Tag Archives: rural america

Is Health Care a Democratic Right?

The bulbous, bloated cartogram meant to render the prospective withdrawal of insurers from individual health-care exchanges under the Affordable Care Act in its deepest colors foregrounded the concentration of a reduced insurance market in rural areas.  The image of a nation seething over, and the ground boiling over as if with discontent, cartogram, distorted “red” and “blue” states alike, but seemed evidence of the ways that the Affordable Care Act warped the even playing field in the United States–as if that ever existed, and could be expected to exist today.  But it might be taken as an emblem of the deep anger and resentment that many areas of the United States felt toward what was labeled “Obamacare,” and the regional markets left with but one insurer–and an inevitable feared rise in their insurance premiums–in the health insurance markets that the Affordable Care Act would create.

The designers of the cartogram warped to counties’ population almost aptly if inevitably rendered the country as boiling over with anger:  it seemed to render a powerful emblem to justify if not inspire broad indignancy about the apparently uneven consequences of mandated insurance exchanges which it argued the less populated–and poorer–areas of the country would be stuck with, as a distillation of social injustice.  For the cartogram captured what its designers argued was the distorted market for health insurance which people on the coasts had designed as destined to shut out large areas of the country shaded in lighter colors–and prevalently light pink.  But the prediction of a contraction of providers that undergirds this ominous scenario, as we now know, didn’t come about at all,–even if the strong passions provoked by the fight over health care did leave the country boiling over with anger and indignancy widely felt to be objectively justified.

 

map2_20170725Warped Map on Insurers Red v Blue Goves

It can be quite forcefully argued that health care deserves to be regarded as a  democratic right–democratic with a “small ‘d,'” in the sense of an egalitarian right, even though debates about access to health insurance are increasingly cast in politically partisan terms.  Although access to insurance exchanges are increasingly treated as a question less of a right than the reflection of a political position, the proposition of guaranteeing health coverage is rejected by champions of the marketplace and its benefits, who argue that its falsity undermines a free market.  As a result, in part, health-care exchanges are increasingly mapped in terms that might well be mistaken for political partisan divisions within the fabric of the nation.  Indeed, the sharp, flat blues, reds, and deep carmine of different regions suggest the hopes and difficulties of providing a uniform insurance plan for a nation of radically different numbers of insured, facing the hope to provide more with coverage in a way that may seem to tilt against the open nature of the marketplace.

 

Us Marketplaces.png

 

But democratic rights include not only political participation but due rights to certain benefits that accord undeniable liberties.  And although liberties which were not defined as including health care in the eighteenth century, leading many strict constructionists to view health insurance as an excessive presence of the state in individual lives, the range of liberties have expanded to-liberties to education, or to health, or protect against race-based, ethnic, or sex-based discrimination–revealing the broadening scope of understanding liberties, and might be  mapped into the fabric of the nation as an individual rights, and a basis for ensuring greater egalitarianism–and social equality–as a right.

 

Obama-healthcareBlack Enterprise

 

Back in 2013, of course, the institution of health-care exchanges set up a new landscape of the Patient Protection and Affordable Care Act–the Affordable Care Act or simply ACA–allowing most Americans to buy insurance on  government-run exchanges (or marketplaces) to have access to health care that they were often lacking in all fifty states, creating the reign of designing data vis of Obamacare exchanges often subsidized by the government for those eligible, to make it available to all, in ways that created some thirty-six exchanges run with the federal government, as states ran the others alone–creating the odd scenario that more enrollees came from red states, where they were run largely by the federal government.

 

state-health-insurance-exchange-landscape

 

But this was not widely accepted, and the rejection of the promise of what is now widely labeled Obamacare reflect the deep divides that its opponents argue the government mandate for buying insurance policies will impose on the nation.  The online popularity of recent projections of a constriction of health insurance options for most counties of the nation that were proposed as recently as the spring of 2017 seemed to reveal the deep dishonesty in the proposals of the Affordable Care Act to level the playing field.  And although the capital of health care as a good to the nation demands to be mapped, the difficulty of parsing the ACA independently of the name of Barack Obama reflect the unfortunate polemic level of debate about seeing adequate health insurance as a right of all–even as fear of losing health insurance dramatically rose across the nation, and the fears of rising premiums posed by its mandates were widely stoked by data visualizations suggesting widespread abandonment of insurance exchanges.

 

Changes in Providers.pngUS Health Policy Gateway

 

The data visualization–which almost amounts to a tool of outright propaganda–uses flat carmine to blanket the real improvements in numbers of the uninsured.  The presentation of an apparent distortion of the market is confirmed by declining insurance policies available on Obamacare exchanges, as its accompanying text assures readers that the real people to benefit from “marketplace enrollment” was the “private health insurance industry” who gained $90 billion in premiums, greatly profiting “publicly traded insurance companies”–distorting meanings of “public” and “private” as if to imply the dystopian nature of health exchanges that benefit coastal states alone.  The map of possible changes in rates of premiums were even more striking, and was presented as evidence of poor policy planning, as well as signs of a grim “slow motion death spiral”–a strategic choice of term suggesting the poor level of health-care it provided, and organically faulty nature of its establishment, but alienating the numbers of premium growth from individuals covered.

 

obamacare-premium-map.png

465213670_wide-c790535ce6a6b088fa700075b6c1de652f014904-s1600-c85.jpg

 

The familiar series of sequential images forecasting the mass exit of insurers from exchanges over a period of four years of the adoption of Obamacare stand at odds with the fears of losing health care and the defense of health care as a right, as well as a national system of insurance.  And despite an onslaught of maps  ostensibly demonstrating the ever-narrowing options for individuals as available insurers in state-based exchanges  dry up; they convey an imaginary future in which few counties with “active markets” of four or more alternatives–apparently compromising the rights of many Americans.

 

nssustaniable decline

Bloomberg Graphics/2017 Health Insurer Exits (projected)

 

Indeed, the image of rapidly dwindling options faced by Americans that such data visualizations claim to be based on data from HealthCare.gov. seem to suggest a focus on individuals.  But the broad brush strokes leave little to the imagination and present an ominous emptying of choice that seems designed to induce panic.  The images are executed with great dramatic effect, but little sense of cartographic skill–they presented a dire picture in which four options would be only available to residents of eight to ten states by 2017, calling into question the ability of much of rural America to remain insured.  The images of rural abandonment by health care exchanges were particularly powerful,  so absolute in their predictions as if to afford little room for interpretation

Yet the projections, for all their power, stand at variance with reality.  There will be, we now know, in fact no Obamacare marketplaces that remain at risk of being without insurers in 2018, as of August 24, 2017–

 

w:o insurers in echange

–and but a smattering of counties that were at risk for being without any insurers:

 

at risk of no insurer

Kaiser Family Foundation Interactive Version/Open Street Maps:  Counties at Risk of Having No Insurer on the Marketplace in 2018 as of August 24, 2017

 

In short, the disruption of the narrative of a dwindling of insurance options has been, after the failure to repeal the ACA, dramatically disrupted.  Even while acknowledging that there was a record low of uninsured in America after the American Care Act was adopted in “Obamacare marketplaces”–a coded term if there ever was one, loaded with disparagement–data visualization were crafted to predict deteriorating coverage options deteriorated in the months ahead in many rural states of apparent objectivity; hastily created maps, at an odd angle to reality, suggested that as much as over a fifth of all federally run marketplaces–predominantly in rural areas–were with only one insurer.

The alleged “bolting” of insurers from such marketplaces were predicted to leave areas like eastern Tennessee without any insurers, like, perhaps, southern Georgia, much of Colorado, almost all of Iowa, many counties in Ohio,  and large numbers of Virginians, as Aetna, Wellmark, and Anthem were predicted to “bolt” from the exchanges, leaving those Obamacare “customers” high and dry.  The argument of the abandonment of rural America was particularly grim.  But as the projections of the “bolting” of insurers fail to acknowledge the sparsely populated nature of many rural areas, the story that they tell of magnifies the poorly managed nature of the marketplace, obscuring the benefits or rights to health care–and the reduction of the number of uninsured across America– that the ACA has managed to create.  By privileging the size of largely unihabited regions of the midwest, maps of uninsured counties presented a decidedly skewed picture of enrollment, where the square miles covered by insurers projected to participate in health exchanges  seemed to outweigh those where insurers participated, irrespective of the sizes of inhabitants.  It is perhaps no surprise that support for Trump’s candidacy did not correlate with support for the ACA–

 

ACA-support-HaystaqDNA-score-by-county

 

 

We focus on individuals to measure popularity for the support for health care reforms across the country.  Although many have recently entertained sustained interactive levels of introspection about where Americans supported  the Affordable Care Act in the months that preceded the election, pouring over the support for the ACA through county-by-county lenses that made sense pretty much only in how they might translate into votes.

While moving toward the acknowledgement of health care as a right is independent from such measurement of support for the ACA, the  low support for the act in sparsely populated areas intensifies as one moved to less populated areas, by and large, to suggest poor penetration of exchanges into much of the nation–and the distance of health care from what seemed in square miles a quite considerable geographic area.  Resistance to the ACA however reflects a rejection of the broad classification of health care as a right–or to even start to affirm it as one nation.   The division of the country, while reflecting the red state/blue state map in many ways, suggest pockets of counties with strong support for the ACA in a surprising range of the south, southwest, and other regions–across the divide between red states and blue.  Haystack’s micro targeting models estimated just under 98,943,000 ACA supporters nationwide–wondering how the electorate would parse on such a push-button issue.  And, indeed, the Senate Republicans were quick to issue a somber grey data visualization that affirmed a clearcut divide suggestive of the status of yellow- or red-alert in areas “abandoned” by Obamacare– in an openly partisan moment, undoubtedly funded by tax dollars.

Senate Republican Party flawed policies 2018U.S. Senate Republican Policy Committee

 

If over a third of the nation, colored an arresting yellow, seem to flounder in facing monopolies of insurance in the image that the Republican Policy Committee in the United States Senate designed for public consumption, who seem to have sought to raise a yellow- or red-alert for subscribers of Obamacare being abandoned, the map foreground  a divide in deeply partisan ways, failing to note persistently steep inequalities among  uninsured across the states, and the difficulties to attract insurers to markets in equal numbers, particularly in regions where up to a fifth remained uninsured in 2001-3.

 

% Uninsured 2002-3.png

 

 

Despite some questions of whether Republicans would be “alienating their own voters” didn’t affect the results of 2016 congressional elections.  But the power of the continued threat of a coming “implosion” of exchanges that upset the level playing field as an inevitable occurrence was successfully manufactured in projections of insurance markets that peddled groundless prognostications as if they were objective fact:  they successfully mobilized fears of the ability to avoid or precipitate a coming crisis by making it concretely manifest for viewers, pushing many to question the benefits that the extension of the ACA would actually bring–and to see it as a promise that would not be able to guarantee continued coverage or familiar premiums, and indeed to be engineered by the coastal elites and insurance companies that so much of the country has already come to distrust.   For the data visualizations that projected the uneven playing field that exchanges would create cast a divided commonwealth as a result of the limited choices restrictive options of health insurance many Americans would face.  These visualizations raised significant alarms bout the fate of Obamacare–and the specter of its undemocratic nature raised questions of what it provided to the country, or what future it might bright–that were deployed in particularly effective ways.

Flat colors of a data visualization communicate as many falsehoods about its actual relation to people as Trump’s favored declamation of Obamacare “very, very bad insurance.”  They obscure satisfaction ratse of over 75% among those enrolled in plans, and of almost 90% in public Medicaid programs for the poor.  Rather, the picture of an implosion of insurance markets garnered ungrounded trust, and became demonized as but “a first step” toward what he presented as the apocalyptic scenario in which the “government basically rules everything”–a fear of the implosion of a free market–ignoring that the American Care Act is premised on encouraging competition among medical insurers.  Yet the image of such an implosion or collapse perpetuated in data visualizations of crude colors was something that was manufactured in projections that masqueraded as objectively designed maps.  In charting decreasing insurer participation in exchanges as actualities, data visualizations seem designed to stoke uncertainty about the future viability of health insurance markets in America.  Yet the uninhabited nature of this landscape of counties–a metric that makes sense only really as a convention of electoral politics, rather than health care or even of individuals residing in different parts of the country, is starkly removed from health care save in terms of how it might translate into a political choice.

The rhetoric of these “maps” uses projections cover the individuals who benefit from medical care.  They encourage voters to feel slighted in new medical marketplaces, and ask them to chose a future–without considering metrics of coverage or the relative quality of medical care.  They serve to map a landscape of fear, encouraging fears of growing premiums and less choice among voters in what is painted as a compromised medical marketplace.

 

1.  Construing health-care as part of a democratic system has been understood in surprisingly partisan terms.  Some would restrict liberties to participation a marketplace, by adopting and privileging the market as a primary metaphor if not end of civil society:  the success of such a distinction has lead to a broad and striking demonization of its mandate, rather than the policies of the health care law signed in 2010–the Affordable Care Act–which as a law has consistently received far less opposition than the change in health insurance provision that mentions President Obama’s name.  The divide in perceptions seems to have been broadened considerably by recent visualizations that project the future market for health care, or project the numbers of insurance carriers available in exchanges, the colors of the availability of carriers overwhelms the presence of individuals, and reveal the new markets that the Affordable Care Act (or ACA) created as if it were an uneven playing field for all Americans.

Indeed, as recently as June 2017 and during the Trump-Clinton campaign, media outlets and websites trumpeted “maps” or “a map” as evidence of the uneven playing fields that the ACA would bring in the country and the restrictive options that were increasingly identified with “Obamacare,” as if it were something different from the health policies that increasing numbers of Americans had enrolled in, but rather a specter of higher premiums, fewer rights, and new restrictions on providers if not health policies that could not be trusted, in ways that continued a drumbeat of visualizations predicting coming imbalances for those enrolled in Obamacare to insurance carriers or to a competitive marketplace–if not rob them of access to insurers–concretizing what Donald Trump cannily called “the broken promise” of Obamacare, as some 2.4 million “customers of Obamacare” would be with but one insurer to select in the coming year.   Health officials in the Trump administration issued a “new map showing in full color how many counties in the United States could have zero or just one insurer selling Obamacare health plans in 2018” as if to provide confirmation of the poor deal that was offered the nation; the data vis produced by the folks at the Heritage Foundation was accompanied by an announcement that, in case any one missed the point, insurance exchanges in 2017 would feature ” a major decrease in competition and choice” (italics added) that exposed the deep failure of the ACA to promote competition as promised:

\

IB-exchange-competition-2017-map-1-825.gifHeritage Foundation (January 30, 2017)

 

The absence of competition left some state, the not unsubtle point was made, that were neglected by the insurance companies that had promoted the ACA:

 

ACA Header?.png

 

The stark contrasts of the data visualization were a rallying cry for a public campaign for the repeal of the Affordable Care Act, designed to activate the red states that were shown to be the largest losers of the insurance exchanges Obamacare created.  The map released shortly after the inauguration in a push to confirm the repeal of Obamacare, showed almost all counties in the southern United States with but one insurer–as almost a third of the counties in the nation–but not the population by any means.  In the rhetoric of an earlier map that described how “large of swaths of yellow cover a number of Southern and Midwestern states, all of Alaska, and elsewhere indicating counties . . .  are projected as of now to have just one insurer selling individual plans next year,” the images of a restriction of opportunities to buy health insurance was alerted, with areas with but one insurer appropriately colored red, as if to convey danger.

