 15% and opioid fatalities by 16%. Similarly, Smith (2020) indicates that deaths from prescription opioid overdose fall 11% after a medical marijuana dispensary opens. These effects are concentrated among non-Hispanic white males.

States adopted doctor-shopping laws (DSLs) requiring patients to report to their health care professional previous prescriptions and, in a broad manner, prohibiting them from obtaining medications through fraud, deceit, or misrepresentation. By limiting the ability of illegitimate patients to access opioids, the policies may reduce use of these drugs. For example, Popovici et al. (2018) show that DSLs reduce opioid overdose deaths and opioid-related admissions to SUD treatment. The authors find no evidence that patients substitute to heroin post-law, suggesting that DSLs do not simply induce patients to use more harmful substances that are procured through illicit markets.

Price Elasticity of Demand

Even though opioids are addictive substances, the demand for opioids and other drugs are responsive to economic factors such as time and financial prices (Becker et al., 1991). Therefore, when examining the evolution of the opioid crisis from prescription to primarily illicit drugs, the role of the price elasticity of demand for opioids is important to consider. Factors that reduce the price of opioids, including insurance expansions such as those due to Medicare Part D, could increase their use (Powell et al., 2018). Soni (2019) uses price variations in Medicare Part D to identify important heterogeneity in elasticity of demand for prescription opioids. Focusing on 55 to 74 year olds, she uncovers an overall elasticity of −0.89; however, almost all of the response is concentrated among new users. Soni also presents evidence that nonprescription pain killers are substitutes for prescribed opioids. Einav et al. (2018) use identifying variation from the design of the Medicare Part D “donut hole” in prescription drug coverage to show that the demand for prescription opioids is price inelastic, with an elasticity estimate of −0.04. However, the local average treatment effect is identified off of individuals at the spending margin of the donut hole; these patients are likely sicker than the typical Medicare Part D beneficiary or younger opioid user. Some evidence suggests that heroin may have a relatively high demand elasticity, when compared with other drugs. For example, Saffer and Chaloupka (1999) estimate price elasticities for alcohol, cocaine, and heroin of −0.30, −0.28, and −0.94, respectively, although it is unclear how applicable these findings are to the present day. However, a more recent study that combines experimental and longitudinal survey data finds that the conditional demand price elasticity for heroin is −0.80, suggesting that demand for illicit drugs continues to respond to price changes in a manner predicted by economic theory (Olmstead et al., 2015).

Harm Reduction Policies

Since there is little evidence that intensifying enforcement has significant potential for decreasing misuse or raising street prices of illicit or diverted drugs, much of the recent policy response emphasizes harm reduction strategies aimed at reducing fatal overdoses and other problems related to the misuse of opioids. Harm reduction policies include, but are not limited to, naloxone access laws (NALs), Good Samaritan Laws (GSLs), and syringe exchange programs (SEPs). NALs provide legal immunity to healthcare providers prescribing or administering naloxone, a medication used to reverse opioid overdoses.21 GSLs grant immunity or mitigated sentencing to individuals who call 911 in the case of an overdose, and SEPs simplify the act of obtaining new, clean syringes for injection drug users and may include the availability of supervised injection sites or other safety measures (e.g., test strips used to determine if heroin contains fentanyl).22

Evidence on the effectiveness of NALs is mixed. Doleac and Mukherjee (2018), emphasizing concerns about possible moral hazard, find that online searches for naloxone increase by 7% and those for opioid-treatment fall 1% post-NAL; opioid-related possession arrests, sales, and emergency department visits increase by 17%, 27%, and 15%, respectively, with no change in opioid-related mortality. Conversely, Rees et al. (2019) show that NAL adoption leads to a 9% to 10% reduction in opioid-related mortality and consistently negative, but less statistically significant, reductions associated with the passage of GSLs. Abouk et al. (2019) highlight the importance of the specific features of NALs, finding that those granting direct authority to pharmacists to distribute naloxone reduce fatal overdoses, whereas other types of NALs do not. An important challenge for all research on this topic is that the enactment of NALs occurred over a short time period that coincides with the explosion of fentanyl. This confluence of rapid policy adoption and changes in substances used implies that uncovering causal effects using standard quasi-experimental methods is difficult, and even more so if the exact timing of when these policies become effective (which may differ from formally legislated dates) is not well understood.

Packham (2019) investigates the effect of SEPs on HIV diagnoses, drug mortality, and opioid-related crime. Using a difference-in-differences framework that exploits variation in timing of SEP openings across counties, she finds that counties in which an SEP opens may realize falling HIV rates but also experience increases in opioid-related emergency department visists, opioid mortality, and arrests for drug possession. Adverse effects are particularly salient in rural areas and high poverty areas, which presumably have limited access to substance misuse treatment. This finding suggests that access to clean needles may have the intended effect of reducing needle sharing and blood-borne disease, while also decreasing the costs associated with intravenous drug use, posing a moral hazard issue.

Health Insurance

A relevant issue in a discussion of prescription and nonprescription opioids is insurance coverage for drug treatment. This coverage is often incorporated with broader changes in the health care delivery system, rather than specifically targeting the opioid crisis. While numerous treatment options are available—including medications such as methadone, buprenorphine, and naltrexone, alongside behavioral interventions like counseling (Murphy & Polsky, 2016)—most substance use disorders, including opioid use disorder, remain untreated. Recent estimates suggest that only 1 in 10 individuals with OUD receive medication for treating it in a given year (Substance Abuse and Mental Health Services Administration, 2019), although there have been recent expansions in availability of buprenorphine. In particular the DEA has expanded the number of health care providers who are allowed (“waivered”) to dispense this medication to patients (Dick et al., 2015).23 While there are many reasons why individuals do not receive treatment, including strong psychological barriers to treatment and stigma, commonly stated causes include inability to pay and lack of insurance coverage (Substance Abuse and Mental Health Services Administration, 2019). Thus, increasing the generosity of insurance, both in terms of the number of individuals eligible and the services included in plans, may facilitate treatment uptake and health improvements.

Research on the effects of health insurance on opioid use disorder frequently uses legal changes resulting from the Affordable Care Act as a source of identifying variation. One important modification under the ACA is that SUD treatment (including for OUD) became listed as an essential benefit that must be covered by most plans (McLellan & Woodworth, 2014). Medicaid, the primarly public health insurance system for low-income Americans, jointly funded by the states and federal government, is the largest insurance payer for SUD treatment (McLellan & Woodworth, 2014). Under the ACA, Medicaid coverage was expanded to include all adults with incomes up to 138% of the federal poverty line. However, in 2012 the Supreme Court ruled that the federal government could not compel states to expand Medicaid, and not all states elected to do so. Several studies have exploited this variation across states to test the impact of expanding Medicaid and show that this had important implications for opioid use disorder treatment access and outcomes.

