The Oxy Electorate

A Scourge of Addiction and Death Siloed in Fly-Over Country

Kathleen J. Frydl
10 min readNov 16, 2016


In 2015, over 3,000 people died of a drug overdose in Ohio. According to the federal government, the national rate for drug overdoses is 14.7 people per every 100,000. In Columbiana County, Ohio, a place where medical staff routinely check for overdose victims dumped in hospital parking lots, the state Department of Health estimates the rate at 19.9.

Columbiana lies in the heart of “steel valley,” a strip of Pennsylvania, Ohio, and West Virginia where most mills closed a long time ago, and those that are left operate with a dramatically reduced workforce. The loss of the region’s main economic engine is like a generational time bomb, a slow-motion detonation that reverberates for decades in empty main streets and collapsing local populations.

But in one area Columbiana recently registered a modest positive gain. The number of people who cast a ballot in the 2016 presidential race was greater than in 2012, even though, as a state, Ohio recorded a net loss in turnout from the previous election. This pattern holds for nearly all opioid-ravaged counties. And not just in Ohio — in Pennsylvania, Wisconsin, and Michigan as well, all of them crucial to the outcome of the election. In 9 of the Ohio counties that Trump successfully turned from Democrat to Republican, six log overdose rates well above the national norm. All of the Pennsylvania counties that chose Obama in 2012 and Trump in 2016 have exceptionally high overdose rates, averaging 25 people per 100,000; in none of these counties did vote totals fall.

Although we await hard data on voters to learn more about the patterns that produced last Tuesday’s shocking upset, we already know that Donald Trump drove rural and ex-urban voters to the polls in numbers that compensated for his weaker turnout in suburban areas. He did not flip every one of these counties; many were already favored to vote Republican. But he did run up the margin separating him from Clinton, accumulating a vote total in his favor that proved impossible for his opponent to overcome.

In fact, other states that remained in the Democratic column show evidence of these same dynamics. New Hampshire and New Mexico, rocked with overdose rates that rank among the highest in the nation, also saw a surge in rural voters in favor of Trump in counties that had no drop-off in overall turnout. Since individual states award the electoral college based upon the popular vote, the “new” voter that Trump was able to produce, convert, or excite suggests that, if this momentum is sustained and unchecked, Democrats will confront an even more unfriendly electoral map in the next election.

Rustbelt and Much of Rural America in Free Fall

In 2014, the last year for which national data is available, 47,000 people died of a drug overdose, now the leading cause of acute preventable death in the country. This figure is comparable to the number US combat deaths listed for the entire war in South Vietnam (58,220). Since 2000, roughly half a million lives have been lost to drug overdoses, exceeding American casualties in World War II. When it comes to drugs, rural and ex-urban America have joined inner cities in fighting a war that the rest of the country has evaded, and refused to acknowledge in its scale and significance.

Although the apparent correlation between the places most under siege by the opioid epidemic and those that flipped or went more heavily for Trump is striking, it is only suggestive, and not by itself explanatory. It does nothing to undermine the convincing evidence that these voters were motivated by an ethno-nationalist appeal, for example. Instead, it supplements it, providing a premise to help explain why a narrative of identity, place, and belonging would have such resonance among them. Success will come to most any strategy that proclaims the value of people living in places that country has, in its collective indifference, discerned little.

In fact, Trump’s “oxy electorate” is swimming in a sea of correlations, not just to high rates of drug overdoses, but, as demographer Shannon Monnat has found, to suicide and deaths from alcohol as well, not to mention dramatic reductions in economic mobility.

Not any single thread can be identified as “causal,” but all help to form a picture of Trump’s appeal — namely, a rhetoric and a campaign strategy that made these voters feel seen and heard, not just as measured against the Democratic Party, but within the power structures of the Republican Party as well. His rage matched their anguish; his gaze met their eyes.

