How Did the Opioid Epidemic Start? The Three Waves

The U.S. opioid epidemic grew out of a well-intentioned but catastrophic push in the 1990s to treat pain more aggressively, amplified by pharmaceutical companies that downplayed addiction risks while flooding the market with powerful painkillers. Public health researchers describe the crisis in three overlapping waves: a first wave driven by prescription opioids starting in the late 1990s, a second wave of heroin deaths beginning around 2010, and a third wave of synthetic opioid fatalities that accelerated after 2013. Each wave fed on the one before it, and a possible fourth wave involving stimulants mixed with fentanyl is now taking shape.

The Cultural Shift That Set the Stage

Before any wave of deaths, there was a shift in how American medicine thought about pain. Through the 1980s and early 1990s, doctors were broadly cautious about prescribing opioids for anything other than cancer pain or end-of-life care. Critics argued this caution left millions of patients suffering needlessly. In 1996, the American Pain Society launched its “pain as the 5th vital sign” campaign, urging clinicians to assess and treat pain with the same urgency they gave to blood pressure or heart rate.¹ The initiative was adopted across hospitals, veterans’ clinics, and accreditation standards. Patient satisfaction surveys began penalizing providers whose patients reported uncontrolled pain.²

On its own, asking doctors to take pain seriously was reasonable. But the campaign created a pressure system that pharmaceutical companies exploited. Clinicians were now expected to do something measurable about every patient’s pain, and opioid manufacturers were ready with a product to sell.

Wave One and the Role of Pharmaceutical Marketing

The first wave of opioid overdose deaths began climbing in the late 1990s, with prescription-involved deaths starting between 1999 and 2005 depending on the state.³ The drug at the center of this wave was OxyContin, a sustained-release form of oxycodone manufactured by Purdue Pharma. OxyContin was approved in 1995 and marketed with an intensity that was unusual for a Schedule II narcotic.

Purdue built its campaign around the claim that OxyContin’s extended-release formula made it less prone to abuse than short-acting painkillers. Sales representatives were trained to tell doctors that the addiction risk was “less than one percent.” That figure was drawn from a misrepresented letter to the editor in a medical journal, not from any controlled trial. The company distributed promotional literature, patient brochures, videotapes, and a website called “Partners Against Pain,” all of which systematically minimized the risk of addiction.⁴ In 2007, Purdue Frederick Company (a Purdue affiliate) and three of its executives pleaded guilty to criminal charges of misbranding OxyContin as less addictive than other opioids and paid $634 million in fines.⁴

The mislabeling and mass-marketing of OxyContin became a catalyst for the opioid crisis not just in the United States but eventually beyond it.⁵ Between the late 1990s and the late 2000s, prescriptions for opioid painkillers soared. Millions of patients received pills for back pain, dental procedures, and post-surgical recovery. Many developed physical dependence. Some progressed to addiction. And as pill supplies tightened through regulation and reformulation, a large population of dependent users was left searching for alternatives.

Wave Two and the Shift to Heroin

The second wave of overdose deaths, driven by heroin, started between 2010 and 2014 across different states.³ A key trigger was the 2010 reformulation of OxyContin. Purdue redesigned the pill so it could not be easily crushed and snorted or dissolved and injected. Research suggests this reformulation substantially reduced access to abusable prescription opioids in the short term.⁶

The reformulation’s effects on individual users were more nuanced than the population-level story suggests. One study found that the reformulation reduced the odds of prescription painkiller misuse and heroin initiation among people who had been misusing OxyContin, without pushing them into greater heroin use.⁷ But at the population level, the transition was unmistakable: as prescription opioid supplies tightened, heroin filled the gap. Heroin was cheaper per dose and widely available in many cities. The people who turned to it were often not the stereotypical users of decades past. They were suburban adults who had started on pills after a surgery or an injury and found themselves dependent.

This transition highlights one of the epidemic’s cruelest dynamics. Policies that successfully reduced prescription opioid misuse did not eliminate the underlying addiction. They changed the supply channel.