For instance, Meinhofer and Witman (2018) find that ACA Medicaid expansion increased prescriptions for medications used to treat opioid use disorder by over 100%, raised admissions to drug treatment centers, and increased the probability that opioid use disorder treatment providers accepted Medicaid as a form of payment. Cher et al. (2019) corroborate this finding for OUD treatment medications. This latter finding is particularly important, as insurance has historically played a minor role in financing SUD treatment Mark et al. (2016). To date, there is limited evidence that this expansion has led to changes in opioid-related deaths (Abouk et al., 2019; Averett et al., 2019). However, the dependent coverage mandate in the ACA (which guarantees that children can remain on their parents’ health insurance plans up to age 26) is associated with reductions in opioid fatalities among young adults impacted by the policy (Wettstein, 2019).

A concern with any insurance expansion is its potential to induce moral hazard. In the context of the opioid use crisis, this would take the form of insurance reducing the out-of-pocket prices of prescription opioids, and potentially spurring misuse and opioid use disorder within the population. The research described in the previous paragraph suggests that this concern is largely unwarranted; however, a clear consensus has not yet emerged. For instance, Powell et al. (2020) leverage the plausibly exogenous variation in prescription drug coverage offered by the introduction of Medicare Part D in 2006 and find that a 10% increase in the supply of medical opioids leads to a 7% increase in opioid deaths among individuals likely ineligible for Medicare. This finding suggests that some of these prescribed opioids are diverted to other users.

Conclusions and Directions for Future Research

As the opioid crisis has emerged as a major public health concern, so has a large and rapidly growing body of economic research analyzing it. These studies use the concepts of supply and demand; habit-forming consumption; substitution, complementarity, and spillovers; and interactions with labor markets and family level outcomes. This research leverages a variety of data sources, with their respective strengths and weaknesses.24 Some analyses are purely descriptive; others attempt to determine causal effects. The research has increased understanding over a variety of important dimensions. For example, supply-side factors (e.g., aggressive pharmaceutical industry promotion of prescription opioids) were a primary cause of the epidemic, but with demand-side factors determining which groups have been most adversely affected. Regulatory and policy approaches have played a role in mitigating these initial harms but have been less successful in addressing the underlying addictions, resulting in spillover effects to the consumption of illicit opioids and the expansion of those markets. Generally, states have played a more active role than the federal government in these policy efforts, although often assisted by federal grant support (e.g., Comprehensive Addiction and Recovery Act of 2016) and advisory statements (e.g., CSA).

Going forward, while PDMPs have been extensively studied, there is room for better understanding the circumstances in which these programs are most beneficial and their long-run impacts. Other policy approaches (e.g., day limits to prescribing, CSA advisories, and most harm reduction policies) have received much less attention. In all opioid policy areas, researchers can benefit from the creation of taxonomies that reduce barriers to studying impacts of the policies (Grant et al., 2020) and from considering the power and appropriateness of the statistical methods chosen (Griffin et al., 2020). There also remains uncertainty about the fundamental causes of the opioid crisis and its relationship with other drug problems, such as the frequency of deaths involving poly-drug use, the rapidly emerging crisis of stimulant use and deaths, and the broader history of rising drug mortality in the United States (Jalal et al., 2018; Ruhm, 2017, 2019b). In addition, the rapid changes in circumstances (like the emergence of fentanyl) and policies (such as NAL implementation) provides severe challenges to the standard quasi-experimental methods economists typically use to study these issues. Considerable flexibility in analysis methods and a comprehensive understanding of research outside of economics will be needed to better determine the best approaches to provide credible estimates in future research.

At the begining of the 3rd decade of the 21st century the United States, and many other countries, are in the midst of the COVID-19 pandemic, which has led to over 400,000 American deaths. Although its implications for the opioid crisis are unclear, preliminary data indicate that COVID-19 is being accompanied by another increase in opioid-related mortality (Centers for Disease Control and Prevention, 2020b). Therefore, assessing the causal toll of the pandemic for opioid use and opioid use disorder will be important. In particular, a better understanding is needed of how opioid problems have been affected by other changes related to COVID-19 such as reduced willingness to seek health care; the growth in telehealth; and other nonmedical factors, such as isolation, strain, uncertainty, economic recession, a large-scale but short-lived government stimulus package, loss of friends and family members to COVID-19, and general disruptions in daily life. Finally, efforts will be needed to determine the most effective policies to address opioid outcomes in the post-COVID19 setting.

While contributions from many disciplines outside economics play enormous roles in answering the questions addressed here, the particular concerns of economists—in areas such as the roles of incentives, opportunity costs and spillovers, causal identification, and health as a production process—provide useful complementary insights.

Acknowledgments

The authors would like to acknowledge helpful comments from Abby Alpert, Janet Currie, Dhaval Dave, Ethan Lieber, David Powell, Molly Schnell, and Bradley Stein. Ruhm thanks the University of Virginia Bankard fund for financial support for this research.