The slightly different visualization from June that accompanied this projected danger suggested that some counties–colored red again–would strip residents enrolled in Obamacare from any insurer for those enrolled in Obamacare–this time in “a new map showing in full color how many counties in the United States could have zero or just one insurer selling Obamacare health plans in 2018“:

 

Conties Analysis ObamacareMSNBC (June 13, 2017)

 

The fear that such maps stoked of an imbalance that cut into the insurance options of many as far as health goes suggested a lack of care and a lack of coverage that suggested a deep disinterest of almost a third of the nation, but did so with little actual grounds.  Those sparsely populated regions loose out in the new marketplace that Obamacare seemed to threaten to impose–even if the Affordable Care Act was created to extend health insurance across the nation:

 

Obamacare 2016_0

McKinsey Center for U.S. Health System Reform (August 26, 2017)

 

Such deep divides within the United States that cut against equal access to health insurance was of course what Obamacare was introduced to prevent, but the exchanges in the less populated states were indeed slow to attract insurers.  However, the terrifying fear of a subtraction of any guarantees of well-being and a level playing field that these projections promote–they are hardly really data visualizations, if they resemble maps–seem as good a definition and a metaphor of undemocratic policies, and a metaphor for the restricted roles people are given a crucial say in the policies and decisions that most affect their lives.  Although the sentiments for including health care as a right has become to a deep divide in the nation, the disadvantages that the initial introduction of the exchanges were cautioned to bring to peoples’ lives and policies were immediately striking.

And the recent success of mapping the actual resurgence of insurers’ involvement in many exchanges in counties nationwide reminds us of–and asks us to reconsider–the deceptive nature of their claims.  Indeed, as recently as June, 2017, media sources presented “a map” or a set of maps as evidence of the imbalances that the previous administration had failed to foresee, or willfully imposed on the nation.

 

1.  The negative benefits to all of health-care being a restrictive good are pretty clearly evident:  healthcare should not be seen as a commodity alone, existing on an open marketplace.  Given the clear negative pressures of lowering access to health care to society, the gleeful prediction by President Trump that Obamacare–as Trump calls the Affordable Care Act (ACA)-, as if it were just not American to promise health care to all–would be implode because of e lack of plans available on exchanges in much of the country thankfully seems untrue.  Indeed, the failure to repeal the ACA by the United States Senate–a failure that seems to have sent a shudder of initial convulsions within the Trump administration, and within Donald J. Trump’s sense of his hold on the Presidency, has led insurers to return to the many counties where they had in previous months left, provided all but one of the counties that seemed to have no clear options in the Obamacare exchanges–and that now-President Trump’s declarations of Obamacare’s demise were quite premature.  Although the graphics of health insurance providers that were available to residents in local exchanges under the Affordable Care Act seemed truly badly served in much of the nation by early 2017–when many of the counties not on the coasts or in coastal states seemed to suffer from a gap in options, as was true even shortly before the 2016 Presidential election in surprisingly effective ways.

As soon as the future markets for insurance were mapped and the maps were released, the revelation of apparent gaps and “dwindling in surname choices” and egregious absences in covering the nation’s populations seemed to show up the falsity of past promises.  The maps gained a polemic authority of their own, confirming lingering suspicions about the poor fit of “Obamacare” to the nation, and providing fodder for raising alarms about the inequitable nature of the exchanges that emerged in different states and counties.  For they seemed to reveal an apparent abandonment of the majority of the country by the coastal elites of California Massachusetts, New York, Chicago and Washington D.C.:  indeed, it triggered a sense of the abandonment of the nation by coastal elites.  The very story that was told about Hillary Clinton and the Democrats in many circles were being repeated in the Presidential campaign were used to lace increasing suspicion about the emergence of fair marketplaces in future years.

 

Graphic-of-Insurance-Providers-number.jpg

County-Level Data on Insurance Providers under ACA/AP  (October, 2016)

 

The deep brown nature of the map didn’t reveal the restrictive choices of insurers, but muddied the picture of the nation, as if throwing into relief a plurality of counties that existed on a higher plateaux of health care, and left behind the rest.  The visualization suggests that a sombre picture of health-care had emerged before the Senate failed to dismantle Obamacare, light tans suggesting the greatest gaps in low-lying lands of few insurance options, and markets where sufficient variability would not bring lower costs.  In those regions, the widespread “lack of choice” appeared so evident in dismaying gaps across the nation, where the departure of insurers from a market seemed that had been seen to rise in 2015 and 2016 had started to fall precipitously, raising the fears of rising premiums.  Several entire states–deep red states, as it happens, like Alabama, South Carolina, Oklahoma, and Wyoming–had only one participating insurer across the entire state, suggesting gaps in the health plan that claimed to be engineered to offer choice.

The mapping of these projections seemed to make manifest the deepest fears of inequality and an unloved playing field, which data visualizations like the above seemed to expose–while dying the projected nature of its claim that insurers’ “departure” had already occurred.  Yet the residents of all counties in the United States but one will be able to purchase an Obamacare plan in the coming year in actuality–the fifth year and enrollment cycle of Obamacare insurance markets, has brought successful expansion, with increasing coverage is provided by insurers across the United States may be even effect a new acceptance of health care as a right.  For despite widely stoked fears of drying out health insurance markets projected in deceptive data visualizations with such particularly alarming effects.

Just a mere two months ago, it was predicted that forty-seven counties would have absolutely no insurers by 2018, and that residents of greater than a thousand counties–and up to 1,200– would be left “bare”–hight and dry–with access to but one insurer in their exchanges.  The alleged analysis of the impending “collapse” of insurer participation nationwide showed an image of “projected insurer participation” as if they described an actuality of declining participation that had effectively fractured the nation–lending currency to pronouncements that struck fear into many voters.

 

County by County analysis.pngExchnage Carrier.pngCNBC, June 13, 2017

 

Despite the manipulative nature of these data visualizations, the recent resilience of markets after election, and specifically the failure to repeal Obamacare, has responded in ways that stand to change.

 

2.  But the picture was indelible when it was framed, forged in the sharp colors of data visualizations which arrived with regularity at the same time as maps of projections of the Presidential election dominated social media and the press.  They created a terrifying image of a divided nation, destroyed by the all but inevitable impending “collapse of Obamacare ‘coverage’ in 2017” as revealed in “stunning maps” released in the late summer during the Presidential election, as if they were the hidden understory of national divisions that some candidates just didn’t get.  These visualizations allegedly revealed divisions of the nation in ways that must have spooked many, weren’t being addressed by the White House or health care officials, and seemed to signal an era’s end–touting “Higher Costs and Fewer Choices for Obamacare Customers in 2017.”  The below-the-radar war of data visualization for national attention suggested nothing less than the erosion of the union that was tied to the encouragement of insurance exchanges.

Such data visualization worked their magic, triggering narratives of abandonment and appearing to reveal an isolation of several of the poorer parts of the nation that set of alarms about the increased division of a nation and an uneven playing field that the Affordable Care Act–now demonized by the name “Obamacare” to distance itself from the actual legislation–that revealed the apparent absence of competition in “stunning” ways.  For by depicting the “epic collapse” of a system that in fact seemed to be give greater stability to a projection and make it manifest as reality.  The magic of the data visualization was that they purported to reveal an actuality the Obama administration seemed to deny as if it were an actuality that denied options to many Americans.   And although the spread of the one-carrier-ounties across much of the “heartland” seemed confirmation, for many Republicans, of an abandonment of the mythic heartland of Trump voters, which pulled from Iowa to New Mexico to West Virginia for Appalachia for Trump–the complexion of where insurance is available.

The alleged objectivity of the visualization left many with breath held, as “stunning maps” released during the heat of the Presidential election in late summer prophesied an impending “collapse of Obamacare ‘coverage’ in 2017” as an all but inevitable reality.

 

Obamacare 2016_0McKinsey Center for U.S. Health System Reform

 

Data visualizations of jarring color selections suggest the discontinuity in a system of health care, using the not necessarily clear metric of the existence of a range of carriers.  The notion of the medical marketplace that such competition was supposed to create however realized clear gaps with the counties in violet, whose disarming continuities suggested pockets of the nation that were unfairly left behind, and others in pink that seemed to be similarly compromised in the notion of options or choice their inhabitants were offered.  But the alarmist cartographies were extremely effective in tellign of a story of those regions that were left out–not only Kansas, but Wyoming, West Virginia, and stretches of North Carolina, South Dakota  and Michigan. The maps spoke to many.

The deeper debate about health care as a right demand to be examined in far greater detail than the polemic nature of such visualizations allowed.  And the recent resurgence of insurers in almost all counties of the nation provides a good occasion to do so.   It’s not a secret that the difficulty of construing health-care as a democratic right has also been rejected by many,–who would  restrict liberties to participation a marketplace.  In doing so, they adopt the market as a primary metaphor if not end of civil society–and view any tampering with the health care market as undue governmental meddling.  Yet the guarantees of well-being and a level playing field seem as good a definition of what is democratic as any, as it affords a needed means to allow people to have greater say in policies and decisions that most affect their lives.  And sentiments for including health care as a right has however come to be one of the deeper divides in the nation.  And the recent success of mapping the real resurgence of insurers’ involvement in many exchanges in counties nationwide reminds us of–and asks us to reconsider–the deceptive nature of their claims about the narrative of the impending collapse of Obamacare that many data visualizations of the nation relentlessly advanced, with minimal questioning or interrogation, and the how the image of the nation they suggest may explain public understandings of health care as a democratic right.

But since the negative benefits to all of health-care being a restrictive good are pretty clearly evident, healthcare should not be seen as a commodity alone, existing on an open marketplace alone.  Given the clear negative pressures of lowering access to health care to society, the gleeful prediction by President Trump that Obamacare–as Trump calls the Affordable Care Act (ACA), as if it were just not American to promise health care to all–would be implode because of e lack of plans available on exchanges in much of the country thankfully seems untrue.  Indeed, the failure to repeal the ACA by the United States Senate–a failure that seems to have sent a shudder of poor guidance and convulsions within the Trump administration, and within Donald J. Trump’s sense of his hold on the Presidency, has led insurers to <em>return</em> to the many counties where they had in previous months left, provided all but one of the counties that seemed to have no clear options in the Obamacare exchanges have gained them.

 

3.  Trump’s declarations of Obamacare’s demise were indeed quite premature.  Although the graphics of health insurance providers that were available to residents in local exchanges under the Affordable Care Act seemed truly badly served in much of the nation by early 2017–when many of the counties not on the coasts or in coastal states seemed to suffer from a gap in available options, as was true even shortly before the 2016 Presidential election in surprisingly effective ways.  The sentiment of a curtailing of options–and of choice, that elusive and so malleable term–became something of a battle-cry against the ACA, which redefined how it was portrayed and cast as an imposition that failed to meet health needs, but whose premiums were substantially more.

If one might say, with poet Elizabeth Bishop, that “more delicate than the historians’ are the mapmaker’s colors,” unlike the color-choices by which cartographers define land and sea as areas viewers can inhabit and read, the stark colors of the data visualization suggest clearcut differences and decisive results–too often just to stark to be lent the credibility that they seek.  Bishop linked the art and science of the cartographer and the art of poetry, in her first published poem, written when staring at a framed map of the North Atlantic that lay under glass as she was ill.  In tracing the mapped waters, and the land that lies beneath the water in maps, shadowed in green, she admired the transformative nature of the cartographer’s art and the expressive license of defining land and sea, and the edges of sandy shelves, as allowing the cartographer to create an aesthetic object able to engage the viewer’s fantasy, through the delicacy of color choices:  the stark, flat tones of the above data visualization–whose colors are all too strict and edges overly severe–work best to create oppositions and manufacture absolutes that offer little distance on the world, or clear purchase on it.

The colors chosen by the cartographer, if at odds with the actuality of the ordering of the land, cannot compare to how the translation of the edges of insured and uninsured are erased in the clear contrasts that compress the actual contours of health care.  If Bishop contrasted the reality claims of the historian to the artifice of the map-maker, whose creations appear arbitrary, but reveal actual complexities, as allowing possibilities for the contemplation of the world.  But rather than presenting an authoritative version of the world, the human measure of a carefully made map, and the invention it offered as an angle at which to examine the world absent from many visualizations, which privilege a single actuality as sufficiently authoritative to orient viewers to the world along a single narrative–and not preserving a human scale to do so.  The deceptive nature with which data visualizations foretold collapsing insurance choices in the Affordable Care Act presented a false reality. about health insurance exchanges, in short, by creating alarming contrasts between sharp colors in maps that offered no opening for interpretation. In contrast, maps of the actual numbers of those without health insurance reveals a landscape of much more complex edges and shadows, as well as deep divides, demanding to be moused over in detail for their interactive experience, if only to come to terms with the changed life experiences of those in many states, as from 2013-16, as the constantly shrinking number of uninsured grew nation wide in ways that attest to the increasing health of the nation–if with considerable numbers of uninsured remaining in may exchanges:

 

shrinkin uninsured.png

 

percentage uninsured.png

New York Times

 

In contrast, the almost uninhabited landscapes bereft of insurers that data visualizations depict to suggest a narrowing marketplaces and dwindling options of Americans offers an image less about “health”–our about our health as a nation–than the problems of creating continuity among the insurance exchanges that underwrite the insurance marketplace.  The lack of perspective that they offer on the residents of each county and of our country–and the forced viewing of “health” in terms of insurance companies which participate in exchanges, suggest what more contemporary poet, Claudia Rankine, called the particularly contentious meaning of “health” today in the United States:  at a time when “Affordable Care Act” is seen as something different from “Obamacare” by most Americans, who want the affordability of health care but suspect the inequality of “Obamacare”–whose repeal Trump declared his first order of business as President.  “We heard health care and we thought public option/we thought reaching across the street across the lines,/ across the aisle was the manifestation of not a red state/ not a blue state but these united states we thought,” Rankine wrote with assurance of a new landscape of health insurance, “we could be sure of ourselves in this one way sure/of our human element our basic decency.”

But the increased decency of providing more Americans with adequate health care, “a kind of human kind of union we were ready to check-up,” as Rankine wrote, in the belief that “in this one way we were ready/to care for each other we were ready to see/our range of possibilities as a precious commodity,” was distorted in a map that focussed on the marketplaces of insurance options that Obamacare–the Affordable Care Act–sought to create.  If in this nation “despite being founded on genocide and sustained by slavery/in God’s country we thought we were ready/to see sanity inside the humanity,” the humanity of health care seems sadly obscured in the exclusive focus of data visualizations that focus on providers absent from the marketplace.