Wave Three and the Arrival of Fentanyl

The third wave, driven by illicitly manufactured synthetic opioids, began between 2014 and 2021 depending on the state and quickly became the deadliest phase of the crisis.³ Illicit fentanyl and its analogs flooded the drug supply, and overdose deaths from synthetic opioids continued rising both before and during the COVID-19 pandemic.⁸

What makes fentanyl so dangerous is its potency. Structural studies show that fentanyl activates the brain’s main opioid receptor roughly 50 to 100 times more potently than morphine, partly because fentanyl’s molecular shape lets it interact with receptor pockets that morphine cannot reach.⁹ In practical terms, a lethal dose of fentanyl can be as small as a few grains of salt. This extreme potency means that even tiny miscalculations in mixing produce fatal batches. Users often do not know fentanyl is in their supply at all: it is pressed into counterfeit pills designed to look like prescription medications, or mixed into heroin, cocaine, and methamphetamine.

Unlike the first two waves, the fentanyl wave did not arise from a medical system. It is a supply-side phenomenon driven by illicit manufacturing, primarily in clandestine labs, using cheap precursor chemicals. The economics are stark: fentanyl is far cheaper to produce per dose than heroin, and a small amount goes an enormously long way, making it attractive to traffickers. This made the third wave harder to address through prescribing regulations or patient education, since the drug was no longer coming through pharmacies.

Signs of a Fourth Wave

Some researchers now describe a fourth wave characterized by polysubstance overdoses, particularly fentanyl combined with stimulants like cocaine and methamphetamine. The share of U.S. overdose deaths involving both fentanyl and stimulants rose from under 1% in 2010 to about a third in 2021, with the steepest climb starting around 2015.¹⁰ Forensic analyses of overdose decedents confirm the pattern: in one study, about 27% of fentanyl-positive cases also contained cocaine, and 15% contained methamphetamine.¹¹

Complicating matters further is the rise of xylazine, a veterinary sedative now frequently found mixed with fentanyl in the street drug supply. Xylazine is not an opioid and does not respond to naloxone, the standard overdose reversal drug. It constricts blood vessels and causes distinctive necrotic skin ulcers in people who inject it, creating a wound-care crisis in cities where the combination is prevalent.¹² The xylazine-fentanyl combination introduces new medical challenges because standard overdose protocols may not be sufficient when a non-opioid sedative is also suppressing breathing and circulation.¹³

The “Deaths of Despair” Debate

A popular explanation for the epidemic frames it as a symptom of economic decline, particularly in deindustrialized communities. The “deaths of despair” hypothesis, associated with economists Anne Case and Angus Deaton, links rising drug deaths, suicides, and alcohol-related mortality to job losses, stagnant wages, and eroded social ties among working-class Americans. There is some truth to the connection: research confirms that economic precarity, opioid accessibility, and working-class population size interact to predict overdose deaths at the county level.¹⁴

But the relationship is weaker than the narrative implies. An analysis of county-level data found that changes in economic conditions explained less than a tenth of the rise in drug and opioid mortality, and even that small association largely disappeared after accounting for confounding factors.¹⁵ The researchers concluded that while improving economic conditions in distressed areas is worthwhile for other reasons, it would be unlikely to produce significant reductions in drug deaths. In other words, economic despair made communities more vulnerable, but the epidemic’s primary engine was the drug supply itself, amplified by aggressive marketing and inadequate regulation.

Why Europe Largely Avoided the Crisis

If economic hardship alone explained opioid epidemics, you would expect similar crises across the developed world. But Europe, despite having its own economically depressed regions, has not experienced anything like the American opioid catastrophe. Several structural differences help explain why.

Direct-to-consumer drug advertising, which Purdue used aggressively in the U.S., is prohibited in European countries. Illicitly manufactured fentanyl, while present in some places like Estonia, has not broadly infiltrated European drug markets the way it has in America. “Doctor shopping” and diversion of prescription opioids are less common in European health systems. And critically, medication-assisted treatment for opioid addiction is generally more accessible and often free of charge across much of Europe, reducing the pool of untreated people who might turn to the illicit market.¹⁶ The American epidemic was not inevitable. It was shaped by specific policy and market conditions.