References

  • Abouk, R., Helmchen, L. Moghtaderi, A., & Pines, J. (2019). The ACA Medicaid expansions and opioid mortality: Is there a link? Medical Care Research and Review, 20(3), 277−282.
  • Abouk, R., Pacula, R. L., & Powell, D. (2019, June). Association between state laws facilitating pharmacy distribution of naloxone and risk of fatal overdose. JAMA Internal Medicine, 179(6), 805–811.
  • Ahmad, F. R. L. S. (2020). Provisional drug overdose death counts (NVSS Vital Statistics Rapid Release Report). National Center for Health Statistics.
  • Alexander, G. C., Frattaroli, S., & Gielen, A. C. (2015). The prescription opioid epidemic: An evidence-based approach. Johns Hopkins Bloomberg School of Public Health.
  • Alexander, M. J., Kiang, M. V., & Barbieri, M. (2018). Trends in black and white opioid mortality in the United States, 1979–2015. Epidemiology, 29(5), 707.
  • Ali, M. M., Dowd, W. N., Classen, T., Mutter, R., & Novak, S. P. (2017). Prescription drug monitoring programs, nonmedical use of prescription drugs, and heroin use: Evidence from the national survey of drug use and health. Addictive Behaviors, 69, 65–77.
  • Aliprantis, D., Fee, K., & Schweitzer, M. (2019). Opioids and the labor market (SSRN Scholarly Paper ID 3179068). Social Science Research Network.
  • Alpert, A., Dykstra, S., & Jacobson, M. (2020). How do prescription drug monitoring programs reduce opioid prescribing? The role of hassle costs versus information (Working Paper 27584). National Bureau of Economic Research.
  • Alpert, A. E., Evans, W. N., Lieber, E. M., & Powell, D. (2019). Origins of the opioid crisis and its enduring impacts (Working Paper 26500). National Bureau of Economic Research.
  • Alpert, A., Powell, D., & Pacula, R. L. (2018). Supply-side drug policy in the presence of substitutes: Evidence from the introduction of abuse-deterrent opioids. American Economic Journal: Economic Policy, 10(4), 1–35.
  • Altekruse, S. F., Cosgrove, C. M., Altekruse, W. C., Jenkins, R. A., & Blanco, C. (2020). Socioeconomic risk factors for fatal opioid overdoses in the United States: Findings from the Mortality Disparities in American Communities study (MDAC). PloS ONE, 15(1), e0227966.
  • Averett, S. L., Smith, J. K., & Wang, Y. (2019). Medicaid expansion and opioid deaths. Health Economics, 28(12), 1491–1496.
  • Bachhuber, M. A., Saloner, B., LaRochelle, M., Merlin, J. S., Maughan, B. C., Polsky, D., Shaparin, N., & Murphy, S. M. (2018). Physician time burden associated with querying prescription drug monitoring programs. Pain Medicine, 19(10), 1952–1960.
  • Baldwin, M. L., & Marcus, S. C. (2014). The impact of mental and substance-use disorders on employment transitions. Health Economics, 23(3), 332–344.
  • Bao, Y., Pan, Y., Taylor, A., Radakrishnan, S., Luo, F., Pincus, H. A., & Schackman, B. R. (2016). Prescription drug monitoring programs are associated with sustained reductions in opioid prescribing by physicians. Health Affairs, 35(6), 1045–1051.
  • Bao, Y., Wen, K., Johnson, P., Jeng, P. J., Meisel, Z. F., & Schackman, B. R. (2018). Assessing the impact of state policies for prescription drug monitoring programs on high-risk opioid prescriptions. Health Affairs, 37(10), 1596–1604.
  • Barnett, M. L., Olenski, A. R., & Jena, A. B. (2017). Opioid-prescribing patterns of emergency physicians and risk of long-term use. New England Journal of Medicine, 376(7), 663–673.
  • Barry, C., & Frank, R. (2019). Evidence for addressing the opioid epidemic: Rounding out the picture. Brookings.edu.
  • Becker, G. S., Grossman, M., & Murphy, K. M. (1991). Rational addiction and the effect of price on consumption. The American Economic Review, 81(2), 237–241.
  • Beheshti, D. (2019a). Adverse health effects of abuse-deterrent opioids: Evidence from the reformulation of OxyContin. Health Economics, 28(12), 1449–1461.
  • Beheshti, D. (2019b). The impact of opioids on the labor market: Evidence from drug rescheduling [unpublished paper]. Economics Department, University of Texas at Austin.
  • Betz, M. R., & Jones, L. E. (2018). Wage and employment growth in America’s drug epidemic: Is all growth created equal? American Journal of Agricultural Economics, 100(5), 1357–1374.
  • Birnbaum, H. G., White, A. G., Schiller, M., Waldman, T., Cleveland, J. M., & Roland, C. L. (2011). Societal costs of prescription opioid abuse, dependence, and misuse in the United States. Pain Medicine, 12(4), 657–667.
  • Bondurant, S. R., Lindo, J. M., & Swensen, I. D. (2018, March). Substance abuse treatment centers and local crime. Journal of Urban Economics, 104, 124–133.
  • Bradford, A. C., & Bradford, W. D. (2017). Medical marijuana laws may be associated with a decline in the number of prescriptions for Medicaid enrollees. Health Affairs, 36(5), 945–951.
  • Bradford, A. C., & Bradford, W. D. (2018). The impact of medical cannabis legalization on prescription medication use and costs under Medicare Part D. The Journal of Law and Economics, 61(3), 461–487.
  • Bradford, A. C., Bradford, W. D., Abraham, A., & Bagwell Adams, G. (2018). Association between US state medical cannabis laws and opioid prescribing in the Medicare Part D population. JAMA Internal Medicine, 178(5), 667–672. |
  • Buchmueller, T. C., & Carey, C. (2018). The effect of prescription drug monitoring programs on opioid utilization in Medicare. American Economic Journal: Economic Policy, 10(1), 77–112.
  • Buck, J. A. (2011). The looming expansion and transformation of public substance abuse treatment under the Affordable Care Act. Health Affairs, 30(8), 1402–1410.
  • Buckles, K., Evans, W. N., & Lieber, E. M. (2020). The drug crisis and the living arrangements of children (Technical report). National Bureau of Economic Research.
  • Bullinger, L. R., & Wing, C. (2019). How many children live with adults with opioid use disorder? Children and Youth Services Review, 104, 104381.
  • Bütikofer, A., & Skira, M. M. (2018). Missing work is a pain: The effect of Cox-2 inhibitors on sickness absence and disability pension receipt. Journal of Human Resources, 53(1), 71–122.
  • Case, A., & Deaton, A. (2015). Rising morbidity and mortality in midlife among white non-Hispanic Americans in the 21st century. Proceedings of the National Academy of Sciences, 112(49), 15078–15083.
  • Case, A., & Deaton, A. (2020). Deaths of despair and the future of capitalism. Princeton University Press.
  • Cawley, J., & Ruhm, C. J. (2011). The economics of risky health behaviors. In M. V. Pauly, T. G. McGuire, & P. P. Barros (Eds.), Handbook of health economics (Vol. 2, pp. 95–199). Elsevier.
  • Centers for Disease Control and Prevention. (2020a). Multiple cause of death, 1999–2018 (Technical report). National Center for Health Statistics.
  • Centers for Disease Control and Prevention. (2020b). Vital statistics rapid release: Provisional drug overdose death counts.