 

4.  It is rather terrifying that the alleged objectivity and authority of such data visualizations were arrogated to make a point that disguised their nature as projections and roles as arguments.  While doing so is tantamount to disinformation, claiming predictive value as declarative statements which has since proved to be without any merit.  For not only did they distort the question of coverage by ignoring that the areas where three or more carriers would be options were most populated–where the best job had been done informing patients of their options to enroll in policies, and also where far better medical coverage existed for Americans in previous years–but the alarms that they sounded were ungrounded, although the image of two coasts and a well-off midwest that suddenly left large parts of the nation in the lurch effectively tapped into deep suspicions and uncertainty.

Rankine persuasively hypothesized–and elsewhere actively protested–the deeply ingrained racism that motivated a nation ready to distinguish between “Obamacare” and the “Affordable Care Act”–valuing the affordability of health insurance, but suspicious of the insurance labeled by the name of Trump’s predecessor.  The motivations for suspicions about “Obamacare” as a tampering with the free market of health providers is unclear, but it undermines the interest in our understanding of the preciousness of health care as a right.  Yet the humanity of health care seems sadly forgotten by the shift from a topography of individuals insured to a topography of the marketplace.  Although Trump seemed to think he had fired Obamacare from the country by declaring it “dead,” and just destined to implode, the markets revealed themselves to have been set up with considerable resilience, despite deeply troubling glitches in its roll out; if more than eighty counties earlier risked offering no options to enrollees, insurers returned overwhelmingly, where they were able, especially when already strongly present in the marketplaces, despite the threat from President Trump to pull federal subsidies.

The presence of mapping future markets for insurance were released with claims to show of apparent gaps, “dwindling insurance choices,” and egregious absences in covering the nation’s populations.  They seemed to show up the past promises of the President to preserve choices for Americans to adopt a health plan that suit them best, and portray them as undue impositions on the marketplace.  The projections acquired a polemic authority, as if confirming lingering suspicions about the poor fit of “Obamacare” to the nation, by providing fodder for raising alarms about the inequitable nature of exchanges in different counties and even in different states.  They seemed to confirm a feared narrative of the abandonment of the much of the country:  indeed, many popular data visualizations triggered a sense of the abandonment of the nation by coastal elites in New York, California, and Washington DC, in particular in Arizona, Oklahoma, Texas, Colorado, Missouri and the Deep South, as well as parts of Michigan.  They confirmed the very story told about Hillary Clinton and the Democrats repeated in the Presidential campaign as if to lace suspicion about the emergence of fair marketplaces in future years.

 

Graphic-of-Insurance-Providers-number

County-Level Data on Insurance Providers under ACA/AP  (October, 2016)

The particularly grim picture that they offered came in for little criticism or rebuttal.  But the data visualizations describe landscapes that are curiously depopulated, even as they present a sobering picture to suggest the withdrawal of insurers from medical exchanges.   The map implies an absence of interest in much of America by the very insurers who claimed to have sponsored the new marketplaces–but had only concentrated on the most profitable regions.  Its implications one of the abandonment of many of the rural areas of the country–the less densely populated–although the greatest success of such exchanges in densely populated urban areas that were liberal-leaning is no secret, they imply an absence of interest in less populated areas of the nation.  The implicit message that little attention was paid to the rural areas was underlined through the strategic colors of the data visualization, which seems to suggest a relief map of areas that would suffer higher premiums:   audiences in much of the country were convinced that they just had it worse in the projections all too often portrayed as eventualities that the nation would stand to suffer.  The tan colors that suggest diminished choices of medical insurance muddied the picture of the nation, throwing into relief a plurality of counties that existed on a higher plateaux of health care, and left behind the rest.

The visualization suggests that a sombre picture of health-care had emerged before the Senate failed to dismantle Obamacare, light tans suggesting the greatest gaps in low-lying lands of few insurance options, and markets where sufficient variability would not bring lower costs.  In those regions, the widespread “lack of choice” that appeared evident in dismaying gaps across the nation, where the departure of insurers from a market seemed that had been seen to rise in 2015 and 2016 had started to fall precipitously, raising the fears of rising premiums.  Several entire states–deep red states, as it happens, like Alabama, South Carolina, Oklahoma, and Wyoming–had only one participating insurer across the entire state, suggesting gaps in the health plan that claimed to be engineered to offer choice.

Such premature projections seemed to make manifest the deepest fears of inequality and an unloved playing field, and invested themselves with an objective authority of exposing an uneven system that was indeed rigged–denying the projected nature of its claim that insurers’ “departure” had already occurred.  Yet residents of all counties in the United States but one will in fact be able to purchase an Obamacare plan in the coming year in actuality–the fifth year and enrollment cycle of Obamacare insurance markets, has brought successful expansion, with increasing coverage is provided by insurers across the United States may be even effect acceptance of health care as a right.  For despite widely stoked fears of drying out health insurance markets projected in deceptive data visualizations with such particularly alarming effects.

 

2.  To better grapple with the readiness of insurers to fill the health-care marketplace, the stunning maps of the presence of insurers who have made health-care policies available demands to be examined through a red state-blue state optic.  For eve if the aversion of Republican-leaning regions in the United States to providing alternatives to health care insurers creates a deep divide concentrated in much of the south, prairie, and southwest, where only 1-2 insurers exist, and despite uneven nature of conditions conducive to access to health services that guarantee well-being–and presumably happiness–the markets have grown.

 

HEalth Care Insureres:Red v Blue Govs.png

 

 

If the divide looks harsh anyway for many rural areas, the red/blue divides cannot reflect the actual availability of health providers to Americans.  Since the notion of the division of the distribution of insurance markets by counties seemed suspect anyway, given the sparser population of many of these states, a more accurate picture of national coverage is offered by a simple proportional warping of the odd division of the electorate by the “county”–an outdated geographical unit if there ever was one.  The mapped that warped counties by their relative populations reveals  an even sharper picture of the actually improving state of availability of insurers–the fewer residents of many of the just-one-insurer regions of the south and indeed midwest shrink, to confirm the growing success of the selection of insurance providers by the ACA, despite some obviously problematic and important to address gaps in coverage.

 

Warped Map on Insurers Red v Blue Goves.png

 

It is striking that these very gaps mirror with a terrifyingly clear correlation both dial-up speeds and broadband technology, as well as intractable bottom-line problems like gaps in the availability of health-care services in rural areas.

 

FQHCs.png

 

federally-qualified-health-centers

Rural Health Information Hub, December 2016

 

 

5.  Abandoning the red/blue divisions, which are taken by the elected governor, we see an even healthier division of the nation, at least in terms of the regions that provide their residents with something like a reasonable variety of possible health care insurers, with large areas of the most populated areas having three potential insurers, rather than insurance markets imploding at all, despite the clear gaps that it reveals in what seem more underpopulated areas–and, quite strikingly, the absence of broadband that would make it easier to enroll for insurance online.

 

5995cec01400001f002c3494Harold Pollack and Todd Stubble

 

Broadband 2014.pngNational Broadband Map, June 2014 (not updated since)

 

These gaps reveal a division of much of America into two regions–no doubt conducive between two expectations of health care or medical provision.  Most southern states indeed had far fewer insurers–left “bare” with but one provider, despite the low populations of such rural regions being just less conducive to insurance markets, and revealing an uneven playing field long preceding the passage of the Affordable Care Act–

 

SOUTHERN states health care insurers 2017-18Harold Pollack/HealthInsurance.org

 

One Provider South Rural.pngHarold Pollack and Todd Schuble

 

The area is not only medically underserved, but suggests a “Southern Problem” having far less to do with Obamacare than with the disproportionate topography of medicine and indeed of those without health-care, but creating many counties including large stretches of chronic undeserved populations.

 

RAC 2014

 

 

Such maps and data visualizations only suggest a need to appreciate and fathom the deeply compromised liberties in areas with few health insurance providers, where insurers haven’t reached clear markets, that not only overlap with many of the more chronically uninsured areas and populations, but with areas of the a terrifying number of uninsured–folks who have decided or been forced to do without health insurance, and where going without health insurance becomes an accepted acceptable alternative, unlike in many regions of the country,

 

 

us-health-insurance-coverage-map

 

which often echoed the very regions of greatest vulnerability in the nation–counties that to be sure often reached out to Trump as a savior for their deep discontent.

 

Rural Assistance Center Underserved.png

 

RAC 2014.png

 

 

2.  The increasing variety of insurance options for much of the nation raises questions about the persistence of a deep inequality–undemocratic for many–in those pockets coinciding with denser votes for Trump, in a normalized choropleth, and more hospitable to an argument of revising current options of health care–and viewing the Affordable Care Act as an imposition of the federal government.

 

Trump votes normalized choropleth

 

There are interesting overlaps on those areas where Trump out-performed previous Republican candidates, notably in Florida’s panhandle and less densely populous counties in the deep south; southern Texas; and Appalachia.

 

trump-increases-republican-votes

 

 

The odd reluctance of these areas to attract anything like a range of possible insurers in lower income areas of low-density where Republicans have recently performed well.

 

counties-1

New York Times

 

trumpland-2.png

They reflect the difficult problem of distinct notions of liberty and rights in the country, corresponding to areas where the civilian population was long underinsured, often by upwards of 15%, and where shortages of health-providers–even if not as readily available in county-level data–are strikingly revealed in a state-by-state survey:  states like Georgia, Mississippi, Arkansas, New Mexico, Nevada and Texas remained significantly below the national average for primary care physicians in 2012, suggesting regions where there were less developed expectations for attaining health care coverage.

 

maptitude-healthcare-providers-by-state-map.jpg

 

 

Increased numbers of uninsured exist in many of the same states are, to recap, unsurprisingly located in some of the same regions–which are less likely to vote for representatives who advocate the belief in health care as a right, and perhaps seeing it as able to be outweighed in importance by an argument of states’ “rights,” even if this discourse is designed to deny health insurance.

 

us-health-insurance-coverage-map.jpg

 

It seems a cruel irony in an era of globalization that the majority of those doctors or members of the healthcare force serving areas of the United States that were most in need, and who see some 14 millions patients every year, were from Iran, Libya, Somalia, Sudan, Syria, or Yemen–citizens of countries included in Trump’s Muslim Ban.  (In other words, these immigrant doctors are filling the increasingly pronounced gap that exists among medical providers in the United States–and getting the job done.)  If medically underserved areas occur in almost every state in the country, the preponderance of medically underserved populations concentrated in less populated areas–as the southwest, southern states, and parts of the Midwest seems to have attracted foreign doctors–and had already led bills to be sponsored to allow Medicare to reimburse pharmacists directly in those communities, to acknowledge the absence of medical services needed by Medicare beneficiaries, to allow clinical pharmacists to work in medical care settings as a health provider.

 

Doctors from coutnries in Muslim Ban.pngSee interactive version of this map here, at the Immigrant Doctors Project

 

The map has some striking overlaps with those regions of rural America that are losing population, although it should be kept tin mind that the above map, which used data from Doximity to suggest the commenting zones for the number of doctors in the United States may distort by expanding the zones of providing services beyond that which physicians actually serve most actively:

:

map-loss.pngRural Communities Losing Populations, United States of America

 

The elevation of the pharmacist to a medical provider may raise ethical questions.  But the existence of hight concentrations of medical physicians from the very countries that were targeted by the so-called “Muslim Ban” that Trump championed had the effect of allowing a crucial degree of medical assistance to reach Americans–although the apparent intent of Trump’s legislation would have been to restrict their abilities to return home freely to visit their families, and compromise the proportion of doctors on call in the cities where they are most concentrated–in Toledo, Cleveland, and Dayton, Ohio as well as Detroit MI.

 

Medical Assistance.pngImmigrant Doctors Project

 

Such pronounced concentrations of physicians which were mapped online in readily seaarchable formats by the Immigrant Doctors Project provide powerful tools to view how the markets for physicians’ skills meets the needs of a marketplace, to be sure, if one recalls the huge numbers of medically underserved counties.  But this is not a marketplace that would be easily filled by our current medical system, or the health-care industries that service more rural or poorer areas.

 

RAC 2014.png

 

Leave a comment

Filed under Affordable Care Act, data visualization, health care, health insurance, Obamacare

Alternative Metrics of America’s Divided Economies #3: Patterns of Drug Addiction and Hidden Pain

In place of topographic local detail, a pastel-hued tiles of psychopharmaceutical pills reveal the extent to which antidepressants have swamped the inhabitants of the United States.  The psychopharmaceutical Pharmakon has provided a new figure of reliance of particular resilience:  Stanford Kay’s graphic registers a flattening of psychic territory, as much as the dramatic increase in the widespread availability of antidepressants across the country.  In this post, I want to drill down into the local topography of dependency, and the reasons for such a broad availability of addictive anti-anxiety medications, in order to create a more detailed–and perhaps just as arresting–image of the circulation of psychopharmaceuticals, and of why we should care.

The distribution of opioids across the United States had a startling concentration in specific areas identified with underemployment, focussed as their spread was in states like West Virginia. The decisive concentration of opioid use in rural areas–seemingly surprisingly–was a decisive strategy to gain currency for the bottom-line drugs of coping, as the multiplication of prescription drugs to deal with pain, from oxocodone to hydrocodone, and by 2006 there had been a pronounced rise in the deaths from opioids in rural communities–Kentucky, West Virginia, and Virginia–that switched the geography of overdoses from urban centers to the bleeding periphery–death rates, reveal newly released government documents, led to an entirely unbalanced share of the 76 billion hydrocodone and oxycodone pills in rural communities.

We can map the disproportionate flooding of these regions that boosted the death rate in these regions by eightfold the national average of 4.6 per thousand, even if those regions were relatively far less inhabited–

Opioid Deaths per 1,000 Inhabitants, 2006/Data from DEA and CDC/WAPO analysis and graphic

–and created a terrifying concentration of deaths by 2006 focussed in rural areas, where companies like SpecGx, a subsidiary of Mallinckrodt, Actavis Pharma, and Par Pharmaceutical, a subsidiary of Endo Pharmaceuticals, created drugs that were so terrifyingly concentrated. By 2006, deaths by overdose were focussed in rural areas, where companies disproportionately concentrated distribution. From 2006-12, the cumulative deaths as a result from Opioids had created a rural epidemic, creating an epidemic like distribution of high rates of death in disadvantaged areas of underemployment, at the same time as the high volume of pills traded on the prescription drug market led to an increase of companies value of 51%, from 8.4 billion in 2006 to 12.6 billion in 2012.

The stealth distribution of opioids with willing knowledge of pharmaceutical corporations was a stealth attack on the nation, from which we may not recover–exploiting the highly addictive and available availability of drugs in an actual belt across the nation from West Virginia to Oklahoma to Utah to Nevada to Northern California, across ninety counties where the death rate from opioids continued to be 4.5 times the national average, often corresponding to areas that were poorly served by doctors, and including some counties where the death rate for opioid overdoses grew to eighteen times the national rate–and opioids were actively shipped to provide over three hundred and sixty pills per person in just a single year.