To slow the traffic in drugs, Trump vowed to “build a wall,” a drastic measure that was widely belittled. But people who live in forgotten places understood that, even if the promise was absurd, their suffering had finally registered. According to StatNews, Trump supporters in Brown County, Ohio, a place that records one of the highest opioid overdose rates I’ve seen (40.2), doubted the Republican candidate could reverse, in any meaningful way, “the damage that has already been done.” Still, Trump was “worth a chance,” one man said, “because nothing’s happening now…We know what the hell we’ve been having, and it’s been continuing on. It hasn’t gotten any better.”

Researchers at The Economist found a “highly statistically significant” correlation between a county’s poor health metrics and a swing in favor of Donald Trump

The Comprehensive Addiction and Recovery Act (CARA), passed this summer by Congress, failed to make an impression in Brown, and other places like it, perhaps because, as President Obama said when signing it, the “modest steps to address the opioid crisis” included in it were not nearly adequate to meet the need. Prescription drug “take back” programs and block grants serve more as symbolic gestures than programmatic responses in places suffering a mortality crisis on levels typically reserved for the developing world.

Nor did Democrats seem “worth a chance.” Days after Obama expressed his disappointment in CARA’s provisions, he traveled to the Democratic convention in Philadelphia to join others in professing his belief that “America is already great.” Even “low information” voters, as defensive Democratic strategists sometimes refer to Trump supporters, can be forgiven for thinking that the Democratic Party discarded them long ago; the failure of the opioid epidemic to penetrate the national narrative, much less the Democratic campaign mantra, only punctuates this indifference, like a bolt of recognition in an otherwise mostly atmospheric impression.

What is most unsettling is that, on the question of the opioid crisis, more “information” would only enhance the notion that Democrats don’t care. Although something on the scale of this epidemic requires participation and enabling from both parties, it is actually the Democrats who stand more to blame.

As this year’s Los Angeles Times investigation revealed, it was in 1995, under a Democratic administration, when Purdue Pharma was able to manipulate Food and Drug Administration patent approval Oxycontin, the drug that launched the opioid epidemic, despite the fact that in drug trials and in subsequent clinical use, patients told their physicians that OxyContin’s slow-release technology — the key to its patent, and the premise of the company’s assertions that the drug was not addictive — did not work for them. Regulators looked the other way, and both the Times and StatNews have since chronicled the ways in which this faulty patent allowed the industry to persuade medical professionals to expand prescription narcotic use beyond the customary realm of palliative (end of life) care or acute pain recovery to much bigger and more profitable market of chronic pain treatment.

Officials in the state of West Virginia — also in the steel valley — were among the first to suspect that Oxycontin was more addictive, and less reliable on slow-release, than Purdue let on. After the state filed suit, the company, desperate to avoid trial, hired a law firm that sent former Deputy Attorney General Eric Holder Jr. to argue for a motion for summary judgment. After he failed, Purdue quickly settled the case, and the incriminating documents from legal discovery, held under legal seal, have come to light only recently via the Times investigation, despite the overwhelming public interest that such material be made available.

Also of note is Eric Holder’s role in the government’s lethargic response to the opioid crisis once he became Attorney General, something that was mentioned in a Washington Post investigation published on October 22, just days prior to the election. Reporters found that, as Drug Enforcement Agents in the field begged to prosecute opioid distributors flagrantly in violation of inventory supervision, senior officials at the Department of Justice took their cues from an Attorney General who remained committed to meeting with the representatives of the pharmaceutical industry and working toward a voluntary solution. Rather than punishment or meaningful penalties — Purdue Pharma has paid $634.5 million fine on a drug that has earned it $31 billion in revenue, or 2 percent of earnings — Holder sought an amicable understanding.

Not surprisingly, in face of powerful appeasement, the prescribing norms for narcotic painkillers pioneered by Purdue remain intact, even as some individual states have moved in favor of limiting the number of days a doctor can prescribe opioids for acute pain. Without disparaging the work of many good people on the ground, there is no such thing as an effective local response to problems that are national in origin and in scope. Where local and state officials have been successful in contracting licit supply, plenty of heroin from Mexico, and illicitly produced synthetic opioids from China, flood the market to satisfy demand.