  • Chang, H. Y., Lyapustina, T., Rutkow, L., Daubresse, M., Richey, M., Faul, M., Stuart, E. A., & Alexander, G. C. (2016). Impact of prescription drug monitoring programs and pill mill laws on high-risk opioid prescribers: A comparative interrupted time series analysis. Drug and Alcohol Dependence, 165, 1–8.
  • Charles, K. K., Hurst, E., & Schwartz, M. (2019). The transformation of manufacturing and the decline in US employment. NBER Macroeconomics Annual, 33, 307–372.
  • Cher, B. A., Morden, N. E., & Meara, E. (2019). Medicaid expansion and prescription trends: Opioids, addiction therapies, and other drugs. Medical Care, 57(3), 208.
  • Cicero, T. J., Inciardi, J. A., & Muñoz, A. (2005). Trends in abuse of OxyContin and other opioid analgesics in the United States: 2002–2004. The Journal of Pain, 6(10), 662–672.
  • Cunningham, S., & Finlay, K. (2016). Identifying demand responses to illegal drug supply interdictions. Health Economics, 25(10), 1268–1290.
  • Currie, J., Jin, J. Y., & Schnell, M. (2018). U.S. employment and opioids: Is there a connection? (Working Paper 24440). National Bureau of Economic Research.
  • Currie, J., & Schwandt, H. (2020). The opioid epidemic was not caused by economic distress but by factors that could be more rapidly addressed (Technical report). National Bureau of Economic Research.
  • Dave, D., Deza, M., & Horn, B. P. (2018). Prescription drug monitoring programs, opioid abuse, and crime. Southern Economic Journal, 87(3), 808–848.
  • Davenport, S., Weaver, A., & Caverly, M. (2019). Economic impact of non-medical opioid use in the United States.
  • Deaton, A. (2020, June 23). Health and wealth inequality in America: How COVID-19 makes clear the need for change. Congressional Testimony to House Budget Committee, Virtual Hearing.
  • Deiana, C., & Giua, L. (2018). The US opidemic: Prescription opioids, labour market conditions and crime (MPRA Paper). University Library of Munich, Germany.
  • Dick, A. W., Pacula, R. L., Gordon, A. J., Sorbero, M., Burns, R. M., Leslie, D., & Stein, B. D. (2015). Growth in buprenorphine waivers for physicians increased potential access to opioid agonist treatment, 2002–11. Health Affairs, 34(6), 1028–1034.
  • Dobkin, C., & Nicosia, N. (2009). The war on drugs: Methamphetamine, public health, and crime. American Economic Review, 99(1), 324–349.
  • Doleac, J. L., & Mukherjee, A. (2018). The moral hazard of lifesaving innovations: Naloxone access, opioid abuse, and crime (SSRN Scholarly Paper ID 3135264). Social Science Research Network.
  • Dow, W. H., Godøy, A., Lowenstein, C., & Reich, M. (2020). Can labor market policies reduce deaths of despair? Journal of Health Economics, 74, 102372.
  • Einav, L., Finkelstein, A., & Polyakova, M. (2018). Private provision of social insurance: Drug-specific price elasticities and cost sharing in Medicare Part D. American Economic Journal: Economic Policy, 10(3), 122–53.
  • Evans, M., Harris, M., & Kessler, L. (2020, April). The hazards of unwinding the prescription opioid epidemic: Implications for child abuse and neglect (SSRN Scholarly Paper ID 3582060). Social Science Research Network.
  • Evans, W. N., Lieber, E. M. J., & Power, P. (2018). How the reformulation of OxyContin ignited the heroin epidemic. The Review of Economics and Statistics, 101(1), 1–15.
  • Finkelstein, A., Gentzkow, M., & Williams, H. L. (2019). Place-based drivers of mortality: Evidence from migration (Working Paper 25975). National Bureau of Economic Research.
  • Florence, C., Luo, F., & Rice, K. (2020). The economic burden of opioid use disorder and fatal opioid overdose in the United States, 2017. Drug and Alcohol Dependence, 218, 108350.
  • Florence, C., Luo, F., Xu, L., & Zhou, C. (2016). The economic burden of prescription opioid overdose, abuse and dependence in the United States, 2013. Medical Care, 54(10), 901.
  • Frankt, A., & Monkovic, T. (2019, November 25). A “rare case where racial biases” protected African-Americans. New York Times.
  • Fryer, R. G., Heaton, P. S., Levitt, S. D., & Murphy, K. M. (2005). Measuring the impact of crack cocaine (Working Paper 11318). National Bureau of Economic Research.
  • Garthwaite, C. L. (2012). The economic benefits of pharmaceutical innovations: The case of Cox-2 inhibitors. American Economic Journal: Applied Economics, 4(3), 116–37.
  • Ghimire, K. M., & Maclean, J. C. (2020). Medical marijuana and workers’ compensation claiming. Health Economics, 29(4), 419–434.
  • Gihleb, R., Giuntella, O., & Zhang, N. (2019a). The effect of mandatory access prescription drug monitoring programs on foster care admissions. Journal of Human Resources, 0918–9729R2. Advance online publication.
  • Gihleb, R., Giuntella, O., & Zhang, N. (2020). Prescription drug monitoring programs and neonatal outcomes. Regional Science and Urban Economics, 81, 103497.
  • Grant, S., Smart, R., & Stein, B. D. (2020). We need a taxonomy of state-level opioid policies. JAMA Health Forum, 1, e200050..
  • Grecu, A. M., Dave, D. M., & Saffer, H. (2019). Mandatory access prescription drug monitoring programs and prescription drug abuse. Journal of Policy Analysis and Management, 38(1), 181–209.
  • Griffin, B. A., Schuler, M. S., Stuart, E. A., Patrick, S., McNeer, E., Smart, R., Powell, D., Stein, B., Schell, T., & Pacula, R. L. (2020). Variation in performance of commonly used statistical methods for estimating effectiveness of state-level opioid policies on opioid-related mortality. (Technical Report). National Bureau of Economic Research.
  • Griswold, M. K., Chapman, B. P., Krotulski, A. J., Friscia, M., Boyer, E. W., Logan, B. K., & Babu, K. M. (2018). Cryptomarket drug acquisition leading to furanyl fentanyl overdose. Forensic Toxicology, 36(2), 534–536.
  • Gupta, S., Nguyen, T. D., Freeman, P. R., & Simon, K. I. (2020). Competitive effects of federal and state opioid restrictions: Evidence from the controlled substance laws (Working Paper 27520). National Bureau of Economic Research.
  • Haffajee, R. L., Jena, A. B., & Weiner, S. G. (2015). Mandatory use of prescription drug monitoring programs. JAMA, 313(9), 891–892.
  • Harris, M. C., Kessler, L. M., Murray, M. N., & Glenn, B. (2020). Prescription opioids and labor market pains: The effect of Schedule II opioids on labor force participation and unemployment. Journal of Human Resources, 55(4), 1319–1364.
  • Hedden, S. L., Kennet, J., Lipari, R., Medley, G., Tice, P., Copello, E. A. P., & Kroutil, L. A. (2020, March). Behavioral health trends in the United States: Results from the 2014 National Survey on Drug Use and Health (HHS Publication No. SMA 15–4927, NSDUH Series H-50). Center for Behavioral Health Statistics and Quality, Substance Abuse and Mental Health Services Administration.