The “overdose belt” that pharmaceutical corporations created had by 2012 led to an epidemic of 76 billion pills being shipped across the nation’s rural regions–places like Kentucky, South Carolina, Tennessee, West Virginia and northern California–where they found large audiences of consumers, who contributed to the consequent rise of uses of heroin and synthetic opioids as fentanyl, as prescription painkillers became less available–hitting areas that had few resources or abilities to contain or act against an unexpected rise of drug overdose deaths that had radically changed the landscape of drug use in America: the flood of addictive drugs had created a new landscape of rural America, increasingly dependent on drugs to assuage difficulties of underemployment that were unable to be otherwise resolved, and all too easy to medicalize in a regime of pain-killers and anti-depressants, without any attention to their increasingly addictive natures: the country became the region of drug addiction, and felt increasingly removed and disconnected to any sense of economic or other benefits, and all too ready to recognize deep difficulties in the status quo, even if they didn’t totally understand or grasp what across the American interior found six companies dominating the distribution of drugs–large starts like McKesson Corp., Walgreens, Cardinal Health, AmerisourceBergen, CVS and Walmart–according to a federal database of all pain pills sold in the United States–who provided legal paths to “pain pills” that fed the opioid epidemic. The concentration of pills/person in a national map from 2006-12 cannot be considered an actual response to therapeutic need–but an orchestrated attack on certain areas of the nation–

Washington Post/DEA data 2006-12

–tat ratcfhed up in intensity, as in a controlled experiment conducted on the nation, by 2012–

–and resulted in something like a flash-flood of pain killers that circulated in the bloodstreams of many unsuspecting who died of opioid deaths, which unsurprisingly reflect an eerily identical digtribution.

In rendering a land whose bounds and contours have become increasingly difficult to navigate, awash with psychic stimulants designed to depress the Central Nervous System, Kay uses the prettiness of pills to paint a sad picture of the spread of increased drug dependency.  For more than anything else, the massive deregulation of drugs and psychopharmaceuticals from the 1990s has led market forces to shape the availability and dosages of medical regimens to regulate nervous chemistry.  Indeed, the shift to market-based models of caring for oneself reflect a triumph of free market forces not only over individual well-being, but the medical pharmakon.

The expansive prescription of psychopharmaceutical drugs maps onto shifting notions of leisure time.  Before his death, the Australian sinologist Simon Leys found it an “ironical paradox of our age” that most classes devote little time to leisure who, as most economic elites, have willingly accepted the “slavery of endless working hours,” and the category has been consigned to the “enforced leisure of demoralizing and permanent unemployment.”  With the conversion of the professional classes into “senseless” money-making machines, Leys worried, leisure has become the destiny of a lumpenproletariat; if Leys bemoaned the tragic loss of a historically cultivated ideal without reference to America, he described decisively shifted in attitudes to work and time he described parallel the deregulation of markets that go under the expansion of free markets in globalization, and the increasingly forced ties between choice and consumption:  we do not choose leisure activities, but consume them; leisure, separated from work, is primarily oriented toward consumption.

Pierre Ryckmans, who wrote under the name Simon Leys, might well have linked globalization to how leisure had become increasingly relegated to the out of work–an unstructured “leisure” time haunted by being unemployed against one’s will, rather than linked to questions of individual choice.  The expansion of time without work has led leisure to fill time in new ways.  Although Ryckmans made no specific reference to America, but offered an abbreviated genealogy lamented the decline in the reconsideration of ‘leisure’ as removed from either gainful activity or worthy pursuit.  To recover a nobler lineage of leisure, Leys rehearsed the long survival of the concept from Chinese literati he studied–“The leisure from learning should be devoted to politics and the leisure from politics should be devoted to learning,” Confucius counseled–through ancient Greeks’ value of scholê as time dedicated to thought, to Nietzsche’s dismay at the “erosion of leisure” he feared an American cultural infection which had communicated the same relentless compulsion to work to Europe.

Much as Nietzsche worried about the infection of a compulsion to work from America that threatened to stigmatize the vita contemplativa, that bedrock refuge of philosophy, by casting it as an altogether shameful fact, Simon Leys sensed the enforced leisure of the underemployed as a deep change, melancholically regarding Don Quixote as the embodiment of leisure past, albeit beset by the combination of too much leisure and few funds that lead him to squander all economic possessions.  The taxing and rather toxic task of filling time removed from reading or reflection seems to define the enforced leisure of underemployment as a terminal condition rather than a choice, in ways that signal either a deep failure of the collective imagination in the valuing of work as a productive activity or conflating choice and consumption–for leisure is essentially defined as a choice in habits of consumption, rather than dedicated to as a pursuit or lifestyle.

This emphasis on work as the sole activity of productivity places such undue stress this places on the individual who, out of work, faces the existential quagmire of the value of leisure time.  The flattening of the topography of America, pictured in the header of this post, as flooded by a free market of prescribed drugs reveals not only a change in the sense of the individual, but the emphasis on the consumption of prescriptions as a form of life.  The recent medicalization of individual psychic and physical experience of pain, itself a correlative of the free-market of pharmaceutical cures, has increased the power of prescription drugs far beyond their earlier clientele.  Embodied in the promise of Big Pharma to provide drugs to reduce “moderate” pain through prescribing opioids, the expansion of prescription drugs has more than anything else encourage the four-fold expansion of opioid drug overdoses across the United States in the past two decades.

Suffering from pain was long tied to work, and accidents of work–for which opioids as Oxycontin were first prescribed, as well as being designed for cancer relief.  The redefinition of “pain” less as an experience but as a topic and subject for medical attention introduced opioids from Percocet to Oxycontin to roadside retail pharmacies  as a condition that is able to be treated in the growth of overdoses of prescribed opioids since 1999, and the tragically decisive distribution of drug poisonings by opioid analgesics in 2006.

Poisonings involving opioid analgesics 2006

CDC

In an odd corollary of the transformation of work, the medicalization of the depression and desperation of the out of work, as a members of a market for the latest innovations of Big Pharma creates a new topography of drug dependence that unsurprisingly partly mirrors that of the out of work.  For as unemployment intersects with the pain of depression to be best alleviated by a virtual pharmakon of prescription drugs, one can see an uptick in the dependency on opioids.  Such a spread does not only follow a tendency of addiction, but the new availability of potent prescriptions of opioid analgesics from the mid-1990s responded to changes in the labor market across the United States–and if the steady in opioid overdoses is statistically but a subset of drug overdoses, it demands greater examination as a geographic distribution than has been the case.

drug overdose deaths

CDC

For even though we have amazingly accurate data on addiction and opioid overdoses for the last twenty years, the actual distribution of both the availability and consumption of prescription drugs reveals deep geographical specificities in relation to concepts of work has not been fully appreciated or examined.

The topography of prescription drug overdoses using painkillers is far more uneven than it is uniform, and the four-fold increase in the distribution of deaths from opioid overdose reveals quite deeply weighted spatial concentrations.  Even after the relatively recent celebrated crackdown on pill mills and pain clinics in America, the culture of pain treatment based on the dissemination of opioid drugs to retail pharmacies across the nation has left an aftertaste of the medicalization the condition of continued underemployment.  The basis for the marketing of painkillers is deeply tied to the growth of dependence on the illegal trade of analgesic opiates, as desensitization masks as a vaguely medical response.  The deepest failure seems to have been the foregoing of pain as an individual project, perhaps, and the topographies of an easy cure that intersected in tragic ways with the decline of the economy after the Great Recession.  For the intensive promotion and marketing of opioids as a highly prescribed drug with little need of oversight or monitoring from 1996, the year OxyContin entered the landscape of public health, in a roll-out that led to almost $1.1 billion in sales by 2000, proved so intensely popular that by the year 2004, or from the Recession of 2001 to the Great Recession, despite a wide array of illicit recreational drugs, it became the number one form of drug abuse across the United States, reaching new markets for drug abuse that had rarely existed, and in so doing utterly changed the national landscape of public health.

1.  It is tempting to try to map what the range of habits of work and nature of leisure time across urban and rural America.  Although the construction of such a relational database of leisure–and ideas of it–are compelling to map, the category of enforced leisure resists clear metrics, and is not clearly present in any single relational database.  Yet what Simon Leys described as something akin to a collective devaluation of leisure as a pursuit has its underside in the expansion of pain–and the difficulty to manage the pain that comes from work, or from lack thereof.  The drive to reduce “pain” precipitated by anxiety, stress, and depression–and the uneasy relation of pain as a category of attention bridging psychic and bodily pain–has helped promote the popularity of prescription painkillers as a tool to change pain perception and relieve both moderate and severe pain alike.

The demand to do so, and to remedy the spectrum of sense-based pain and anguish, or sensory and affective pain, provided a basis for pharmaceutical companies to promote OxyContin and other opioids to all willing to listen to the possibilities for easing suffering from the mid-1990s.  The recent maps of a hidden topography of pain that the geographic distribution of the prescriptions of painkillers and the addiction to such synthesized opioids that has come to plague much of the country–manifested in the deadly effects of addiction to painkillers across much of rural America–maps onto a massive shift not only in ways of spending time but of the expansion of those out-of-work and indeed the pain of those underemployed, whose result may be linked to the terribly shocking visualization prepared by the CDC of the topography of drug-poisoning across much of rural America.

drug poisoning rates

What sort of topography both of pain and of drug abuse does this data visualization reveal?

If the rise of drug poisoning in much of America seems clearly tied to the flooding of much of America with easily procurable and widely prescribed painkillers, promoted across America since the mid-1990s, the morass of the effects of the wide availability of analgesic opiates has created a crisis of public health we are only beginning to confront, as well as an economic crisis of how those afflicted with addiction to painkillers have lost control over how to manage their own time.  The plague of analgesic opiates in American that Big Pharma, eager to expand the market for painkillers, has broadly disseminated to patients, downplayed the risks of their addictiveness, and drastically underplayed its risks–or the inevitable possibilities for its abuse–in ways that constituted a true public health tragedy by promoting a regimen for diminishing pain as without risk.  The promise of reducing–or absenting oneself from pain–has created the promise of a life free from pain, removed from the subject, unlike the sort of indices that Dr. Ronald Melzack used to allow patients to map pain in the body by the pathways along with pain travelled in the Central Nervous System and “body-self neuromatrix” as measured in the McGill Pain Questionnaire, parsing sense-based (one through ten) to affective (eleven to twenty) pain.

McGill Questionnaire

Melzack’s McGill Pain Questionnaire

Unlike the “Gate Control Theory of Pain,” opioids were marketed as a means to pre-empt pain’s sensation that blurred the distinction between sense-based pain afflicted in work and the psychic affective pain that was newly burdened on the out-of-work; the prescription of painkillers designed to be prescribed for chronic pain to alleviate individual suffering blurring.  Did this quite different remapping of pain shift coincide with a shift in the landscape of public health costs?

Based on a range of new data–from rates of addiction, prescriptions of painkillers, prescription drug poisoning, overdoses of first-time users of opioids, and methadone treatment centers across the country–a topography of addiction and overdose emerges that seems to have paralleled the increased availability of “pain relief” in retail pharmacies across America.  The arrival of cheap and widely prescribed opioid drugs–opioids easily available with largely unmonitored prescriptions–have created deep social and medical costs we have yet fully to ken.  The mapping of increased drug-poisoning from overdoses matches the increased circulation and abuse of cheaper opiates in rural areas rarely affected by illicit drugs in previous years, and a rash of first-time users have all been tied to painkillers, whose specific causal relations to one another demand further analysis.  For the flooding of the market with low-cost opioid analgesics effectively promoted a self-induced regimen of desensitization that has huge costs on people’s lives and well-being, and of which many are still seeking to gain control.

Rather than pursue a model of pain management, the offering of an option of prescription-strength analgesics as a cure for pain provoked a rash of addiction that suggests a huge tragedy of public health–and reveal a geography of painkillers that cannot purely map suffering or pain.  From an apparent high in which some one in twenty adults regularly used prescription painkillers for non-medical reasons, in 2010, the epidemic of painkillers is only beginning to be controlled–although the networks or drug addiction it created have grown entrenched, as they have provided both a gateway drug to tar heroin, and increased the geography of a demand for drugs of mental desensitization that seems closely tied to the spread of the terminal out of work:  for the “pain portrait” that such a pharmakon of drugs provide is not a detailed or sharply chiseled portrait of palliative care, but marketing a promise of wiping pain out–rather than addressing suffering.

So much is evident in the new topography of drug abuse and addiction across America.  While opiates were synthesized for cancer-induced pain and acute injuries, the dramatically sudden dispensation of new levels of analgesics lacked clear overview, but were aggressively marketed for “non-malignant pain” that bridged psychic and physical pain, and its risks for addiction downplayed.  Yet an increasing range of areas of the country are currently struggling with the range of narcotics abuse, seeking to contain the costs of over-prescribed opiates from OxyContin to Vicodin, advocated as non-addictive cures for pain, Narcotics Anonymous has grown across the country that reflect an expanded topography of suffering, rivaling twelve-step self-help programs as Alcoholic Anonymous, as if to testify to a need to remedy addictions.

5f66af15e

The growth of the program intersects in interesting ways with states with the greatest prescriptions for painkillers, reflecting not only the skewed demands for prescription painkillers in America, but  topographies of addiction.  High rates of prescription opioids and self-help groups for narcotics appear particularly pronounced where poverty, unemployment and job losses are particularly acute at the start of the new millennium, which is the subject of this post.

Pinkiller prescriptions

2.  Can one use data visualizations of opioid abuse to read the costs of an absence of dedicated leisure time?  The growing demand to assuage the psychic pains effected by a lack of work almost reflect a declining ability to order time.

Simon Leys clearly voiced concerns that the “enforced leisure” of the early twenty-first century was the other side of the slavery of work–so unlike the concealment of work by leisured early modern nobility or aristocracy–and a decline of the liberal arts.  To be sure, the drastically diminished attention to the pursuit of idleness as is linked to the rise of new networks of opioid addiction that painkillers across America, so unlike the pursuit of recreational drugs among the upper or middle classes in earlier eras.  The paradoxical expansion of the use of opioid analgesics designed to reduce suffering have created new networks of addiction as pain-relievers and dependence across rural areas, tied to the past dependence on painkillers, that has necessitated new methadone treatment centers as the number of deaths from opioid abuse has decisively risen in the twenty-first century:  despite actual suffering from chronic pain, over-prescription of such easily addictive and abusable drugs have diminished well-being with rapidly expanding social costs that are difficult to  foresee, as the daily toll of death from overdoses of prescription painkillers has grown to almost forty-five in recent years.   Total fatal overdoses in such addictive prescription opioids tripled from 2000 to 2010, as drugs of such elegantly different neologistic marketing trademarks–Opana, OxyContin, Vicodin or methadone–have come to claim lives in ways never foreseen.

For medical interpretations of the range of mental trauma that afflict the underemployed and out-of-work are deeply tied, in the United States, to how diagnoses that cast depression or anxiety as a form of pain to be alleviated by prescribed pills of pain relief–opioid analgesics that are increasingly recognized as having a darkly addictive side, whose costs are difficult to map in objective fashion.  While consumed to induce desensitization or produce periods of euphoria, the rash and resale of prescription painkillers that are inhaled, injected, or smoked in different forms has created a public health epidemic of growing proportions of addiction and drug overdose that has spread to many of the nation’s rural areas–even as law enforcement has persistently focussed policing efforts on poorer urban slums.  Even if the DEA found in 2010 that ninety of the top one hundred doctors prescribing oxycodone were located in Florida, the far broader geography of overdose and death is increasingly distributed nationwide, if with a clear clustering in areas of deepest addiction in the oldest manufacturing states.  Is this a coincidence?

culture of pain?