Policy Responses and Their Limits

The most widespread regulatory tool has been the Prescription Drug Monitoring Program, or PDMP. These are state-run databases that track controlled substance prescriptions so doctors and pharmacists can spot patients receiving opioids from multiple providers. When states mandated that prescribers check the PDMP before writing opioid prescriptions, opioid prescribing rates dropped, opioid-related emergency visits fell, and opioid-related hospital stays declined, saving an estimated $155 million in Medicaid spending annually.¹⁷

That sounds like a success, but a systematic review of PDMP research found the broader picture was less clear. PDMPs did not show consistent evidence of reducing illicit opioid use, opioid dependence, or emergency department visits at the population level. Results on overdose deaths and treatment admissions were conflicting, with some studies showing improvements and others showing none.¹⁸ The basic problem is the same one that drove wave two: reducing the prescription supply does not automatically reduce the demand.

The CDC’s 2016 opioid prescribing guideline had a similar double-edged effect. It successfully shifted chronic pain patients toward non-opioid alternatives and encouraged dose tapering, with no increase in work limitations due to pain among those who discontinued opioids.¹⁹ But the guideline also triggered a wave of aggressive tapering. The rate of tapering and rapid discontinuation both jumped immediately after the guideline was released and continued climbing month over month.²⁰ Among patients who were tapered, more than a quarter experienced a maximum tapering rate above 40% per month, which is much faster than most clinical experts recommend.²¹ Rapid tapering can cause severe withdrawal symptoms and has been linked to patients seeking illicit opioids. The CDC itself later revised its guidance in 2022 to explicitly caution against abrupt dose reductions.

Medications That Reduce Overdose Deaths

The most effective intervention for people already living with opioid addiction is medication-based treatment, sometimes called medication for opioid use disorder (MOUD). A large cohort study of people who survived a nonfatal overdose found that methadone treatment was associated with a 53% reduction in all-cause mortality and a 59% reduction in opioid-related mortality compared to receiving no medication. Buprenorphine was associated with a 37% reduction in all-cause mortality and a 38% reduction in opioid-related mortality.²² Naltrexone, a third option, did not show a statistically significant mortality benefit in that study.²² When comparing buprenorphine and methadone head-to-head, mortality while in treatment was low for both, and the statistical difference between them was ambiguous.²³

Naloxone, the emergency overdose reversal drug, has become a critical tool distributed through pharmacies, community programs, and first responders. In the fentanyl era, there has been debate over whether standard doses are sufficient. Some researchers have argued that fentanyl’s rapid onset and high receptor occupancy necessitate higher naloxone doses.²⁴ But a review of available evidence concluded that most fentanyl overdoses can be reversed with two standard intramuscular or intranasal doses, with exceptions for extremely potent analogs like carfentanil. High-dose naloxone formulations carry a greater risk of precipitated withdrawal, which can itself be dangerous.²⁵ The practical takeaway for bystanders is that carrying naloxone and being willing to administer a second dose matters more than having a single ultra-high-dose formulation.

Surgery as an Overlooked Entry Point

Much of the conversation about the epidemic’s origins focuses on chronic pain prescribing, but surgery remains a significant gateway to opioid dependence. A study of more than 30,000 patients prescribed opioids at discharge found that the risk of new persistent opioid use increased with initial prescription size. Patients who received the largest prescriptions had roughly double the odds of persistent use compared to those who received the smallest.²⁶ Longer hospital stays, high-dose initial prescriptions, and female sex were all associated with greater risk of continued use after general surgery.²⁷

Being prescribed an opioid at discharge after surgery was not itself a risk factor for persistent use. The issue was how many pills the patient received. This distinction matters because overly restrictive policies that eliminate all post-surgical opioids can leave patients in unnecessary pain, while overly generous prescriptions hand patients more pills than they need, creating leftover medication that can be misused or diverted.

Genetic Vulnerability and Who Gets Addicted

One of the most persistent questions people have about the epidemic is why some individuals become addicted while others take the same drugs without developing a problem. Family and twin studies have consistently shown that genetics plays a substantial role in susceptibility to opioid addiction, as it does for other addictive disorders.²⁸ But the genetics are complex, involving many small-effect variants across multiple genes, with significant variation between populations. No single gene test can predict who will develop an addiction. The practical implication is that addiction is not a moral failure or a simple choice. It is a medical condition with biological roots, influenced but not determined by environment and behavior.