  • Hedegaard, H., Warner, M., & Miniño, A. M. (2017). Drug overdose deaths in the United States, 1999–2015. NCHS data brief, no 273. National Center for Health Statistics.
  • Hill, C. S., Jr. (1993). The barriers to adequate pain management with opioid analgesics. Seminars in Oncology, 20(1): 1–5.
  • Hollingsworth, A., Ruhm, C. J., & Simon, K. (2017). Macroeconomic conditions and opioid abuse. Journal of Health Economics, 56, 222–233.
  • Holmgren, A. J., Botelho, A., & Brandt, A. M. (2020). A history of prescription drug monitoring programs in the United States: Political appeal and public health efficacy. American Journal of Public Health, 110(8), 1191–1197.
  • Hummels, D., Munch, J., & Xiang, C. (2016). No pain, no gain: The effects of exports on effort, injury, and illness (Working Paper 22365). National Bureau of Economic Research.
  • Ihongbe, T. O., & Masho, S. W. (2016). Prevalence, correlates and patterns of heroin use among young adults in the United States. Addictive Behaviors, 63, 74–81.
  • Jalal, H., Buchanich, J. M., Roberts, M. S., Balmert, L. C., Zhang, K., & Burke, D. S. (2018). Changing dynamics of the drug overdose epidemic in the United States from 1979 through 2016. Science, 361(6408).
  • Jayawardhana, J., & Fernandez, J. M. (2021). The associations of medical marijuana policies with opioid‐related health care utilization. Health Services Research, 56(2), 299–309.
  • Kaestner, R., & Ziedan, E. (2019). Mortality and socioeconomic consequences of prescription opioids: Evidence from state policies (Working Paper 26135). National Bureau of Economic Research.
  • Kilby, A. (2016). Opioids for the masses: Welfare tradeoffs in the regulation of narcotic pain medications (Working Paper). MIT Economics Department.
  • Krueger, A. B. (2017). Where have all the workers gone? An inquiry into the decline of the U.S. labor force participation rate. Brookings Papers on Economic Activity, 2017(2), 1–87.
  • Laird, J., & Nielsen, T. (2017). Physician prescribing behaviors on prescription drug use and labor supply: Evidence from movers in Denmark (Harvard University Job Market Paper).
  • Lamy, F. R., Daniulaityte, R., Barratt, M. J., Lokala, U., Sheth, A., & Carlson, R. G. (2020). Listed for sale: Analyzing data on fentanyl, fentanyl analogs and other novel synthetic opioids on one cryptomarket. Drug and Alcohol Dependence, 213, 108115. Advance online publication.
  • Leonhardt, D., & Thompson, S. A. (2020, March 6). Is the rise in illicit opioids affecting labor supply and disability claiming rates? New York Times.
  • Lin, D., Liu, S., & Ruhm, C. J. (2020). Opioid deaths and local healthcare intensity: A longitudinal analysis of the US population, 2003–2014. American Journal of Preventive Medicine, 58(1), 50–58.
  • Lin, D. H., Lucas, E., Murimi, I. B., Jackson, K., Baier, M., Frattaroli, S., Gielen, A. C., Moyo, P., Simoni-Wastila, L., & Alexander, G. C. (2017). Physician attitudes and experiences with Maryland’s Prescription Drug Monitoring Program (PDMP). Addiction, 112(2), 311–319.
  • Lippold, K. M. (2019). Racial/ethnic and age group differences in opioid and synthetic opioid–involved overdose deaths among adults aged ≥18 years in metropolitan areas—United States, 2015–2017. MMWR. Morbidity and Mortality Weekly Report, 68.
  • Lyapustina, T., Rutkow, L., Chang, H. Y., Daubresse, M., Ramji, A. F., Faul, M., Stuart, E. A., & Alexander, G. C. (2016). Effect of a “pill mill” law on opioid prescribing and utilization: The case of Texas. Drug and Alcohol Dependence, 159, 190–197.
  • Maclean, J. C., Horn, B. P., & Cantor, J. H. (2020). Business cycles and admissions to substance abuse treatment. Contemporary Economic Policy, 38(1), 139–154.
  • Maclean, J. C., & Saloner, B. (2018). Substance use treatment provider behavior and healthcare reform: Evidence from Massachusetts. Health Economics, 27(1), 76–101.
  • Maclean, J. C., Webber, D., & French, M. T. (2015). Workplace problems, mental health and substance use. Applied Economics, 47(9), 883–905.
  • Macy, B. (2018). Dopesick. Little, Brown.
  • Mallatt, J. (2018). The effect of prescription drug monitoring programs on opioid prescriptions and heroin crime rates (SSRN 3050692). Social Science Research Network.
  • Mallatt, J. (2019). Individual effects of prescription drug monitoring programs: How prescriber information targets misusers while preserving access for treatment of legitimate pain (SSRN Scholarly Paper ID 3411199). Social Science Research Network.
  • Mallatt, J. (2020a). The effect of opioid supply-side interventions on opioid-related business establishments (SSRN Scholarly Paper ID 3697626). Social Science Research Network.
  • Mallatt, J. (2020b). Policy-induced substitution to illicit drugs and implications for law enforcement activity (SSRN Scholarly Paper ID 3418615). Social Science Research Network.
  • Mark, T., O’Brien, J., Mendell, G., McLellan, A., & Arsenault, S. (2018, January 5). Improving addiction treatment with consumer report cards. Health Affairs Blog.
  • Mark, T. L., Yee, T., Levit, Y. R., Camacho-Cook, J., Cutler, E., & Carroll, C. D. (2016). Insurance financing increased for mental health conditions but not for substance use disorders, 1986–2014. Health Affairs, 35(6), 958–965.
  • Martins, S. S., Santaella-Tenorio, J., Marshall, B. D., Maldonado, A., & Cerdá, M. (2015). Racial/ethnic differences in trends in heroin use and heroin-related risk behaviors among nonmedical prescription opioid users. Drug and Alcohol Dependence, 151, 278–283.
  • Mauri, A. I., Townsend, T. N., & Haffajee, R. L. (2020). The association of state opioid misuse prevention policies with patient-and provider-related outcomes: A scoping review. The Milbank Quarterly, 98(1), 57–105.
  • McCance-Katz, E. F. (2018). Presentation: The national survey on drug use and health 2018. Substance Abuse and Mental Health Services Administration (SAMHSA) Presentation, Center of Behavioral Health Statistics and Quality.
  • McCance-Katz, E. F. (2020). Death certificate files: Wide-ranging online data for epidemiologic research (WONDER). CDC, National Center for Health Statistics.
  • McClellan, C., Maclean, J. C., Saloner, B., McGinty, E. E., & Pesko, M. F. (2020). Integrated care models and behavioral health care utilization: Quasi-experimental evidence from Medicaid health homes. Health Economics, 29(9), 1086−1097.
  • McGreal, C. (2018, November 19). The making of an opioid epidemic.
  • McLellan, A. T., & Woodworth, A. M. (2014). The Affordable Care Act and treatment for “substance use disorders:” Implications of ending segregated behavioral healthcare. Journal of Substance Abuse Treatment, 46(5), 541–545.
  • McMichael, B. J., Van Horn, R. L., & Viscusi, W. K. (2020). The impact of cannabis access laws on opioid prescribing. Journal of Health Economics, 69, 102273.