And this seems a crucial point: just because counties that favored Trump have comparatively low rates of immigration, does not mean that the people living in these same places feel immune to foreign influence. It is more likely they feel besieged by it. Over the summer, readers of local papers like the one in Luzerne County, Pennsylvania — one of three counties that Donald Trump flipped — learned that fentanyl, “concocted in Chinese labs,” posed a potency problem for which no one seemed to have an answer. Its even more potent successor carfentanil, also manufactured in China, is so dangerous that first responders must use excessive caution when exposed to it. Turns out, there was a foreign power that illegally interfered in this election in nefarious ways; it’s just not the one you think.

Democrats in Trouble

Prior to the election, some Democratic officials, no doubt expecting victory, promised to form a 527 (a group with a tax status that enables participation in politics) devoted to breaking down voting barriers, and they put Eric Holder’s name forward as a potential leader. It would be hard to find a more efficient way to communicate utter indifference to the struggles of Americans than the appointment of someone who, in two successive encounters, fanned the flames of the opioid epidemic. Clearly, the perception of the Democratic Party’s indifference to rural and ex-urban voters comes from more than just biased media outlets exploiting the fertile ground of discontent in regions where entire industries have gone extinct. It derives as well from specific actions that convey contempt.

As Senator Bernie Sanders said recently, white working class voters “turned their backs on the Democratic Party.” Many stayed home, others voted third party, and some responded to the call of Donald Trump’s campaign, a splenetic diatribe of insularity and disgust.

Nothing suggests to me that voters in this last group should be characterized as “economic populists,” an implicit vindication or ennobling of the cause for which they cast their ballots. But I am not surprised by their selection either. Then again, after decades of suffering decades of disdain and divestment, when some African-American neighborhoods and prison networks revitalized the Nation of Islam in the early 1990s, I also wasn’t surprised. Though the Nation offered a more tightly-stitched and explicitly racialized narrative than Trump, the neglect and the need in the places they recruited was only that much greater.

Sometimes people answer the door for the only person knocking.

Below is an Airtable comprised of Ohio Counties that record overdose fatalities at (or nearly at) 20 people per 100,000 or higher. All saw at least a 10% surge in the number of voters who favor Trump over Romney, and/or a 10% drop from Obama to Clinton, with the two exceptions of Butler, home to Miami University, and Hamilton, the county jurisdiction for Cincinnati. Five of them flipped from Democrat to Republican.

Below is the data for Pennsylvania, which is particularly important given its overall rise in voter turnout, and New Hampshire, which Hillary Clinton won…barely.

In Pennsylvania, only 4 of 33 high-overdose counties fail to comport to the pattern of flipped from Democrat to Republican, and/or a surge of more than 10% in the number of voters choosing Trump, and/or a decline of more than 10% from Obama to Clinton. One is Alleghany, home to Pittsburgh, and another is Lehigh, home to Allentown; only Philadelphia surpasses them in population — and in fairness, I think Philadelphia should be grouped with them, since it’s probably reasonable to conclude that factors other than a drug epidemic led Donald Trump’s popularity among voters there. Delaware County, another exception, fits with a pattern already noted in Ohio: it is home to three major universities. So Dauphin County stands as the single exception to what might be expected; there Donald Trump fell just shy of surging 10%.

I included New Hampshire even though I don’t have 2015 county level overdose rates because the counties most affected are well known to state authorities and tracked on a month-to-month basis. The Democratic stronghold of Strafford withstood the challenge of the 10% threshold, and I am assuming Portsmouth and Exeter helped Rockingham do the same. Still, they provide a feint echo of the pattern that can be discerned elsewhere, as the other three counties attest. Even in victory, the Democrats should be alarmed.