This has been the week where everyone is scrambling to come up with The Take about Luigi Mangione, alleged shooter of Brian Thompson. In the scramble, an interesting bit of trivia has been overlooked except as a joke among certain kinds of terminally online communities: Mangione’s Goodreads review of Industrial Society and its Future by Theodore John Kaczynski, a.k.a. the Unabomber manifesto. As a long time Manifesto-head, this obviously drew my interest in a more-than-extracurricular way.
The basic overview of the Manifesto is essentially summed up in its opening line: “The Industrial Revolution and its consequences have been a disaster for the human race.” The arguments, which are not really all that outlandish (or all that original), are that the requirements of industrial society – technologically advanced society organized on a large scale such as we’re living in today – tend necessarily and, for Kaczynski, unacceptably, towards restriction of human freedom and autonomy, mostly (though not completely) via disruption of what Kaczynski terms the “power process.” (The power process is not all that important for the rest of this newsletter, so I’m not going to go into it further.) The terse manifesto (though not as terse as Mangione’s own) argues that the good parts of technological advancement are inseparable from the bad parts; that in the final analysis, there’s no way out but revolution (at best) or individual withdrawal and acts of disruptive violence intended to accelerate the painful demise of the system, which is obviously the route Kaczynski himself opted to take.
It’s hard to think of a technological apparatus that subverts human freedom and corrodes human dignity as efficiently as the US health care system. The exuberant, rattling anger that has bubbled over like boiling milk from a pot in all corners of US society, all ends of the ideological spectrum, and all walks of life following the shooting attest to how much this particular way of organizing advanced technical society degrades and dehumanizes people – and pisses them off. Horror stories abound. So too does reporting on the bottomless evils of the health insurance industry. Everyone knows. In particular, the highly technified aspects of this industry have come under scrutiny in recent years, though as I will ultimately argue, that scrutiny (much like Kaczynski’s manifesto) is more than slightly misguided.
Just last year, United Health Group got in trouble for issuing so-called “algorithmic” denials of care in its Medicare Advantage programs (for last week’s post with more info on Medicare Advantage, see here). Here’s how Stat News summed up the lawsuit:
The class action suit, filed on behalf of deceased patients who had a UnitedHealthcare Medicare Advantage plan and their families by the California-based Clarkson Law Firm, follows the publication of a STAT investigation Tuesday. The investigation, cited by the lawsuit, found UnitedHealth pressured medical employees to follow an algorithm, which predicts a patient’s length of stay, to issue payment denials to people with Medicare Advantage plans. Internal documents revealed that managers within the company set a goal for clinical employees to keep patients rehab stays within 1% of the days projected by the algorithm.
The basic story is this. In 2020 United Health Group, as part of its aggressive vertical integration strategy (involving lots of mergers and acquisitions), acquired a subsidiary called NaviHealth which had developed algorithmic software called nH Predict. It’s worth quoting from the news story about the acquisition at length (bolded emphasis mine):
Founded in 2012, Brentwood, Tenn.-based NaviHealth teams with health plans, hospital systems, risk-bearing physician groups, and other health care providers to coordinate clinical decision-making and track patients’ data via its technology tools. The company manages post-acute services for about 4.5 million Medicare Advantage members and serves more than 140 hospitals in the Centers for Medicare and Medicaid Services’ Bundled Payments for Care Improvement Advanced program.
The company developed decision-support tools and uses standardized clinical operations and local care coordinators to help identify existing care resources in the community and the best next course of treatment for each patient.
NaviHealth's approach has shown to help patients recover more quickly from an acute health episode and hospitalization, according to the company.
The startup will continue to serve its clients and members as a standalone entity aligned with OptumHealth, the health care services and delivery business of Optum. The company will maintain its current leadership team under CEO Clay Richards.