  • Meara, E., Horwitz, J. R., Powell, W., McClelland, L., Zhou, W., O’Malley, A. J., & Morden, N. E. (2016). State legal restrictions and prescription-opioid use among disabled adults. New England Journal of Medicine, 375(1), 44–53.
  • Meinhofer, A. (2016). The war on drugs: Estimating the effect of prescription drug supply-side interventions (SSRN Scholarly Paper ID 2716974). Social Science Research Network.
  • Meinhofer, A., & Witman, A. E. (2018, July). The role of health insurance on treatment for opioid use disorders: Evidence from the Affordable Care Act Medicaid expansion. Journal of Health Economics, 60, 177–197.
  • Metcalf, G. E., & Wang, Q. (2019). Abandoned by coal, swallowed by opioids? (Working Paper 26551). National Bureau of Economic Research.
  • Miller, J. N. (2020). The war on drugs 2.0: Darknet fentanyl’s rise and the effects of regulatory and law enforcement action. Contemporary Economic Policy, 38(2), 246–257.
  • Mulligan, C. B. (2020). Prices and federal policies in opioid markets (Working Paper 26812). National Bureau of Economic Research.
  • Murphy, S. M., & Polsky, D. (2016). Economic evaluations of opioid use disorder interventions. Pharmacoeconomics, 34(9), 863–887.
  • Musse, I. (2020). Employment shocks and demand for pain medication (SSRN Scholarly Paper ID 3646543). Social Science Research Network.
  • Nguyen, T. D., Bradford, W. D., & Simon, K. I. (2019). How do opioid prescribing restrictions affect pharmaceutical promotion? Lessons from the mandatory access prescription drug monitoring programs (Working Paper 26356). National Bureau of Economic Research.
  • Nicholas, L. H., & Maclean, J. C. (2019). The effect of medical marijuana laws on the health and labor supply of older adults: Evidence from the health and retirement study. Journal of Policy Analysis and Management, 38(2), 455–480.
  • Olmstead, T. A., Alessi, S. M., Kline, B., Pacula, R. L., & Petry, N. M. (2015). The price elasticity of demand for heroin: Matched longitudinal and experimental evidence. Journal of Health Economics, 41, 59–71.
  • Packham, A. (2019). Are syringe exchange programs helpful or harmful? New evidence in the wake of the opioid epidemic (Working Paper 26111). National Bureau of Economic Research.
  • Pacula, R. L., & Powell, D. (2018). A supply-side perspective on the opioid crisis. Journal of Policy Analysis and Management, 37(2), 438–446.
  • Paice, J. A., Toy, C., & Shott, S. (1998). Barriers to cancer pain relief: Fear of tolerance and addiction. Journal of Pain and Symptom Management, 16(1), 1–9.
  • Pardo, B., Taylor, J., Caulkins, J. P., Kilmer, B., Reuter, P., & Stein, B. D. (2019). Future of fentanyl. RAND Corporation.
  • Park, S., & Powell, D. (2020). Is the rise in illicit opioids affecting labor supply and disability claiming rates? (NBER Working Paper w27804). National Bureau of Economic Research.
  • Paulozzi, L. J., Kilbourne, E. M., & Desai, H. A. (2011, May). Prescription drug monitoring programs and death rates from drug overdose. Pain Medicine, 12(5), 747–754.
  • Peiper, N. C., Clarke, S. D., Vincent, L. B., Ciccarone, D., Kral, A. H., & Zibbell, J. E. (2019). Fentanyl test strips as an opioid overdose prevention strategy: Findings from a syringe services program in the southeastern United States. International Journal of Drug Policy, 63, 122–128.
  • Pierce, J. R., & Schott, P. K. (2020). Trade liberalization and mortality: Evidence from US counties. American Economic Review: Insights, 2(1), 47–64.
  • Plüddemann, A., Flisher, A. J., McKetin, R., Parry, C., & Lombard, C. (2010). Methamphetamine use, aggressive behavior and other mental health issues among high-school students in Cape Town, South Africa. Drug and Alcohol Dependence, 109(1–3), 14–19.
  • Pollack, H. A., & Reuter, P. (2014). Does tougher enforcement make drugs more expensive? Addiction, 109(12), 1959–1966.
  • Popovici, I., Maclean, J. C., Hijazi, B., & Radakrishnan, S. (2018). The effect of state laws designed to prevent nonmedical prescription opioid use on overdose deaths and treatment. Health Economics, 27(2), 294–305.
  • Pouget, E. R., Fong, C., & Rosenblum, A. (2018). Racial/ethnic differences in prevalence trends for heroin use and non-medical use of prescription opioids among entrants to opioid treatment programs, 2005–2016. Substance Use & Misuse, 53(2), 290–300.
  • Powell, D., Alpert, A., & Pacula, R. L. (2019). A transitioning epidemic: How the opioid crisis is driving the rise in Hepatitis C. Health Affairs, 38(2), 287–294.
  • Powell, D., & Pacula, R. L. (2020). The evolving consequences of OxyContin reformulation on drug overdoses (Working Paper 26988). National Bureau of Economic Research.
  • Powell, D., Pacula, R. L., & Jacobson, M. (2018, March). Do medical marijuana laws reduce addictions and deaths related to pain killers? Journal of Health Economics, 58, 29–42.
  • Powell, D., Pacula, R. L., & Taylor, E. (2020). How increasing medical access to opioids contributes to the opioid epidemic: Evidence from Medicare Part D. Journal of Health Economics, 71, 102286.
  • Quinones, S. (2016). Dreamland. Bloomsbury Press.
  • Rees, D. I., Sabia, J. J., Argys, L. M., Dave, D., & Latshaw, J. (2019). With a little help from my friends: The effects of Good Samaritan and naloxone access laws on opioid-related deaths. The Journal of Law and Economics, 62(1), 1–27.
  • Reuter, P. (1986, January). Risks and prices: An economic analysis of drug enforcement. Crime and Justice, 7, 289–340.
  • Rice, J. B., Kirson, N. Y., Shei, A., Cummings, A. K. G., Bodnar, K., Birnbaum, H. G., & Ben-Joseph, R. (2014). Estimating the costs of opioid abuse and dependence from an employer perspective: A retrospective analysis using administrative claims data. Applied Health Economics and Health Policy, 12(4), 435–446.
  • Rietveld, C. A., & Patel, P. C. (2020). Prescription opioids and new business establishments. Small Business Economics, 1–25.
  • Rudd, R. A., Paulozzi, L. J., Bauer, M. J., Burleson, R. W., Carlson, R. E., Dao, D., Davis, J. W., Dudek, J., Eichler, B. A., Fernandes, J. C., Fondario, A., Gabella, B., Hume, B., Huntamer, T., Kariisa, M., Largo, T. W., Miles, J., Newmyer, A., Nitcheva, D., Perez, B. E., … Zehner, A. M. (2014). Increases in heroin overdose deaths—28 states, 2010 to 2012. MMWR. Morbidity and Mortality Weekly Report, 63(39), 849.
  • Ruhm, C. J. (2000). Are recessions good for your health? The Quarterly Journal of Economics, 115(2), 617–650.
  • Ruhm, C. J. (2017). Drug involvement in fatal overdoses. SSM-Population Health, 3, 219–226.
  • Ruhm, C. J. (2018). Corrected US opioid-involved drug poisoning deaths and mortality rates, 1999–2015. Addiction, 113(7), 1339–1344.
  • Ruhm, C. J. (2019a). Drivers of the fatal drug epidemic. Journal of Health Economics, 64, 25–42.
  • Ruhm, C. J. (2019b). Nonopioid overdose death rates rose almost as fast as those involving opioids, 1999–2016. Health Affairs, 38(7), 1216–1224.