Although the CDC and other government agencies can measure the increase in prescriptions of powerful pain-killers, the drug-induced deaths caused by such addictive narcotics, and the increase of deaths from first-time use of heroin, the dangers of opioids both to individual health and as gateway drugs create complex networks of dependence whose social costs are difficult to measure or objectively view.   The spread of such addictive opiates is fueled by increased tolerance of opiates, but also by the effective social acceptance of such self-induced desensitization by the massive marketing of painkillers as forms of relief, despite their evident costs and consequences, costs bound to increase in future years in ways difficult if not impossible to map in their complete range of consequences–and the cascading spiral of addiction and dependence on painkillers, heroin, and other high-grade opiates, which are simultaneously entering the country in increasingly cheap form.  Increasing social and health costs are in part reflected in growing archipelagos of addiction clinics promising rehab across the country, but also in the deep difficulties of addicts to regain control over their lives and their health, given the steep costs of addiction and the inroads of illicit drug markets that the ready availability of OxyContin, Percocet, Vicodin, and other opioids create–and the spread of methadone clinics across the country.

The rapid retail marketing of painkillers across America from the 1990s first fostered a dangerous of network of addiction across rural America has much to do with the absence of employment.  Recourse to painkillers and anti-anxiety medication have brought a course of self-induced desensitization to the doldrums of demoralization for those enforced to spend life in the leisure of unstructured time.  For painkillers come to constitute a poisoned remedy for the depressed, injured, and out-of-work, which encouraged or contributed to a recent epidemic of heroin in much of rural America, where painkillers were neither tightly controlled, closely supervised, or adequately overseen.  While the Belgian-Australian translator of Confucius bemoaned the metamorphosis of “leisure” to a condition of the out-of-work, growing costs of depression and anxiety reveals a spread of addiction quite unlike the market for recreational drugs, particularly prominent in those areas of the country marked by the disappearance of manufacturing jobs and repeated waves of recession, and the clearer topography created by expansion of those out-of-work in specific areas of the country, among whom the resale of relatively cheap opiates is particularly pernicious, and deeply tied to the increased dependence on what Olga Khazan has called “a cheaper, more accessible high” that sadly reflects the promotion and provision of painkillers to the same demographic in often unsupervised and unmonitored fashion.

It is tempting to look for echoes of the new economics of addiction by their eery intersection with the spread of the out-of-work during the same decade as the Great Recession:

onthemap_fig01_1_
map_6

The growing costs of the increasingly pharmacological response to anxiety, depression and physical pain is evident in the current epidemic of the devastatingly widespread abuse of over-prescribed analgesic opiates in specific, because these drugs have been unleashed in unmonitored and uncontrolled ways as a cheap high whose addictive nature encourages both tolerance and abuse.  Many abuse painkillers to seek temporary euphoria, but with the often unintended consequences creating a chemical tolerance of opioids and dependence on them that has meant that networks of addiction and illicit sales, once a phenomenon purely urban in scope, have expanded to rural areas, as have unintended overdoses.

While we cannot clearly map such networks with the requisite objectivity of a map, an archeology of the growing consequences of the dissemination of opioids across America in retail stores provides perspectives on the pathways of addiction and networks of dependence in which we are submerged–networks which data visualizations may claim they objectively represent, but which Americans are increasingly immersed in so many ways, and that a single point of view or gaze isn’t able to dominate or comprehend:  from the rise of geographic spread of drug-induced death and overdose to the dramatic geographic expansion of first-time heroin users, and the rise of rehab clinics, the social costs of such addiction can barely be measured.

3.  Or do painkillers’ wide over-prescription suggest a deadly pharmakon of cures promoted as bridging both psychic and bodily pain, but whose squelching of pain provides an entrance into quagmires far greater than we are able to map?

Whereas Plato contrasted philosophy to other occupations as providing a pharmakon of greater riches to its practitioners, the absence of work has been tied to a deadly pharmakon of painkillers in the United States.  The often largely unreflective promotion of such poisonous remedies as an alternative to the despair of those with few hopes for stable work, promoting a palliative transportation to an alternate state of mind, particularly disquieting given the deep economic disequilibrium in our society.  Despite the false dichotomies drug companies persist in duplicitously drawing between legal painkillers/illicit drugs, non-addictive/addictive drugs, prescription sales/narcotics, Oxycontin/heroin, or Big Pharma/cartels, the ballooning of prescription sales has caused each of these binaries to very quickly collapse.  Rather than map onto urban centers of reflection, or onto points of drugs’ arrival in the United States, as, say, the opium dens of leisure that quickly emerged from San Francisco to Victoria to Vancouver after the arrival of Chinese immigrants in the 1880s, in ways that created sudden panics mapped as threats to civil society, the epidemic of opiates in America is less widely mapped, or even seen as the threat to civil society that it increasingly constitutes, perhaps because it constitutes such a mirror of the steep inequalities of economic life and work in many regions of the United States.  The strong ties between addictive painkillers and new networks of purchasing and selling heroin indicated in a recent New York Times editorial, that noted the otherwise inexplicable tripling of heroin overdoses over the past three years. Opioid overdoses have claimed over 8,000 in the US in ways hard to process apart from the rise of a market for opioids larger than in either Canada or Mexico.

The chilling statistics of such offers a sort of mirror of the spread of addictive opiates that entered much of the United States as prescription drugs in retail pharmacies, and that must be clearly separated from the traffic in recreational drugs in the pathways of their diffusion and toxic combinations found in the bloodstreams of those who have died from overdoses.   The new geography of addiction is, in fact, a particularly scary mirror to hold up to ourselves or our economy.  The geographic distribution of the plague of overdoses and addiction to opiates mirrors a demoralizing and incapacitating “enforced leisure” seems bolstered by the inability of adequate resources to respond to an endemic unease and anxiety among the out-of-work.  The spread of opiates to rural areas where they are often dispensed or prescribed as a pain relief is far from akin to a leisure activity or pattern of recreation drug use, but suggests a trap of addiction that sneaks up on its users, and arrives as easily from ready-to-fill prescriptions obtained from “pain clinics” or from referrals, and the ubiquity of brand-name drugs that are heavily marketed to all Americans.  The striking demographic roots of the rise of prescription opiates across the country demand to be mapping in ways quite distinct from the pathways for the arrival and transmission of drugs into the country when we chart the increasing illegal transport of heroin to rural and non-urban areas along a network of transportation routes that run along the infrastructure of highways–

and-the-same-goes-for-heroin

Strategic Forecasting Inc.

–the routes of heroin transportation mapped above reflect the increased demand for the narcotic opioid across non-urban America.

We have often examined such a map’s primary routes of transportation by the overland transport routes of criminal cartels across the border, and overseas shipping routes from Colombia, Venezuela, or Brazil.  But the rise in demand for drugs of deeper degrees of desensitization must be more carefully compared to the rapid rise of opiates across America–as much as the permeability of our borders.

heres-a-look-at-which-cartels-tends-to-handle-which-drugs-though-the-dominant-zetas-are-conspicuously-missing-on-this-map-1

Strategic Forecasting Inc.

The sources for the relatively recent growing demand for narcotic opioids might be better grasped by the new geography of addiction and overdose that the pathways of cartels, urban gangs, or airports associated with urban clearing houses.  It is striking that unlike recreational drug use patterns of past years, addiction to medical opiates that were so widely prescribed across the United States have grown in many regions of America not previously marked by drug use–and not as leisure drugs, but as drugs foster by opiates’ deep addictiveness.  But if we all continue to focus on cartels as the source of the heroin that enters America, and especially of the stronger strains that of tar heroin particularly plaguing rural areas because of their relative cheapness and availability, we inadequately measure the demand for the greater degrees of desensitization it provides, or to adequately forecast the new patterns for its future growth.  The past decade alone has seen a 60% growth in first-time heroin users in America, that cannot be blamed only on its ease of entrance into America.  The unassuming Vermont Governor Pete Shumlin to call attention to the 770% rise of heroin use in his state since 2000 as an “epidemic” in the entire State-of-the-State address entirely dedicated to heroin addiction–and speculating on the weekly expenditure of some $2 million in his state alone on heroin, at huge local cost to state residents.  It is profoundly striking that unlike recreational drug use patterns of past years, addiction to medical opiates that were so widely prescribed across the United States have grown in many regions of America not previously marked by drug use–and not as leisure drugs, but as drugs foster by opiates’ deeply  nature.

prescrip painkillerssols

It is striking that the addictive dependency on opioid painkillers indeed expanded in the recessions of the early 2000s in unforeseen ways, coinciding with the large number of jobs shed among less-educated workers in the 1990s, before the Recession of 2001, or the job-losses suffered since–estimated from 2.4 to 2.6 million–in the very areas of the country where workers entered waves of unemployment that sent shocks through the country’s core. The pronounced patterns of job loss seem starkly evident as a new topography of the economy in the grim grey maps that offer data visualizations of job loss from the Economic Populist to the Economic Policy Institute

waves of unemployment
Jobs Displaced 2010

–where declines in the daily wages seem especially evident in those counties associated with manufacturing–deriving a quarter of employee earnings from manufacturing as of 2000, according to the 2004 ERS County Typology–based  overwhelmingly in Southeastern and Midwestern states, which have suffered marked losses of earnings:

finding_joblosses_500new

The very areas of sudden shedding of jobs during the economic downturn provide a screen on which the subaltern routes of the transportation of heroin and addiction can be projected.  The turn to opioid-based drugs might be grasped in part in the shifting economic structures of much of middle America, that seem to respond the new networks that rise of the use of prescribed opioids created–from oxycodones from OxyContin to Percocet, and Percodan, or of hydrocodone, and hyper-analgesics as Vicodin, Lorco and Nortab.  Data on the recent spread of opioid prescriptions and sales, and of the addiction to increasingly available opioids clearly coincides with the new geography of the out-of-work across America, and seems closely tied to a widespread trend to the medicalization of alleviating psychic pain.  The mapping of such prescriptions and overdoses from their abuse suggest a topography of social costs that demand to be better mapped.

The enforced leisure of long-term underemployment and ensuing anxiety dovetail eerily with the diagnoses of “moderate to severe pain” as suitable for painkillers and the marketing of opioids at most retail pharmacies–drugs of considerable potency, often earlier associated only with acute pain, and the growth of prescription and non-prescription opioid abuse, are marketed as able to remedy an expanding varieties of ‘pain’ at most retail pharmacies.  Mostly marketed under familiar brand-names, and widely available at retail pharmacies, such narcotics often distributed through prescriptions increasingly long-term, in ways that further invite abuse, as is reflected in the expansion of painkiller addiction during the last fifteen years in the United States.  The availability of opioids suggest not only an easy fix across the United States, but the increasing tolerance of opioids, and pathways of mental addiction they create, suggest a gateway drug has introduced and created a new topography of addiction that poses substantial health costs, even as the prescription of such psychopharmaceuticals was developed as a way to alleviate psychic pain.  That the Great Recession facilitated the expanded illegal and legal circulation of powerful painkillers in an unsupervised fashion cannot be denied.  The plague of unnecessary prescriptions issued with increasing abandon since the late 1990s, in ways that have created one of the worst and most serious drug epidemic in US history because of their intensely addictive character, whose topography is particularly striking, as are the rash of fatal overdoses opioid addiction has apparently caused in much of middle America.

Simon Leys’ observation may strike one as overly strident in its elitist tone, and is almost openly dismissive of the lumpenproletariat of the long-term unemployed and the mass-marketed “leisure” to which they seem consigned.  But it reflects on the debasement of leisure, rather than the class-system that it describes.  The grimness of this “enforced leisure” is reflected in the range of psychic pain being out-of-work has provoked among the long-term unemployed, lying far beyond either a Kierkegaardian angst–an inability to enjoy nature and life–or even a Knausgaardian time-stopping despair, perhaps tied to the inherited tendency to alcohol abuse.  If Karl Ove Knausgaard is known for tracing a trajectory from possible promise to an intense experience of the ineluctable passing of time in ways that the author’s mind is almost unable to process fully, lengthy meditation is less the scope of such addiction to opioids than the starving for a stronger sort of fix, sadly tied to a desperation of often undetected repercussions and consequences.

prescrip painkillerssols

The costs of such desperation and the failure to address the psychic pain it creates are increasingly tragically evident in the spread of painkillers across America.  The epidemic of addiction across much of the middle of the country–the southeast and midwest–undeniably dovetail with the effects of the Great Recession, during which access to opioids has become something of a terminal activity across areas of America that one would not characterize as likely sites of an epidemic of drug overdoses, where the promises of relieving pain have created black markets that have fueled new markets for illicit drugs able to provide a stronger high.  The dangers of painkillers are particularly pronounced not only in their sheer addictiveness, but the rising rise of overdose that they provoke in combination, known or unknown, with the anti-anxiety medications now so widely prescribed that similarly depress the Central Nervous System (CNS) of individuals and which are further depressed to dangerous levels by opioids:  of the 6.8 million Americans who filled prescriptions for opioid painkillers in America between 2009 and 2013, some 60% took them in possibly fatal combination with counter-indicated drugs that further depressed the Central Nervous System to fatal levels, inviting a new geography of drug overdose.

The geography of overdosing in America, once confined to urban areas, has so grown during the first decade of the twenty-first century to numb the mind, as many areas of middle America where such fatal overdoses were almost absent have become part of a new national landscape, in which, according to the authors of a CDC study of age-adjusted death rates related from prescription drug poisoning across America, some 90 percent of which were tied to prescription drugs, of which opioids were tied to 21% of drug poisoning deaths in 1999, but 42% by 2009–or over 15,000 individual deaths:

CIty lab Drug Deaths 1999-2000
2004-5, 8-9

The crippling county-by-county spread of deaths by overdose that was mapped by Lauren M. Rossen, Diba Khan and Margaret Warner for the CDC in a small-area distribution featured some time ago on City Lab showcase a dangerous national trend.  For the deeply addictive nature of painkillers is a potentially crippling in public health costs.  If this created a shift in policy of the Office of National Drug Council’s policy on anti-overdose drugs, New Mexico–which led in the number of overdose opioid-caused deaths–was quickest to act on revisiting its own policies.  Will the nation need to revisit its policies as a whole?

Although it is rarely linked to the economic downturn, it suggests the new pharmacopoeia of the long-term out-of-work or underemployed.   Addiction creates an alteration in brain chemistry, and the addictive nature of opioid painkillers themselves rises both on account of the notorious rise in tolerance of painkillers, and the intensity of the addictive demand for opioids.  The spread of this  poisonous pharmakon has created, moreover, an increasing demand for heroin across America among those in search for a greater fix, forging both networks of sales and pathways of popularity in regions of the country where drug abuse was rarely known in ways only beginning to be mapped, or examined as new networks of pharmaceutical dependence.

