Meanwhile, the institute shed some of its original congressional mandate for treatment and prevention, by giving the job of funding mental-health services to a new federal entity, the Substance Abuse and Mental Health Services Administration. From then on, as one former N.I.M.H. director told me, undertreatment and homelessness were SAMHSA’s responsibilities. As the Human Genome Project launched, and brain-scanning technology leapt forward thanks to functional MRI, pressure to find genetic and brain signatures for psychiatric illnesses grew. Yet, as the new millennium commenced, a specific scan for disorders such as schizophrenia remained elusive. Dreams of single genetic causes were dissipating. A crisis was brewing.
History holds a large, unmarked graveyard filled with the ideas of those who tried to pin down the ultimate causes of mind/brain illnesses. Critics and scholars have portrayed some of the memorable failures—a procession of phrenologists, degeneration theorists, germ enthusiasts, wild psychoanalysts, political revolutionaries, and sexual liberationists. All of them pushed for their cherished notion, only for it ultimately to be found misguided, wanting, or worse.
There’s an underlying reason for all this zigzagging. Picture an archer’s target; at the bull’s-eye, place the most basic of possible causes for mental illness—say, genes. Huntington’s chorea, a fatal disease that affects cognition and movement, is solely genetic: it sits wholly at the center of the target. But other, more common conditions, such as schizophrenia and bipolar disorder, can only partly be predicted in terms of genetic risk, and, in most other forms of mental illness, genetic determinism further diminishes. To fully understand those diseases, we have to start looking to the next ring in the target. What else might be at work? We may now take aim at neurons, then jump out to neural circuits and networks, then to the entire brain with its hundred billion neurons and trillions of synapses. At any of these different levels of biology, a pathogenic event might disrupt us.
As if that were not enough to overwhelm us, there is much more to consider. Next, our psychiatric archers must move their focus from the brain to the mind, bump up against the mind-brain problem, hurry past dozens of philosophers, and simply grant that minds, in part, can cause things to happen. After that, they must turn their attention to the other outer rings, such as the self, individual behavior, the social world, and the nonhuman environment. Each of those holds the possibility of specific kinds of trouble: negative thought patterns; chronic affects like fear or shame; relationships filled with abuse; deprivation, poverty, and our catchall term for many horrors, trauma; and then, in that very last circle, poisons, bacteria, and viruses.
By taking up all of the rings on that target as potentially interacting causes of illness, psychiatry captures a rich set of human possibilities, from errors in our molecules to forces like racism. Clinicians can take a shot at any of them, unleashing as many arrows as they need. Prozac, psychotherapy, leaving a brutal spouse? Yes, yes, and yes. But experimental science requires studies that reduce a vast field of variables to an independent one whose effect can be tested. Unlike a psychiatrist working with a patient, a scientist in search of a soluble problem must limit herself to only one spot. And so psychiatry has long been a scientifically unstable discipline; it has veered back and forth between different explanatory models because its object of study, the mind/brain, presents the most overwhelming array of epistemological problems in all of medicine. In our quest for valid and reliable answers, it’s easy to get lost.
In 2002, at a moment when the clinical promise of the Decade of the Brain remained unfulfilled, it came time to choose a new N.I.M.H. director. Thomas Insel, a leader in the quest to find biological explanations for complex behaviors, got the job. Insel was brilliantly successful, famous for illuminating the role of the hormone oxytocin in eliciting bonding behavior in voles—an important finding in the emerging field of social neuroscience. Frustrated by the weaknesses in the nation’s mental-health-research program, he concluded that they stemmed from one of the foundations of clinical work: DSM-III was serving practitioners and patients well enough, but its categories were sometimes too muddy for researchers in search of well-defined scientific targets. How many specific kinds of depression lurked in “D.S.M. 296.31, Major Depressive Disorder, Recurrent, Mild”? Most experts would guess that there were many. And, if one did not tease apart those variants, how could anyone figure out what caused Jim’s depression but not Jane’s, why Zoloft worked on Amelie but not Eli?