  • Rutkow, L., Chang, H. Y., Daubresse, M., Webster, D. W., Stuart, E. A., & Alexander, G. C. (2015). Effect of Florida’s prescription drug monitoring program and pill mill laws on opioid prescribing and use. JAMA Internal Medicine, 175(10), 1642–1649.
  • Sacarny, A., Yokum, D., Finkelstein, A., & Agrawal, S. (2016). Medicare letters to curb overprescribing of controlled substances had no detectable effect on providers. Health Affairs, 35(3), 471–479.
  • Sacks, D. W., Hollingsworth, A., Nguyen, T. D., & Simon, K. I. (2019). Can policy affect initiation of addictive substance use? Evidence from opioid prescribing (Working Paper 25974). National Bureau of Economic Research.
  • Saffer, H., & Chaloupka, F. (1999). The demand for illicit drugs. Economic Inquiry, 37(3), 401–411.
  • Saloner, B., Akosa Antwi, Y., Maclean, J. C., & Cook, B. (2018). Access to health insurance and utilization of substance use disorder treatment: Evidence from the Affordable Care Act dependent coverage provision. Health Economics, 27(1), 50–75.
  • Saloner, B., & Barry, C. L. (2018). Ending the opioid epidemic requires a historic investment in medication-assisted treatment. Journal of Policy Analysis and Management, 37(2), 431–438.
  • Savych, B., Neumark, D., & Lea, R. (2018). Do opioids help injured workers recover and get back to work? The impact of opioid prescriptions on duration of temporary disability (Working Paper 24528). National Bureau of Economic Research.
  • Schnell, M. (2018). Physician behavior in the presence of a secondary market: The case of prescription opioids (Working Paper). Economics Department, Princeton University.
  • Schnell, M., & Currie, J. (2018). Addressing the opioid epidemic: Is there a role for physician education? American Journal of Health Economics, 4(3), 383–410.
  • Scholl, L., Seth, P., Kariisa, M., Wilson, N., & Baldwin, G. (2019). Drug and opioid-involved overdose deaths—United States, 2013–2017. Morbidity and Mortality Weekly Report, 67(51–52), 1419.
  • Schuchat, A., Houry, D., & Guy, G. P. (2017). New data on opioid use and prescribing in the United States. JAMA, 318(5), 425–426.
  • Schuler, M. S., Heins, S. E., Smart, R., Griffin, B. A., Powell, D., Stuart, E. A., Pardo, B., Smucker, S., Patrick, S. W., Pacula, R. L., & Stein, B. (2020). The state of the science in opioid policy research. Drug and Alcohol Dependence, 214, 108137.
  • Scrivner, O., McAvoy, E., Nguyen, T., Choedena, T., Simon, K., & Borner, K. (2020). Interactive network visualization of opioid crisis related data- policy, pharmaceutical, training, and more.
  • Smart, R., Kase, C. A. Taylor, E. A., Lumsden, S., Smith, S. R., & Stein, B. D. (2020). Strengths and weaknesses of existing data sources to support research to address the opioids crisis. Preventive Medicine Reports, 17, 101015.
  • Smart, R., Pardo, B., & Davis, C. S. (2020). Systematic review of the emerging literature on the effectiveness of naloxone access laws in the United States. Addiction, 116(1), 1−216.
  • Smith, R. A. (2020). The effects of medical marijuana dispensaries on adverse opioid outcomes. Economic Inquiry, 58(2), 569–588.
  • Soni, A. (2019, July). Health insurance, price changes, and the demand for pain relief drugs: Evidence from Medicare Part D (SSRN Scholarly Paper ID 3268968). Social Science Research Network.
  • Spencer, M., Warner, M., Bastian, B. A., Trinidad, J. P., & Hedegaard, H. (2019). Drug overdose deaths involving fentanyl, 2011–2016. National Vital Statistics Reports, 68(3).
  • Substance Abuse and Mental Health Services Administration. (2019). 2018 National Survey of Drug Use and Health (NSDUH) detailed tables.
  • Substance Abuse and Mental Health Services Administration. (2020). The opioid crisis and the Black/African American population: An urgent issue (Working Paper).
  • Swensen, I. D. (2015). Substance-abuse treatment and mortality. Journal of Public Economics 122, 13–30.
  • Ullman, D. F. (2017). The effect of medical marijuana on sickness absence. Health Economics, 26(10), 1322–1327.
  • United States Department of Justice. (2018a). 2016 heroin domestic monitor program. DEA Intelligence Report.
  • United States Department of Justice. (2018b). 2018 national drug threat assessment. DEA Intelligence Report.
  • Venkataramani, A. S., Bair, E. F., O’Brien, R. L., & Tsai, A. C. (2020). Association between automotive assembly plant closures and opioid overdose mortality in the United States: A difference-in-differences analysis. JAMA Internal Medicine, 180(2), 254–262.
  • Venkataramani, A. S., & O’Brien, R. (2020). Economic opportunity, drug overdose mortality, and disability (Working Paper WI20-08). Retirement and Disability Research Center University of Wisconsin.
  • Wang, L. X. (2020, April). The complementarity of health information and health IT for reducing opioid-related mortality and morbidity (SSRN Scholarly Paper ID 3176809). Social Science Research Network.
  • Weissman, D. E. (1993). Doctors, opioids, and the law: The effect of controlled substances regulations on cancer pain management. Seminars in Oncology, 20, 53−58.
  • Wen, H., & Hockenberry, J. M. (2018). Association of medical and adult-use marijuana laws with opioid prescribing for Medicaid enrollees. JAMA Internal Medicine, 178(5), 673–679.
  • Wen, H., Hockenberry, J. M., & Cummings, J. R. (2017). The effect of Medicaid expansion on crime reduction: Evidence from HIFA-waiver expansions. Journal of Public Economics, 154, 67–94.
  • Wen, H., Hockenberry, J. M., Jeng, P. J., & Bao, Y. (2019). Prescription drug monitoring program mandates: Impact on opioid prescribing and related hospital use. Health Affairs, 38(9), 1550–1556.
  • Wen, H., Wilk, A. S., Druss, B. G., & Cummings, J. R. (2019). Medicaid acceptance by psychiatrists before and after Medicaid expansion. JAMA Psychiatry, 76(9), 981–983.
  • Wettstein, G. (2019). Health insurance and opioid deaths: Evidence from the Affordable Care Act young adult provision. Health Economics, 28(5), 666–677.
  • White, A. G., Birnbaum, H. G., Rothman, D. B., & Katz, N. (2009). Development of a budget-impact model to quantify potential cost savings from prescription opioids designed to deter abuse or ease of extraction. Applied Health Economics and Health Policy, 7(1), 61–70.
  • White House Council of Economic Advisers. (2017). The underestimated cost of the opioid crisis.
  • Young, H. W., Tyndall, J. A., & Cottler, L. B. (2017). The current utilization and perceptions of prescription drug monitoring programs among emergency medicine providers in Florida. International Journal of Emergency Medicine, 10(1), 16.
  • Ziedan, E., & Kaestner, R. (2020). Effect of prescription opioids and prescription opioid control policies on infant health (Working Paper 26749). National Bureau of Economic Research.