Mapping the meanings of the availability and outcomes of opioids across America is particularly important in tracking the costs of the broad dissemination of easily obtained and over-prescribed palliatives at a time when anti-anxiety drugs (designed to reduce depression) are already widely prescribed to the out-of-work, and the adoption of a regimen of pain relief increasingly undertaken, if quite misguidedly, as a viable long-term treatment.  The popularity of prescription painkillers during the last two decades arrived in retail pharmacies just before recurrent threats of recession in the early 2000s and in late 2007, coincided with an expanded pharmacopeia that came to reflect our uneasy economic divides with particularly dangerous consequences:  for not only are these heavily addictive opioids often characterized as gateway drugs for further forms of depressing the Central Nervous System, but their combination with anti-anxiety meds creates a potent cocktail of psychopharmaceuticals more apt to lead to fatal drug overdoses than most illicit drugs–and open opportunities for fatal drug overdoses across much of rural America.  The combination created a startling postcard in 2010 of drug overdoses in America that maps terrifyingly onto the greatest sites of opioid sales, which made deep inroads into the central midwestern and southeastern states that don’t seem only associated with recreational drug use.

Poisonings involving opioid analgesics 2006

Drug Poisonings involving Opioid Analgesics in America, 2006
2010 Age-Adjusted Death Rates for Drug Poisonings

drug poisoning postcard

CDC

4.  In ways that oddly parallel the shifting topography of the recent recession, the rising use of opioid analgesics, based on promises of pharmaceutical industries to reduce pain, has created an expectation of its alleviation and network of the reliance on opiate analgesics which poses huge economic and social costs, as the Venn diagram indicating the intersection between those taking prescription painkillers as well as antidepressants or anti-anxiety medications has rapidly and perilously grown with ten-fold increase of pain medications since 1990, when Big Pharma marketed a medical culture able to control pain, and to contain “moderate pain” as if it were a form of anxiety–a promise that seems more linked to marketing than diagnostic skill.  Indeed, their combination–as the combination of opiates and alcohol–has led to expanded numbers of fatal overdoses.

The rapidly altered landscape of prescribed anti-anxiety medications from 2009 is famous, if not notorious, as shifting the lay of the medical land.  And the intersection between opioids and anti-anxiety CNS depressants a train wreck waiting to occur of tremendous health care and social costs, whose costs rarely mapped.  Stanford Kay‘s map of the regional patterns of the dominant psychiatric drugs in the country in 2009 indicate the birth of a nation increasingly dependent on the over-prescription of antidepressants, ready to prescribe pills for anxiety disorders, bipolar conditions, depression, obsessive-compulsive disorders, panic or post-traumatic stress, with one pill indicating a million prescriptions, from the figurative west coast of Xanax to the east coast of Valium encompasses the zones of Prozac, Zoloft, and the plains of Ativan, it only remains to determine at what point the number of psychiatric prescriptions surpassed the number of people in America.

antidepressants in USA 2009

Most-Prescribed Psychiatric Drugs by Generic and Brand Name in America (2009); Stanford Kay/GOOD

src.adapt.960.high.oxycontin2_0830.1394646385288

OxyContin–a painkiller claiming more lives in the United States than heroin and cocaine combined by 2012/Getty Images

Despite the diffusion of pain diagnoses and prescriptions, the costs of such opiates are not that widely appreciated.  Dr. Nora Volkow of the National Institute of Drug Abuse has begun to draw public attention to these social and psychic costs; she has estimated some 2.1 million in the United States suffering from substance abuse disorders in 2012, and many die from the abuse of prescribed opioid analgesics, in ways that demand to be mapped, remapped, analyzed and unpacked.  Although but a fraction of the circa 30 million abusing opioids worldwide are in America, there are huge social costs of such poor medical oversight, as opioid analgesics flood a free market in what seems a market-driven flood.

The expansion of opioid prescriptions form the late 1990s may have helped to create the quadrupling of overdose rates seen in the US since 1999, greater health care costs in rehab and recovery clinics, and costs in crimes, and in lives, and created a new culture of the provision of medical care for pain in readily available form that was widely unmonitored and easily abused–many of which were overly-prescribed and widely diverted for “non-medical” uses as extended-release tablets are ground up, snorted, smoked or injected for recreational use.  The public health impact of opioids and heroin are seen as joined by the Office of National Drug Control Policy, and not because of differences in data, even as the number of overdose deaths from opioids has jumped over 20% during 2006-2010, when death from opioid overdoses jumped to 16,661 and overdoses from heroin declined to close to 4,000.

The inappropriate use and abuse of prescription drugs has grown dramatically, despite a recent 2012 citizens’ petition urging stricter labelling of opiates, and their designation for “severe pain,” the costs of opioid addiction and overdose have not been as closely tied–and mapped–to the despair and anxiety of rural underemployment as one might expect.  Although Physicians for a Responsible Opioid Prescribing has called attention to the failure of the FDA and US Government to monitor their use, and act to curb the epidemic of OxyContin, the economic origins for the recourse to a poisonous pharmakon of addictive painkillers have been glossed over and not been addressed.  Opioid painkillers have become the most prescribed drugs, leading to public editorials to beyond recommendations for their curbing; but the relations between high rates of prescription to the increased economic disparities of job loss are less often unexamined, however striking the distribution of prescription drugs in areas of greatest job-losses.

paink scrip rates 2012

Rates of Painkiller Prescriptions in America, 2012 (USAToday)

The relatively rapid rise of opioid addiction conceals the fact that most rural counties in middle America, so affected by addictive analgesics, still lack treatment centers for their abuse.  Their addictive properties, apparent in Oxycodone, the active ingredient in OxyContin, similar in its chemical structure to heroine, create similar depression of the central nervous system particularly perilous to patients.  Such addictive analgesics can commonly potentially create quite fatal cocktails paired with widely prescribed anti-anxiety medications that also act to depress the Central Nervous System, inviting unmonitored psychopharmaceutical experiments self-administered by unknowing patients–whose growing tolerance may encourage them to search for potent ways to diminish their pain.  (Abuses of Oxy have also been convincingly argued to create a gateway to the consumption of increasingly cheaper heroin often of far greater strength, after it was found in 2012 that abuse of heroin tar among first-time users had grown nation-wide by 60%, as users searched for a similar overwhelming of neural pathways.)

The danger of the distribution of an addiction to opioids mirrors the rise of psychopharmaceuticals for anxiety and anti depression across much of America.  Can the anxiety of economic downturn of 2006 have spurred the spread of prescription abuse, or did the consumption of opiates that are branded for regular sale–Vicodin; Percocet; etc.–and created an environment of availability of prescription medicine in the US to alleviate and reveal suffering at such significant costs?

rise in opioid prescriptions

The deep social costs of such desensitization to pain were felt not only by insurance companies–although pain relievers costs insurers companies some $72.5 billion annually in health-care costs[1], but contribute to deep health costs of the growing illegal trade in addictive opioids–whose tolerance, as their effect on the brain’s own natural opioid system grows less responsive to them over time, increases their abuse and possibilities of their overdose.[2]   The flooding of markets with such CNS depressants that flood the system with dopamine has fostered their abuse, and revealed a failure to resolve deeper problems–and has created a culture of abuse, encouraged by the increased tolerance of opioids among patients, necessitating higher dosages, often without clear monitoring or supervision to deal with chronic pain–placing them at risk of addiction and transitioning to heroin or other drugs:  the death-certificates issued for the overdose of opiate analgesics have been more common than for either heroin or cocaine.

The broad dissemination of painkillers means that members of any demographic might become an addict, and most anyone might die of a drug overdose–and indeed anyone can be subject to disrupting their own neural networks–especially when combined with antidepressants or anti-anxiety medications like Xanax and Ativan that are so widely prescribed and which slow down the Central Nervous System (CNS)–and for which most who held prescriptions for opioids also had a prescription, greatly increasing their risk of death–during 2013, some 60% of those taking opioids legally were hazardously combining them with other medications, and one third taking anti-anxiety medications, creating a climate of neuropsychopharmaceutical experimentation on patients done without their knowledge, putting them at risk of an unintentional deadly overdose.  (Just below a third raised similar risks by combining different opioids in an unsupervised manner.)

The philosopher Søren Kierkegaard, who also wrote under a pseudonym, famously saw the torments of anxiety and depression as a potential stage of transition in life.  He described angst not as a medicalized condition–and rather as a possibility for learning to be anxious “in the right way” to gain individual freedom–but his discussion of “anxiety” as an existential condition may have contributed to its adoption as a term of medical diagnosis and identification as a condition medicine could relieve:  the expectation for such a cure may have contributed to the currency of anxiety as a state of mind that could be cured.  But sales of addictive painkillers at retail pharmacies across America, if by no means confined to being a palliative for anxiety, foreclose much opportunity for individual reflection–and indeed interact with anxiety and depression medications in ways that seem to run experiments with brain chemistry, to create some of the most toxic combinations of greatest risk.  The unsupervised spread of opioid prescriptions suggests more of an abandonment to the world. Kierkegaard saw deep danger in the very ease of an ability for “losing one’s self” in the turmoil of the world, even as he prized the potential benefits of such a withdrawal from the world as offering an opportunity to learn true humility.

Rather than offering a cure, all too often the quenching of suffering by painkillers in concert with alcohol or anti-anxiety medications that have also become so widely prescribed suggests far less positive outcomes, and openly courts both accidental overdose and death.  Far more bitter than a perception of life’s nauseating air, the growing addiction to painkillers is unlike and lies far beyond the Knaussgardian abandon of confronting and monumentalizing seemingly squalid personal details and emotional impoverishment:  if Karl Ove Knaussgard excavated the seeds of deep discomfort approaching alienation to his father’s alcoholism or alcoholic abandon, the addictiveness of opioids presents an invitation to deeper discomfort as well as a possibly fatal gateway drug.   The topography of reliance on painkillers paints a picture of a country of sufferers without time for expansive introspection, but paints a picture in which, if Knaussgard described the quotidian detail of a struggle anyone could face, most anyone could die by overdose.

brain

5.  We’ve much more concentrated on policing the entry of drugs across national borders–much as the panic that surrounded the historical arrival of opium from China–by mapping the routes of the illegal transport of alien substances as if they run into our lands along smuggling routes and hidden corridors rooted in urban markets and clearing houses.  Preoccupied by the sites of entrance of drug traffickers linked to sites lying exclusively outside of our borders, we’ve projected risks  outside our borders and less closely scrutinized the mechanics and networks of the illicit dissemination of prescriptions drugs, and the accidental and unmonitored combination of psychotropic substances–and the network of providing opiates not only at pharmacies but pain clinics across America.  But the extent of demand for painkillers and opiates is home-grown–as much as they depend on cross-border smuggling, or cartels whose routes permeate borders’ safeguards, the networks that provide a spur to drug consumption are all too often dictated not by leisure drug-use, but by new patterns of job loss.

It is, after all, a bit of a truism that the United States is the largest market for illegal drugs, as one of the most profitable in the world, with heroin and marijuana entering the country since the 1970s across its Southwest border, as heroin spreads from distribution hub in New York City, rerouting heroin from Southeast and Southwest Asia, while Ecstasy (MDMA) entered from European and Israeli drug-trafficking syndicates, border-crossing corridors of trade are only a partial  map of the expanding network of market for illicit drugs within our borders.  For the rising availability of pharmaceutical opioids in much of middle America has not only grown such border crossing, but fed a growing demand for painkillers in areas where no or few earlier networks for drug sales earlier existed, and created an almost reflexive recourse to pill-popping or tar heroine as overpowering palliatives.

Yet despite the fact that the traffic in drugs does cross borders, smuggled on boats, airlines, cruise ships, and package carriers, a growing demand for opioids fed by the rise of a regimen painkillers, that did much more to attract other opioids into United States markets that never–or rarely–existed before, effectively manufacturing a deep demand for pills and narcotics where none existed before.  To map the growing networks of addiction and the dangers of personal and public health that they reveal, we might move beyond the cartels, gangs, or foreign dealers that are often portrayed as purveyors of narcotics to unloosing an unhealthy regimen in much of the country.  The acceptance of an addictive regimen of opiates that have become dispensed in drastically different proportions across America, suggest a division of the country only recently examined, but suggests a difficulty of coping with pain of suffering, and something like a readiness of newly emboldened pharmaceutical companies to minister to a broadly diffused geography of suffering, and the desperation that such readiness creates. Although pain-related health issues don’t themselves vary much by region, reliance on and over-prescription of such addictive painkillers creates a geography of social detachment not elected by its victims–but whose increasingly reflexive recourse to prescription drugs creates a growing health risk.  Indeed, the networks of providing painkillers to much of rural America–and especially throughout the Deep South–has fueled a shifting topography of addiction that would not be recognizable in earlier decades.

The rash of overdosing from painkillers and illicit drugs that seem provoked by such a now-diffuse diagnosis seem a sign, as Knausgaard may have meant to imply in the title to his book, that anyone can become an addict that has rapidly altered the geography of drug overdoses in the United States–and has the corollary that increasing numbers are dying from drug overdoses or with opiates in the blood.   The all-too-nauseating air of such enforced leisure reveals its darkest side in the tragic expanding geography of a newly toxic pharmakon of drugs across the United States, evident in the pronounced pockets of mortality in the new age-adjusted rate of death by drug-poisoning by opioids or heroin across the lower forty-eight, released by the CDC for 2006-10, revealing unexpectedly concentrations of high mortality rates from drug overdoses in states such as Washington, Arizona, Florida, New Mexico, Utah, Oklahoma, Virginia, West Virginia, and Maine–as well as northern California, across demographic groups and rural and urban areas:

Drug Poisoining by Heroin or Opioids, 2006-2010

Drug Poisoning by Heroin or Opioids in US counties, 2006-2010:  #DrugPolicyReform/data from CDC

The broad dissemination of narcotics into discrete pockets in Texas, Louisiana, Florida, and Utah is by no means limited to the urban slums where folks might imagine themselves starving for a midnight fix.  Rather than lying in the urban streets where one might be starving for a cheap fix, the dissemination of opiates is fostered by an expanding network from CVS to Walgreens, which facilitate the surprisingly varied topography of their consumption and sale in select areas at ten per 100,000.

6.  The growing network providing opiates nourishes an annual market now valued at $60 billion that is rapidly growing each year.  For each of the past fifteen years, and pretty much since the last recession of the late 1990s, a startlingly increasing incidence of death from pharmaceutical overdose have mushroomed across the US; at the same time as addictive narcotics are marketed as remedies to depression or psychological isolation–but whose interaction with antidepressants is extremely dangerous, and tantamount to running an uncontrolled experiment on the brain chemistry of many Americans. The all too dramatic diffusion of prescription painkillers in America provoked a parallel increase in recent years in deaths due to overdoses from opioid analgesics, whose numbers had risen 16,000 by 2012, between 1999 and 2010, increased at astronomical rates of 265% among men and 400% among women that demand to be explained.  (At their greatest growth, or between 1999 and 2006, perhaps in parallel to a pronounced economic downturn, they expanded expanded by 16% every year.)  The landscape of painkillers and pill mills created along highways, in non-urban areas, and even mini-malls–often using regional dealers, more than wholesalers–has not only expanded access to opioids, but a network for encouraging and proselytizing painkillers as a cure-all drug for all with minimal supervision of their addictiveness.  The spread of age-adjusted death rates for drug poisoning, including opioids as Oxycontin and heroin, dramatically grew between 1999 to 2009 in ways that reveal a strikingly changing complexion of the country as a whole over only a decade.