Recall that Optum is part of United Health Group, of which United Health Care is the insurance arm that is a major player in Medicare Advantage. Here, “decision support tools” means algorithms for making care determinations. nH predict is an algorithm designed to “support” clinical decisions about post-acute care for Medicare Advantage beneficiaries — like all algorithms, it is optimized to some kind of criterion. In this case, I would guess that the criterion is some kind of metric of cost-per-length of stay in a post-acute care setting like a nursing home.
Okay, fine. The controversy comes in because following its acquisition of NaviHealth, United Health Care began using the nH Predict software to issue psychotically cruel denials of payment, which amount to denials of care. Basically, this amounted to kicking elderly people out of long-term care where they were recuperating from orthopedic surgeries and things like that after, like, eight days, dumping them back at home without appropriate care resources to support their recovery. (The Stat News reporting explores several such cases in wrenching detail.) Employees report being heavily pressured to obey the algorithm’s clearly bananas outputted recommendations, on penalty of being fired.
The response to the reporting was, somewhat predictably, focused on this technological, algorithmic aspect. It is horrifying, after all. And it does seem, at first blush, like a technical nightmare straight out of Ted Kaczynski’s acid-zapped incel brain. Let us quote from Lewis Mumford, historian of technology whose influence on the Manifesto is clear:
The center of authority in this new system is no longer a visible personality, an all-powerful king: even in totalitarian dictatorships the center now lies in the system itself, invisible but omnipresent: all its human components, even the technical and managerial elite, even the sacred priesthood of science, who alone have access to the secret knowledge by means of which total control is now swiftly being effected, are themselves trapped by the very perfection of the organization they have invented. Like the Pharoahs of the Pyramid Age, these servants of the system identify its goods with their own kind of well-being: as with the divine king, their praise of the system is an act of self-worship; and again like the king, they are in the grip of an irrational compulsion to extend their means of control and expand the scope of their authority. In this new systems-centered collective, this Pentagon of power, there is no visible presence who issues commands: unlike job's God, the new deities cannot be confronted, still less defied. Under the pretext of saving labor, the ultimate end of this technics is to displace life, or rather, to transfer the attributes of life to the machine and the mechanical collective, allowing only so much of the organism to remain as may be controlled and manipulated.
Do not misunderstand this analysis. The danger to democracy does not spring from any specific scientific discoveries or electronic inventions. The human compulsions that dominate the authoritarian technics of our own day date back to a period before eve n the wheel had been invented. The danger springs from the fact that, since Francis Bacon and Galileo defined the new methods and objectives of technics, our great physical transformations have been effected by a system that deliberately eliminates the whole human personality, ignores the historic process, overplays the role of the abstract intelligence, and -- makes control over physical nature, ultimately control over man himself, the chief purpose of existence. This system has made its way so insidiously into Western society, that my analysis of its derivation and its intentions may well seem more questionable -- indeed more shocking -- than the facts themselves.
To be clear, I have not read Lewis Mumford, except in brief, excerpted, and summary form. (I’m interested to but you know, who has the time for anything!) If you’ve read Industrial Society and its Future, you can read Mumford all over it. (Jacques Ellul’s more pessimistic The Technological Society was apparently a post-facto influence.) But I think the Congressional response points to an important way in which Kaczynski actually misread Mumford, one that has implications for us now. With a particular sort of inauthentic shock, members of Congress issued Statements about how machines should never be allowed to make important care decisions. Left untouched and unmentioned was whether it’s okay or not for humans to make those same decisions — or the fact that it was not nH Predict compelling UHC employees to follow its dictates, but rather other humans, those in management.