Notes

  • 1. CDC

  • 2. See Barry and Frank (2019) for an eloquent discussion of the importance of incorporating evidence from fields beyond economics.

  • 3. See HHS Expands Access to Treatment for Opiod Use Disorder; however, as of this writing, it is unclear whether HHS has authority to issue this decision without congressional action.

  • 4. All figures on drug deaths cited in this article are based on information contained on death certificates. These certificates understate the actual death numbers by 20% to 25% because of incomplete reporting of drug involvement (Ruhm, 2018).

  • 5. Patients with substance use disorder were also increasingly more likely to seek medical care for their condition over that time period, as substance problems have been increasingly more likely to be seen as a condition to be treated medically in recent decades. This change in how SUDs are viewed likely also contributed to the increase in diagnoses, independent from increases in underlying substance misuse.

  • 6. There was a spike in fentanyl deaths between 2005 and 2007 in Chicago, Detroit, and Philadelphia related to a single lab in Mexico that was shut down in 2006 (Pardo et al., 2019).

  • 7. For further studies on age and race dimensions of the crisis, please see Rudd et al. (2014), Martins et al. (2015), Ihongbe and Masho (2016), Hedegaard et al. (2017), Pouge et al. (2018), and Scholl et al. (2019).

  • 8. Relative risk factors are taken from Altekruse et al. (2020).

  • 9. While both medications are used to treat chronic pain, Cox-2 inhibitors do not have the same addictive properties as opioids.

  • 10. Insuring individuals with opioid use disorder is estimated to cost employers an extra $10,000 per a year in medical expenses and $1,200 annually in work losses (Rice et al., 2014).

  • 11. Swensen (2015) shows that fatal drug overdoses also decrease as treatment access increases.

  • 12. See Case and Deaton (2020) for a detailed discussion of these issues and for their arguments for why they believe that declines in social capital are the key factor explaining increases in “deaths of despair.”

  • 13. Some scholars may categorize these as supply-side policies because they relate to the health care system. However, for the purposes of this review, these policies are treated as predominantly demand-side as they criminalize patient behaviors. For example, patients are prohibited from seeking prescriptions from multiple providers without disclosing this information.

  • 14. The Drug Enforcement Administration assigns substances a Schedule I through Schedule V classification status based on the medical usefulness of the drug balanced with the potential to misuse the drug. Schedule I drugs include heroin and are strictly illegal with no medical benefits, lower schedules include drugs that, in descending order, have relatively higher harm risk and lower medical benefits. Schedule II through Schedule V drugs include opioids like oxycodone, prescription fentanyl, and hydrocodone, among many others, and require a prescription.

  • 15. Missouri was the last state to implement a PDMP, and the law has been contested since its effective date, so some studies consider Missouri to not have an operating PDMP.

  • 16. Note, however that some states are able to provide risk scores of all patients in PDMPs; for example, see NarxCare.

  • 17. The PDMP may also have a direct effect by reducing the prescribing of other scheduled drugs, such as sedatives or stimulants.

  • 18. This represents one of the few policies that have been used at a federal level.

  • 19. See Griswold et al. (2018), and Peiper et al. (2019) for examples of fentanyl precursors sourced abroad and mailed to the United States.

  • 20. In the Holder memo, then Attorney General Eric Holder directed federal lawyers to stop prosecuting nonviolent drug crimes. However, the reformulation of OxyContin and other regulatory measures, noted previously, over this time period may preclude clean identification of Holder memo effects.

  • 21. See Smart, Pardo, and Davis (2020) for a recent review of NALs.

  • 22. Generally, SEPs allow individuals to return used syringes for safe disposal (which also minimizes exposure to discarded “dirty” syringes for non-users) and receive clean syringes without risk of legal consequences or financial costs.

  • 23. The Drug Addiction Treatment Act of 2000 permits health care providers (initially physicians only, but later additional providers such as nurse practitioners were included) who meet specific qualifications to treat OUD with Schedule III, IV, and V narcotic medications that have been approved by the Food and Drug Administration for treating this condition.

  • 24. See Smart, Kase, et al. (2020) for a review of data resources sorted by strategic priorities set out by HHS, by type of data (national surveys, electronic health records [EHR], claims data, mortality records, prescription monitoring data, contextual and policy data, and level [national, state, local]. See also Scrivner et al. (2020) for an interactive visualization of commonly used data resources at Public Data Resources Used in Studying Opioid Crisis that enables navigation through the data dictionaries and recent publications by data set.