The total sales of prescription opioids, if $3 billion was spent on Oxycodone in 2009, grew to $9 billion in 2012.  With growing sales of  such synthesized drugs including Oxycodone, the highly addictive main active ingredient in Percocet, OxyContin, and Percodan, or hydrocodone in Vicodin, Lorco and Nortab, across America, they have brought increased numbers and waves of overdose deaths.  Indeed, the most pronounced rates of growth nation-wide in sales and addiction have brought a huge rise in health-risk, as well in addiction-associated crime in the first decade of the new century, changing the topography of opioid consumption in ways difficult to imagine or comprehend; one can also unpack its consequences one can examine in detail by scrolling across its interactive map:

AP opioid

Associated Press

The starkly shifting geography created by growth in sales of painkillers marketed as legitimate relief suggest the networks of dependence that have grown up around their use.  For the wide marketing of such painkillers has concealed their addictive dangers, and the relatively small timeframe in which such different rates of painkillers’ consumption were established–and the patterns of dependence that they have come to create.

Perhaps the acceptance of such enforced leisure is not far removed from the dependence on the desensitization painkillers promise–encouraged by medical advocacy of treating chronic pain by opioid painkiller prescriptions advocated during the mid-1990s.  After the return of vets from Iraq and Afghanistan in the new century, the diagnosis was prevalent for veterans, as well as the unemployed, or out-of-work, creating a problematic topography for painkiller prescriptions throughout the United States.  Strikingly uneven practices of painkiller prescriptions have been made with an abandon some reckon sufficient to provide every inhabitant of the country with a bottle of pills in the first decade of the new millennia, and by 2012, the contrasting topography of dispensing painkillers created a clearly defined zone in the United States that eerily mirrors what one might call a “rust” belt of a decline of manufacturing, but spread to much of the southern and midwestern states, as well as the western states–and come to outnumber the population.

Pinkiller prescriptions

Some rural areas where painkiller prescriptions are so popular outnumber the state’s population.  Indeed, the numbers of prescriptions suggest that addicts may hold several.  And these are also often some of the same states without easy access to opioid abuse therapy–which some 53% of the counties in the nation lacked in 2012, despite their overwhelmingly disproportionate affliction by painkiller addiction.  Access to needed Opioid Relief Therapies (ORT) are, for example, conspicuously absent for prisoners in many of southern states where they are most sorely needed, from Kentucky and North Carolina to Georgia to Nebraska and to New Jersey:

us_map_ort

When Bloomberg News prepared a brittle visualization of a similar time period for sales of opioids, they not inadvertently highlighted the predominantly rural zones of the most intense painkiller sales across the country’s least urban areas.  In such regions, easily accessible opioids provided the cheapest pain relief:  blue marks the index not of pain, but of the circulation of prescribed painkillers to paying customers–a record not only of “where it hurts,” but of where we are most hurt by the networks of opioid sales across America, where addiction is apt to feed a further demand for harder drugs, particularly as the snorting, injecting, and inhaling of pulverized pills creates a level of tolerance creates a mental demand for greater highs, creating a deeper hurt than the providing of such opioid analgesics may have been intended to respond!

BW23_feat_greece_405

Such data visualizations inescapably map a topographies of deep despair–of desperate searches for painkillers that doctors will not supply, of the resale of prescribed drugs, and of the deception of over-the-counter sales–despite their clear absence of qualitative information.

Multiple stories ran that year, based on similar data, announcing with amazement the emergence of rural West Virginia, Broward County, Tennessee or Missouri, as alternate epicenters of pill-popping, beside Orange County or Staten Island–where Oxycodone sales surged 1200% from 2000-2010–to frustrate law enforcement oversight.  Ten of the most opiate-prescribing states were located in the South by 2014.  The potential  possibilities of empathy are not often evident in data visualizations or maps, whose format often obscures individual cases:  the data visualization is difficult to process for its opacity and absence of qualitative details.  The distribution of the demand for painkillers as palliatives seems a tragic mirror of the growth of underemployment and unemployment across rural and exurban counties for a random month in 2009–

2009 unemployment in rural counties nonurbs

and broadly echo the decline of manufacturing across the old industrial belt:

Industrial Belt

Deeply analogous disparities emerged in the disproportionate distribution of prescriptions for painkillers in 2012, a decade after the rise of pill hills, which uncannily reflect drastic losses of jobs exurban areas during the previous five years:

job loss 2007-11
legend job losses and gains

Such data maps of economic divides in America should be compared directly to the mapping of increasingly evident coincidence of opioid sales in 2010 and fatal drug overdoses released by Dr. Ileana Arias, of the CDC:

opioid sales:drug overdose 2008-10

The juxtaposition of the layers of these datasets again dramatically illuminates a deeply poignant distribution of despair. If it cannot in any way definitively suggest the geographical basis for networks of demand for painkillers across the country, it raises questions about the abuse of painkillers, and the culture of pain relief that grows with increased recourse to retail pharmaceutical stores as outlets for relief.  Whereas maps once offered ways to imagine lands on which viewers had never set foot, but the data visualizations of addiction to painkillers offer someone of a truly terrifying mirror in which to contemplate the state of the country.  Increased pharmaceutical dependence make the US the largest consumer by far of opiate pain killers in the world, and the painkiller industry particularly conspicuous as a network of health.  CDC Director Tom Frieden acknowledges that “opioid overdoses tend to be highest where opioids get the highest use,” lamenting the potential abuse of medications “can be an important tool for doctors to use … but … not the answer every time someone has pain.” The tacit acceptance of opioids across America–concretized in the spread of almost unregulated “pill mills”–has created a virtual free market for narcotics that all but invited abuse.

law
mill1

7.  What and where are these new networks of addiction?  Whereas cities have been long seen as the hubs for foreign drug cartels, the new networks of prescription drugs from pharmacies or pain centers created a network of drug consumption and a demand for opioids located far outside of urban areas often lying in rural and exurban areas across new networks of consumption.   Indeed, the recent migration of outlets for the dissemination of prescription painkillers to such online sales outlets as Pharmacy4Pills, based in the Bahamas, and rogue pharmacies for OxyContin, methadone, and Vicodin have in a sense confirmed the ubiquitous availability cultivated over the past ten years for painkillers that expanded a growing market for deeply addictive substances across the country. Indeed, one such network lies in the increased availability of painkillers at Veteran Affairs hospitals across America.

The curious coincidence reveals a temporary tolerance of opiate-based pain-medication–and encouragement of a network of painkiller provisions–able to be mapped in somewhat terrifying manner by the widespread provision of painkillers by VA hospitals in the US that have wrought increased addiction among returning vets in an all too bitter resolution for their calls of duty in roughly.  The expansion of prescriptions of painkillers across roughly the same timeframe follows an almost identical topography, as if a culture of pain medication–or a culture of mistreatment–existed in regions of the country.  The disproportionate demand for pain medications may suggest a distinct market for prescription drugs–rather than disproportionate bodily suffering–as was recently mapped as a prescription epidemic by the Center for Investigative Reporting.

culture of pain?

Center for Investigative Reporting

There should be no surprise that this network of painkiller providers feeding Veterans’ addictions maps so clearly onto a network of addiction–and indeed a network of illicit opiate drugs.  The network has created ties among painkillers’ illegal resale and the resale of prescriptions that raise complicated questions about the intersections between a poor economy and painkillers’ prescription.  For the consumption of painkillers has so grown across the country in the past fifteen to twenty years that areas which saw  low rates of drug overdose facilities in the midwest and southwest transform into regions faced with widespread death by overdose–as prescription painkillers from Vicodin, OxyContin, or methadone are prescribed in the millions, often taken in dangerous doses, frequently illegally sold–to meet diagnoses of PTSD, depression, anxiety, brain injuries, or guilt–with the result of often overloading opiates on PTSD victims with a particular intensity. Such an overloading oddly persists despite the  actual possibility of increasing their chances of suicide:   VA hospitals over-prescribed painkillers to those suffering from PTSD, despite the possibility that their depression and devastation can lead far more easily to overdoses:  indeed, twice as many veterans were found dead from prescription drug overdoses in 2010 as the national average.  Such conspicuous variations in the rates of prescription for opiates present a painfully grim picture that fails explanation through any variations in health, health care, or the diagnosis of acute pain.

8.   Despite fairly uniform levels of pain in individual states, inadequate assessment of dangers of addiction has created a startlingly differentiated topography in the market for painkillers in relief:  the spatial differences in the dispensation of opiates reflects an increased reliance on pain medicine that ranges from prescribed to un-prescribed and illicit drugs that derive from opium that are equally addictive, if more powerful, from heroin to codeine.

prescription-bottle-120214

The recent legalization of medical marijuana–now expanded to twenty-three states–may provide some relief, as medical marijuana sales of $1.7 billion in 2011 are poised to rise above over $8.9 billion by 2016.  Despite clear dangers of long-term use that legalization of medical marijuana may provoke, the turn to medical marijuana is, in many ways, a search for the legalization of potential palliatives–an expansion of the pharmakon of remedies for ongoing pain–to ones that may offer less of a poisoned remedy, but a remedy nonetheless with its own distinct set of risks–if ones we hope more diminished from the opioids so widely available during earlier decades.

Marijuana Legalization map

Yet are the dark spots in the below map, where painkiller prescriptions are so widely sold to be able to provide the nation’s  population, symptomatic of the increased introduction of opioids for a range of symptoms, or the over-prescription of drugs whose true danger was not readily ascertained?  Even though many of such pain clinics or fairly satanic “pill mills“–which require no medical records, and exchange cash for pain medication–have been shuttered in severe cases of this national drug abuse epidemic, leading deaths from Oxycodone to plummet, there is strong resistance to creating a database to monitor drug prescriptions–or to insist that the problem does not exist.

map-opioid-prescriptions-624

The varying density across the nation reveals a stark topography of addiction that reflects the increased circulation not only of illegal drugs, but the availability of prescription opiates across the country.  The density of addiction levels maps precisely onto those areas where there is documented use of illicit drugs other than marijuana among those over age twelve–and reveals a strikingly surprising network of addiction across much of the south, southwest and northwest.

Other Than MJ

Such admittedly schematic and somewhat superficial visualizations force us to ask what sort of pharmakon exists in America. Did the availability of painkillers come to constitute a poisoned remedy for the depressed, injured, and out-of-work, which encouraged or contributed to a recent epidemic of heroin in many rural states, as painkillers were more tightly controlled?  Whereas the ancient philosopher Plato contrasted philosophy to other occupations as providing a pharmakon of greater riches to its practitioners, have patterns of unemployment unreflectively promoted poisonous remedies as an alternative to the despair of those with few hopes for stable work, and promoted self-administered substances and self-induced withdrawal to promote a palliative transportation to an alternate state of mind?

Despite the false dichotomies drug companies continue to duplicitously draw between legal painkillers/illicit drugs, non-addictive/addictive drugs, prescription sales/narcotics, Oxycontin/heroin, or Big Pharma/cartels, the ballooning of prescription sales has largely itself caused each of these binaries to collapse. The increased slippage and play between such binaries is indeed hard to deny in the booming of Oxycontin sales. The increased degree of play between these binaries seems most pronounced in areas nicely mapped in the indelibly dark data visualization below of the boom in prescription sales 2000-2010–despite the data visualization’s lack of qualitative detail, and focus on sheer growth of Oxycodone sales per capita alone.

painkiller nation

Statistics:  DEA; AP analysis–Phil Holm and Michelle Markoff

The visualization mirrors a geography of death by overdose from prescription drugs–a stubbornly grizzly distribution indeed.  The data visualization begs not only for correlation with the economics of the out of work, as the increasing money spent on drugs begs for correlation with the contracting availability of jobs.  A visualization of such sky-high rates of the prescription of painkillers in the decade from 2000-2010 is an indictment of a medical culture of diagnosis and the provision of drugs, the widespread acceptance of painkillers as a strategy of coping suggests a drastically diminished creativity and resourcefulness in public health options that is almost bound to have high future costs. The overly blind promotion of desensitization to pain through often unsupervised self-administered pharmaceutical dosing undeniably encourages recourse to pain killing drugs.

Is it a coincidence that the increased promotion of opiates for pain reduction dovetailed both with the renewed recession of the early 2000s, and the pain-killers prescriptions provided to veterans returning from war and tours of duty.  It bears investigation whether combat fatigue led Americans to consume a whopping 80% of opiate painkillers produced globally by 2012, or what was the role of the economy.  But the combination no doubt set the stage for the telling expansion of an economy of pain medicine that dispensed some 110 tons of addictive substances from Oxycontin to harsher drugs as codeine, long a preferred drug across the US.  With some 92 Oxicodone-based medications on the market and another 218 containing hydrocodone, the possibilities of addiction are not only endless, but generated by an expanding market for opioids of different brands.

The complex choropleth map of the increasing consumption of painkillers in America is rapidly expanding, even as billions are spent on the fight against drugs in foreign countries, and some fifty billion has been spent on patrolling the nation’s borders, or police patrols’ almost exclusive focus on poorer urban slums–as if addiction were still an urban phenomenon alone, without clear understanding of the new topography of addictive prescription drugs.  The health care costs and social consequences of addiction to opioids are bound to increase in future years, with sharply escalating mortality rates by opiate overdose in many parts of the country that are rarely on the map–the growing rates of deaths from overdoses of opioid pain relievers has grown sharply among women since 1999 and 2010 at a five-fold rate of increase, versus by a factor of 3.6 among men, in large part because of unmonitored dosages, but also because of the rapid rise in illicit drug sales.

While for many years, the majority of OxyContin consumption was located in Florida, For the expansion of cheap tar heroin sales in the US that have exploded over the past ten years with a rapidity that has created a public health epidemic of increasing proportions–generated in large part by the pharmaceutical industries. Such advocacy of self-induced desensitization, rather than offering an illuminating perspective on life, has created a cheaper fix than anti-anxiety meds, often themselves offered across the midwest and much of America to ease the depression of unemployed–creating a rush on heroin, it has been argued, as access to Oxycontin was restricted.  One saw between 2004-8 a  66% increase in heroin-related treatments in Ohio among suburban caucasians, as demand for black tar heroin first led Ohio to become an entrepôt in its illicit trade.  The effects are cast into relief in a map comparing overdose fatalities to traffic deaths from 2008–a year when the majority of such drug-induced deaths in the state of Ohio were in fact due not to heroin but to Oxycontin.  Although the DEA found that in 2010, ninety of the top one hundred doctors disseminating Oxycodone were based in Florida, a shifting geography of overdose and death was in fact widely distributed nationwide:

fatality_map_slideshow

It is instructive to compare the distribution of painkiller sales in the United States, shown here in 2010, by which time admission to treatment clinics had quadrupled nationwide, and review it in comparison to the geography of jobs in America pictured above–both to the loss of jobs after the 2007 Recession in rural and exurban areas, and to a reverse map of those cities where the most jobs were advertised in the last quarter of 2014, if only to suggest in broad strokes that the prescription of painkillers is increasingly tied to questions of our economy.  The focus on a loss of jobs in rural or non-urban areas, rather than urban or ex-urban areas, reflects both the greater loss of jobs in those areas, and the unique topography of opiate addiction and use that has emerged across much of the country.

state-info-mapa_626px-1
Jobs after Recession 2007-10

Bureau of Labor Statistics/Daily Yonder

The dramatic growth in Age-Adjusted Drug-Related Deaths across the country has created a landscape few would imagine in 2000, even though the increase of drug addiction has grown since the 1970s. For the deaths related to overdoses of such pain relievers as Oxycontin and its pharmaceutical cousins rose fourfold between 1999 and 2010, when they accounted for 16,650 deaths, and deaths from heroin overdoses rose by almost half from 2006 to 2010, as the number of heroin users rapidly rose from 239,000  to 335,000 between 2010-2012 (3)(4)(5).