My (again, admittedly very limited) understanding of Mumford’s argument is that the history of technology reflects and solidifies human choices and priorities — not the dark inverse of the Silicon Valley type of techno-optimism that Kaczynski envisions, of detached autonomous machines overtaking us and beating us into submission. The machines, in an important sense, are us. In the much more recent book Fossil Capital, Andreas Malm demonstrates (convincingly, in my view) that it wasn’t mere technical superiority or cost-effectiveness that really gave coal-fired steam power the kick in the pants needed to overtake other forms of power like water — it was the transportability, reliability, and constancy of coal, which allowed for an imposition of labor discipline necessary to the burgeoning Industrial Revolution that water power, for example, didn’t allow. (My understanding of Mumford is way too incomplete, but from what I understand this glances off of an argument he makes in Technics and Civilization).
Way back at the beginning of the Industrial Revolution that started humanity down this disastrous path, a guy by the name of Friedrich Engels (who is very familiar to anyone that has been reading this newsletter) coined a neat little term to describe a phenomenon particular to the emerging industrial-technical organization of life. The term, which will also be extremely familiar to readers of this newsletter or Nate’s, is social murder. Via a long analysis of the Marxian concept of “mute compulsion” elaborated recently in a book by Søren Mau (and covered by yours truly in this post), we can show straightforwardly that social murder issues from what Ted Kaczynski might call the technological organization of society, and which I would call the “rules of the game” of capitalism itself.
Through working on this newsletter, I have started to theorize social murder as a subcategory of structural violence, one that can serve as a theoretical basis for many rather adrift concepts in public health such as so-called “health disparities.” Social murder helps us describe actually-existing population health and ground it conceptually in a capitalist political economy, be it global or national (or local). One of these aspects of population health that social murder helps us make sense of is health insurance – in fact, I am planning a whole post about health insurance, qualitatively, as a social determinant of health, so stay tuned for that.
Returning to United Health Group and nH Predict, the issue is not exactly that machines were making decisions that should be made by humans. It’s not exactly that the decisions rendered would be fair or just had they been made by humans. It’s that the decisions, wherever their enforcement authority ultimately lay, were optimized to an unjust criterion, one that reflects what we might variously call the logic of capitalism or the profit motive or the valorization imperative. To see this more clearly, let’s look at how the nH Predict actually functioned within the context of United Health Group in the broader economy — something that has been generally overlooked in the coverage of the care-denial scandal.
As I’ve mentioned previously, United Health Group has two big arms: the insurer, United Health Care, which is (again) a major player in Medicare advantage; and another entity called Optum, which is an organizational umbrella encompassing a bunch of different things, from so-called PBMs (pharmacy benefit managers), to a whole health informatics and data analytics vertical, to a variety of “senior care assets” like nursing homes, actual brick-and-mortar long-term care facilities. UHG’s strategy of aggressive acquisitions of all these things on the Optum side, including NaviHealth and with it the nH Predict software, is an approach called “vertical integration.” This is exactly what Andrew Carnegie pioneered back in the OG Gilded Age: establishing control over, in Carnegie’s case, all aspects of the steel production process under one business entity. I (following other business and health care writers who identified this) have mentioned that United Health Group generates a lot of its revenue via “flywheeling” — which means its different business components doing business with each other.
The acquisition and use of nH Predict was probably not carried out under the sign of some sinister and inscrutable technical mandate. It was done in the context of UHG’s vertical integration push, to squeeze as much value out of its flywheeling business dealings as could be squeezed. United Health Care insures a large number of Medicare Advantage beneficiaries; United Health Group - Optum owns a lot of nursing homes where Medicare patients are often sent to recover from surgery and things like that. Those care settings are expensive. In one sense, UHG is paying itself by insuring those patients for post-acute care stays, so you might imagine that they would be incentivized to pay for longer stays. However, the proof is in the pudding (the purpose of a system is what it does) — in another sense, there’s probably not a great return on investment for the insurer in paying for longer stays. At some point, I’m guessing, the “costing the insurer” and “paying the nursing home” curves cross one another. What nH Predict really helped UHG do is squeeze money from this circuit by hustling Medicare Advantage Patients that they insure out of expensive care settings that they own after an “optimal” length of stay — optimal as defined by the algorithmic workings of the software, which mystify the straightforward business calculus of UHG.