The Epidemic’s Youngest Victims

The opioid crisis has also created a generation of affected newborns. Neonatal opioid withdrawal syndrome, or NOWS, occurs when infants exposed to opioids in the womb go through withdrawal after birth. Its incidence has risen substantially alongside the epidemic.²⁹ Symptoms include tremors, irritability, feeding difficulties, and in severe cases the need for pharmacological treatment. Whether the mother is receiving medication-assisted treatment matters for the baby’s outcomes. Infant mortality was higher among babies whose mothers did not have medication for opioid use disorder claims, and sudden unexpected infant death was the primary cause of death in both groups.³⁰ Treating the mother’s addiction does not just protect her; it appears to protect the baby as well.

1
British Journal of Anaesthesia. How Did the Opioid Epidemic Start? The Three Waves
2
PubMed Central. Moving Beyond Pain as the Fifth Vital Sign and Patient Satisfaction Scores to Improve Pain Care in the 21st Century
3
PubMed Central. A state-level history of opioid overdose deaths in the United States: 1999-2021
4
PubMed Central. The Promotion and Marketing of OxyContin: Commercial Triumph, Public Health Tragedy
5
PubMed. Interconnected influence: Unraveling purdue pharmaceutical’s role in the global response to the opioid crisis
6
PubMed Central. THE EVOLVING CONSEQUENCES OF OXYCONTIN REFORMULATION ON DRUG OVERDOSES
7
PubMed. The impact of the abuse-deterrent reformulation of extended-release OxyContin on prescription pain reliever misuse and heroin initiation
8
PubMed Central. The rise of illicit fentanyls, stimulants and the fourth wave of the opioid overdose crisis
9
Cell. Structures of the human μ-opioid receptor bound to morphinan and fentanyl analogs
10
PubMed. Charting the fourth wave: Geographic, temporal, race/ethnicity and demographic trends in polysubstance fentanyl overdose deaths in the United States, 2010-2021
11
PubMed Central. The Fourth Wave of the Opioid Epidemic: Increasing Combination of Fentanyl With Stimulants
12
PubMed Central. Xylazine-Induced Necrotic Skin Ulcers in a Fentanyl-Injecting Individual in South Florida, United States: A Case Report
13
PubMed. Xylazine Adulteration of the Heroin-Fentanyl Drug Supply: A Narrative Review
14
PubMed. The needle and the damage done: Deaths of despair, economic precarity, and the white working-class
15
NBER. Deaths of Despair or Drug Problems?
16
PubMed Central. Is Europe facing an opioid crisis like the United States? An analysis of opioid use and related adverse effects in 19 European countries between 2010 and 2018
17
PubMed Central. Prescription Drug Monitoring Program Mandates: Impact On Opioid Prescribing And Related Hospital Use
18
PubMed Central. The effectiveness of prescription drug monitoring programs at reducing opioid-related harms and consequences: a systematic review
19
PubMed Central. Opioid and non-opioid analgesic prescribing before and after the CDC’s 2016 opioid guideline
20
PubMed. Long-term opioid therapy tapering: Trends from 2014 to 2018 in a Midwestern State
21
JAMA Network Open. Trends and Rapidity of Dose Tapering Among Patients Prescribed Long-term Opioid Therapy, 2008-2017
22
PubMed Central. Medication for Opioid Use Disorder After Nonfatal Opioid Overdose and Association With Mortality: A Cohort Study
23
JAMA. Buprenorphine/Naloxone vs Methadone for the Treatment of Opioid Use Disorder
24
PubMed Central. Higher doses of naloxone are needed in the synthetic opiod era
25
PubMed Central. High-dose naloxone formulations are not as essential as we thought
26
PubMed Central. Postoperative Opioid Prescribing and New Persistent Opioid Use: The Risk of Excessive Prescribing
27
PubMed. Postdischarge opioid use and persistent use after general surgery: A retrospective study
28
PubMed Central. Genetic Vulnerability to Opioid Addiction
29
PubMed Central. Neonatal Opioid Withdrawal Syndrome (NOWS): A Transgenerational Echo of the Opioid Crisis
30
PubMed. Differences in Mortality Among Infants With Neonatal Opioid Withdrawal Syndrome