While the abuse of painkillers were previously contained to areas of rural Appalachia, the spread of Age-Adjusted Death Rates from drug poisoning throughout the United States has, to put not too fine point on it, developed a distinct complexion that parallel an ineluctable economic downturn, as a level of incidence confined to rural Appalachia has become the new national normal–this is not an expansion of “leisure” in the ways Leys described, entirely, but an acceptance of an entrance into self-imposed lethargy of desensitization, brokered by blocking  the transmission of pain to the brain by the spinal cord, often both by blocking the sensation of pleasure and often prompting threats to depression of breathing and oxygen inhalation that leads to death by overdose.  If Leys contrasts the imposed leisure of unemployment to the leisure of Chinese literati’s time of learning, this is closer to acceptance of haze-inducing opiates that has so expanded that over 5,500 Americans are beginning to mis-use painkillers every day.

DrugPoisonDeaths1
AADR chrono
Heroin admission 2005

At the midst of this shifting geography of painkillers emerged a striking map of heroin use across the country, tied less to the growth of cartels than to the growth of a quite steady demand that seems localized in depressed economies, even if incomplete data is available in several states, but seemed shockingly diffuse throughout the country.

The topography of admission rates for heroin is strikingly similar to that of the prescription of Methadone as a painkiller:     General rates of death by drug-poisoning in 2010 had grown along a slightly different distribution, with some similarities, clustered in the Northwest, the poorer southern states, and the midwest.  The telling topography of painkillers has received increasing attention, but is difficult to resolve, so closely is it tied to the increasing economic chasms between parts of American society, and the numbness with which those out of work seem to look at them with mute despair.

2010 drug poisoning:10,000

9.  Even if one is not a huge fan of the qualitatively thin nature of most data visualizations, the stark changes in the country’s complexion over a decade suggest both massive misguided management of pain and a desperate search for quick alleviation of depression tied to expanding economic gaps–whose contours are less likely to be legible in the above maps, so widespread is the reliance on drugs to repair the psychic wounds of an extreme distribution of wealth. The distribution reflects a tolerance and expansion of pain medications has forced Dr. Margaret Hamburg of the FDA to defend their continued sale as serving the needs of an important “niche.”

Yet is a profitable “niche” market able to justify the growing levels of drug-poisoning so prevalent in the past five years as a national divide,–increasingly linked to a national economy’s lethargy?   Can the rise of dependence on opiates be ignored, given the striking rapidity of the nation-wide trends of numbers of admission to treatment clinics for non-heroin opiates since 1995, when they first emerged in select areas –Michigan, Montana, and Utah, and Mississippi, all afflicted by deep unemployment?

TEDS2k5AdFig5

SAMSHA

The newly prominent role of painkillers in the levels of drug-related deaths by 2010 presents a staggering picture of a nation increasingly abusing–consciously or unconsciously–prescription drugs:

2010 drug poisoning:10,000

The above distribution clearly reflects the increasing number of kilos of painkillers legally sold in the same year, a statistic that has undeniable tie to economic depression:

prescrip painkillerssols

The almost endemic spread of an epidemic of painkillers has occurred at considerable social costs, but stands to create far greater costs in health care.  If Oxycontin is not always a stepping stone to heroin, the widespread reliance on pain medication–and widespread marketing of the notion that chronic pain as able to pharmacologically alleviated, that allows “pain medicine” to exist as a separate field–has generated growing abuse over a ten-year period.

10.  Police data from Washington state provides a picture of growing drug abuse across that state that suggests a pattern to image addiction to painkillers across the rest of the nation.  We have all too readily identified Florida as a prime case study of the capital of pain-killers, moving from Palm Beach and to gain an audience on account of their relatively unregulated spread in pain clinics.  But another geographically quite removed micro-history of the expansion of painkillers might be effectively illustrated at closer grain in a choropleth generated at the University of Washington.  The study raises pressing questions about the transition from opiates to heroin, as a growing market for the lethargy of leisure led to rising mortality rates due to the abuse of drugs with little quality control–or indication of their potency and strength. The rapid spread throughout Washington of opiates over ten years maps in some cases onto clear communities–as the finer grain of the distribution of a period over which deaths due to opiates more than doubled over the decade, 2000-2010, when pain clinics and treatment centers emerged across this state.

police_evidence_rxop_2000
police_evidence_rxop_2009
police_evidence_heroin_2009
treatment_rx_10
deaths_county_aggregated_2001
deaths_county_aggregated_2010

The data visualization raises questions about the transition between drugs, and the growing markets for heroin, as much as its abuse–revealed best in the shocking growth of drug overdose deaths to a point at which over two people died with opiates in their blood each day in the state, raising questions about the safety of prescriptions and the possibilities of abuse.  Researchers at Brandeis found a similar link between addiction to pain-killers and subsequent treatment for heroin, and the rise of individual admissions into drug treatment centers nationwide has rapidly ballooned in somewhat surprising regions by 2011:

IN Treatmnet

Center for Behavioral Health Statistics and Quality, Substance Abuse and Mental Health Services Administration, National Survey of Substance Abuse Treatment Services

Although misleading in ways, given the spread of marijuana abuse as something of a constant in many regions, the map is most striking for the prevalence of opiates and heroin as dominant addictions.  This chart of what drugs primarily lead to admission in addiction treatment centers reveals a disturbing topography of its own, linked not only to pathways of supply, but suggests the relation between the high number of clients in treatment for heroin and opiates to large numbers of folks in treatment in the Northeast–although the surface of this rather opaque data visualization has yet to be fully scratched, and its content less suggestive of trends than of the damage that opiates and heroin have already wreaked.

diction by state

Suspending judgment on the reporting of marijuana use in addiction centers in just one month in, one can identify a topography of drug use that is particularly striking– the black spots of the nation are revealed.

Other Than MJ

The tripling of drug overdoses from painkillers over ten years has created a terrifying topography of addiction, mappable by the growing number of drug overdoses from 2008.

state-info-mapB_626px

National Vital Statistics System, 2008/“Prescription Painkiller Overdoses in the US”

The topography reflects the range of kilograms of prescription painkillers per 10,000 people across the United States.

state-info-mapa_626px

SOURCE: Automation of Reports and Consolidated Orders System (ARCOS) of the Drug Enforcement Administration (DEA), 2010

In focussing on overdose deaths among the demographic of women alone, the Center for Disease Control found a startling geographic concentration between 2009-10 in middle America which we will never know if it could probably have been prevented or forestalled.  The strikingly suggestive spatial clustering in states that one would not associate with urban areas–and might even label rural–demands investigation.  The below map of deaths per 100,000 across the union–the vast plurality of which, somewhat shockingly, are opioids and prescription drugs–reveals a landscape of addiction and dependence that is non-urban in nature, and relies on networks of the transportation and marketing of drugs among a large suburban or rural populace, whose centers of consumption may be more easily mapped than the pathways by which it arrives to prescription customers and illicit clients–which would of course require far less readily available data about its transportation.

m6226a3f1

11.  The existence of such open data is most widely available from causes of mortality, as well as numbers on those who seek help with addiction from clinics.  Morgues and hospitals are the sources of data that are most striking by far, if the self-selected numbers of those addicts who seek treatment also illuminate a less well-known landscape of desperation preferred not to be discussed, if only because it indicates an epidemic for which no easy answer or solution can be found.  In 2010, painkillers alone killed more individuals than overdoses from heroin and cocaine combined; as of 2012, some 12.5 million people admitted to regular use of painkillers without a prescription.  Indeed, as expenditures on cocaine decreased substantially 2000-2010, the weighted average of monthly expenditures on heroin regularly grew by half.  The numbers on age-adjusted drug-related deaths per 100,000 people mirrors a somewhat similar distribution for that year:

PDO-2013-Map

Prescription painkillers provided a tragically similar topography of death by overdose:

state-info-mapb_580px

12.  We are however all too apt to direct attention to drug smuggling and the currents of illicit narcotics trade that enter the nation’s borders, and the global market for leisure drugs.  The demand for painkillers and narcotics has grown in a distinct topography that demands clearer mapping, given the inescapable rises that it will be soon provoking in health costs, as well as the absence of a large share of the nation’s citizens. Global data is hardly encouraging–it reveals the increasing saliency of the US and other wealthy countries with a considerable number of unemployed creating a large market for opiates:  the rise of opiates in 2010 suggested the rise of a market considerably larger than in Canada or Mexico, or in much of Europe–save England and the Ukraine, although Russia, Afghanistan and Iraq are off the charts.

opium-use-in-north-america-and-europe-is-stable-or-shrinking-while-africa-and-asia-account-for-70-percent-of-global-opiate-users

UNODC/2012 World Drug Report

The unsurprisingly parallel rise of the consumption of Ecstasy (MDMA) as a preferred drug in the US makes it difficult to ignore as an epicenter of consumption in the western hemisphere.

the-us-narcotics-market-is-worth-about-60-billion-annually-according-to-united-nations-estimates

UNODC/2012 World Drug Report

Even as the consumption of cocaine seems to have stabilized or decreased in the US in recent years, compared to its rise in Europe, and growth in Australia, its use is shockingly widespread.

in-the-past-four-years-cocaine-use-has-decreased-in-the-us-while-increasing-in-australia-and-stabilizing-in-europe

UNODC/2012 World Drug Report

The number of seizures of stimulants like methamphetamines quite dramatically grew in the previous decades’ final years, skyrocketing from less than 500 to 6,000 and then 12,000 kg.  If one is all too apt, at times, to blame the arrival of synthetic methamphetamines on Mexican cartels, the geography of their transport is determined by a basic principle of supply and demand, a principle that expands the roles of suppliers, and is fed by the attraction of demand.

the-supply-routes-for-meth-follow-the-demand

DOJ National Drug Threat Assessment 2011

The skewed topography of the preference for methamphetamines in much of the Midwest, designated below by red dots, seems almost stanched less by border guards than the clearcut preference across the east coast for crack cocaine.

demand-is-geographically-skewed-in-the-us-as-the-west-prefers-methamphetamine-red-and-the-east-prefers-cocaine-blue

NPR recently charted a “drug super-highway” that paralleled the economic decline of our own national infrastructure, as well as globalization.  It reflects the rise of seizures of cocaine in what seem its two largest markets by far–Europe and the United States by 2010, rather than numbers of deaths or self-reported statistics.

all-things-considered-the-drug-superhighway-is-running-smoothly-as-cocaine-is-causing-more-trouble-than-ever-in-its-largest-markets-us-and-europe
in-the-past-four-years-cocaine-use-has-decreased-in-the-us-while-increasing-in-australia-and-stabilizing-in-europe

2012 World Drug Report

The growing national epidemic concentrating in particularly increased regions of addiction–no doubt prompted by “forced leisure” and long term low employment–if often described as being fueled by both the drug’s growing potency and its cheaper costs.  It is concentrated in the midwest, rural states, rather than urban areas, and reflects both an easier access and increased dependency of relatively well-off groups–who consume heroine that is often derived from poppies growing in Mexico, but are nurtured by a reliance on the prescription (or non-prescription) pain-relief pills, and according to Sam Quinones, the growth of pain clinics, rather than a topography of crime.

heroin2

SAMHSA

The global market of illegal opioids, however, is immense.  Of the approximately 820 tons of heroin and 994 tons of cocaine the circulate annually in the world, one might detail a global weather pattern of intoxication fed by individual desperation, and a sort of mania for the consumption of painkillers–followed by, if not creating–as has been sustained–a network of narcotic opioids such as heroin in later years.

151

In ways that feed these circuits, the growing global demand for desensitization seems driven by the United States and Russia, two of the greatest engines for opoid demand.  The startling escalation of the consumption of opioids in specific–drugs increasingly prescribed and relied upon for extra pain relief, or numbing bliss, is dangerously high in much of the world–although not in Europe, so far–in ways that constitute a global health risk and a national health risk as well to which we’ve turn a cold shoulder.  And the United States seems to currently consume some 80% of painkillers worldwide, largely legal, according to the American Society of Pain Intervention, using drugs like Percocet which are funneled to clients by prescription (or not) by drug companies who make a huge profit on them–without any sense of their future costs.

Opoids

UNODC/2012 World Drug Report

13.  The easy entrance of heroin as it has been tracked so far suggests an established set of channels of transportation.  But the growth of these channels are also of course fed by deeply seated demand, crossing a permeable border in ways that border control seem unable to stop–even while consuming an increasing amount of resources, public funds, and dehumanizing many.

In ways that might help to visualize how the arrival of the demand for opiates as distinct from home-grown demand, STRATFOR has mapped the varied routes by which cartels have helped move heroin into the United States, at considerable profit and indicated the global nexus of a narcotics trade that Mexico has effectively become.  But the deeply set currents of demand seem something we are more likely to continue to turn a blind eye–as well as to the health risks we are increasingly facing, and even not so unconsciously encouraging by the growth of painkillers and pain killing drugs.  Indeed, in looking at the spread of cartels and drug routes in Mexico, are we doing our best to ignore how Big Pharma and a free market for drugs–and drug sales without consultation of medical records–has created a large social science experiment without our borders that have prepared the ground and the network for the diffusion of opioid drugs to exurban areas?

The pathways of the consumption and arrival of drugs–leisure or addictive–are perhaps best understood as relative.A sobering realization is that, from the Canadian point of view, the source of all those drugs is, rather, the routes that they travel across the US and North America, and the cartels that stretch from the Gulf of Mexico to Arizona, or from Sinoloa to Arizona, and across across that all so porous Canadian border:

NA Smuggling Routes

Would that it were so easy to locate and situate the single source of the problem as lying across a border.  Opiates in the form of heroin continues to arrive at consumers five years ago from a variety of routes in a globalized economy, moving from Afghanistan and Myanmar; UNODC charted the global trafficking of heroin from Asia in 2008, revealing a shadow economy still insufficiently measured, to its largest markets in Europe, before reaching global customers.

Global Heroin0UODC
as-a-comparison-heres-how-heroin-makes-it-to-its-largest-markets

But the driving demand for opiates that structured much of the market is something very much within our own country and control.  The public health risk that have been created by the frequent prescription and provision of painkiller pills has created a network of demand–and a network of addiction–into which truly anyone may be vulnerable.

2 Comments

Filed under antidepressants, drug addiction, opioid epidemic, opioids, prescription drugs