The crucial point is: it’s not technology per se. It’s capitalism. (Ted Kaczynski, who devotes a substantial portion of the Manifesto to whining about “leftists” and their weakling psychology of ressentiment, did not really grasp this). The Industrial Revolution wasn’t simply the rise of the machines, it was the development and use of machinery out of the dialectical unfolding of the capitalist mode of production, as Marx describes in Capital Vol. 1 in the chapter “Machinery and Large-Scale Industry.” I think it matters whether we want to understand the health insurance industry and its evils as an autonomous technical behemoth, superficially at the level of algorithmic decisions, or whether we want to understand it as a way of organizing technological and economic resources, incentives, and choices given the political-economic structure of capitalism which, after all, we’re all workin’ with. The former – the story of autonomous technology – kind of strips away the industry’s guilt and lets the evil little nuclear reactor at the core of our fucked-up health care system – which, again, we might call the logic of capitalism or the profit motive – off the hook. Crucially, it misdirects attention that should be focused like a white-hot laser beam on the real problem: private, for-profit health care, and the thugs who administer it.
Proponents and lackeys of the health insurance system as such make technological-sounding claims about it, that all the furtive and nefarious little dealings they do are really about “bending the cost curve” or (ironically echoing Ellul) something about “efficiency.” Everyone knows this is bullshit. In this sense, I have a little bit of hope. Reading Stuart Hall as I have been, I’ve been asking myself – what are the points of contradiction and change revealed here? One big one I can think of is that people have been beaten up so terribly by the baldly criminal misconduct of these increasingly vertically-integrated companies that you really don’t find anybody willing to defend health insurance or its practices anymore, not even like you did back pre-ACA in 2007 or 2008. Nobody fucking cares about the abstract value of “choice” or getting to “keep your doctor,” nobody fucking believes that we have the “best health care system in the world.” Everybody is tired of getting surprise-charged $1000 for routine blood work, or worse. Even doctors (the ones that don’t work for insurance companies) are tired of spending valuable time on the phone fighting with insurers over their medical recommendations.
Everyone in the country hates their health insurance, everyone in the country has at least one heart-rending horror story of trying to do something like get an X-ray for a broken limb or seek cancer treatment. The aftermath of the shooting has been a moment of collective catharsis, but nothing has really changed in this fundamental structure, and, concerningly to me, avenues for change are still opaque. Posting feels good, but won’t do it. Neither — to sort of paraphrase one emerging line of critique and praxis around this — will simply breaking up vertically-integrated insurers like UHG, welcome as that would be.
It’s true that the rules of capitalism are doing this to us in an abstract sense, but they’re not doing it without a lot of help and facilitation from conscious human agents, some of whom are getting very, very rich. It feels like the first big revelation to emerge in the wake of the shooting is just this: there are people that are doing this to us, and they have goofy LinkedIn pages. I hope that this can open into a further revelation: that what seem like the adamantine mandates of technical organization and efficiency are, in equal measure, human choices and adaptations to the political-economic incentive structure. There could be other choices and adaptations, just as there could be deep changes to the structure itself. I hope we (vague “we,” as a country or something) can nurture this more open-ended sense of possibility and not short-circuit into Unabomber-style impossibilism and violent nihilism — the belief that technological society is a unidirectional vise, ever-tightening around us, and we have no choices but to submit or withdraw. It doesn’t work like that. It is, as I said earlier and say a lot, an unfolding dialectic, with internal contradictions that are constantly ripening. Rather than a problem of technical domination, the health care system poses a political problem with very, very possible political solutions – lots of other countries have universal health care. It can be done.
I won’t insult you by offering more bullshit handwaving about “organizing,” but it is my sense that to get universal health care in this country, we need something like a popular movement with substantial organizational and political might. Beyond that, I leave it to people like Tim to outline the actual practical process of working towards this.