In 2010, Insel and his team unveiled the Research Domain Criteria, or RDoC, a new framework for the study of mental disorders that introduced its own nomenclature and benchmarks. The move seemed to separate scientific research from the language and culture of patient treatment—a divorce made more bitter when Insel suggested that DSM categories were mere constructs “based on a consensus about clusters of clinical symptoms, not any objective laboratory measure.”
RDoC’s influence was felt across the institute’s research portfolio. It aspired to be objective. It also prompted scientists to ask questions about disorders in a very specific way. You could study problems like hyperactivity in kids, post-traumatic stress disorder in rape victims, or self-harm in adolescents—but, to maximize your chances of being funded, your study had to incorporate a measurable characteristic, such as a gene or neural circuit, that reflected an underlying biological process. In a 2013 TED talk, Insel, standing before mesmerizing brain scans and images of neurons, assured his audience that new knowledge based on this approach—he had previously called it “clinical neuroscience”—would soon sweep away two centuries of psychiatry.
Since then, this new paradigm has powerfully altered what psychiatric scientists look for—and what they look past. For example, researchers have discovered hundreds of genetic loci associated with schizophrenia and with major depression, and more than fifty for bipolar disorder and autism. Each time a new correlation is found, geneticists celebrate. But, as E. Fuller Torrey, an advocate for the severely mentally ill, told me, so much success has added up to failure. “They have identified a lot of risk genes, not any that cause a disease,” he said. “That’s very embarrassing to them.”
Insel left the N.I.M.H. in 2015, and later confessed that one of the reasons he did so was this same dispiriting realization. In a 2017 interview, he elaborated on his departure. “I spent thirteen years at N.I.M.H. really pushing on the neuroscience and genetics of mental disorders,” he said. “And when I look back on that I realize that while I think I succeeded at getting lots of really cool papers published by cool scientists at fairly large costs—I think $20 billion—I don’t think we moved the needle in reducing suicide, reducing hospitalizations, improving recovery for the tens of millions of people who have mental illness. I hold myself accountable for that.” When I spoke to Insel recently, he said, of RDoC, “I think it became an academic exercise. . . . You want to pick up measures that actually are of value to patients, families, and providers. And RDoC got way too complicated. It wasn’t really tied to clinical outcomes in a way that would matter.”
Meanwhile, a decade spent in search of so-called biomarkers crushed clinical investigators who had been trained to use descriptive DSM categories and who aimed their studies at symptom relief and therapeutic impact. By 2015, only around ten per cent of the N.I.M.H. budget was directed toward clinical research. Psychotherapy researchers, who had made much progress before RDoC, saw their funding dry up. Barbara Milrod, of Albert Einstein College of Medicine, told me, “I am angry, as a psychiatrist and as a clinical researcher, because we are doing nothing for our patients and losing generations of researchers and methods.”
Imagine a lighthouse keeper whose beam and horn guide ships in storms. Imagine that this operator, in an epiphany, realizes that all the difficulties he encounters come from water and air. He determines to study the chemistry of H2O and O2. This steward is no eccentric but rather a prestigious and powerful voice in his field; thanks to his financial largesse, many others follow his lead. They all stop worrying about their beacons and foghorns, and no longer bother with weather reports, tides, or distress signals from vessels. When called to task, they assure those whose loved ones have drowned that, though it might take fifty or a hundred years, the riddle of water and air will eventually be solved.
Sound ludicrous? But where were our psychiatric sentinels as opioids, alcoholism, and suicide ripped through the struggling towns of middle America? It took two economists, Anne Case and Angus Deaton, vacationing in Montana, to notice the early mortality of white males around them, and, in 2015, alert us to the shocking numbers of what they called “deaths of despair.” Why did the mental-health-research community fail to notice this? One reason, perhaps, is that there is no gene for social collapse.
The COVID pandemic, with its complex biopsychosocial effects, was a cataclysm that emerged from our environment, and its psychiatric consequences have only begun to be understood. Much of its impact, I fear, might simply be ignored, because many of our lookouts remain intently focussed on threats from the opposite end of the causal spectrum. While we concentrated on things like neural circuits, a viral menace attacked. The fear, helplessness, and isolation that it created roiled our communities and families, put great pressure on our emotional and psychic lives, and deeply affected our children. We need to pivot so as to better comprehend those realms, for the pandemic has thrown overly reductive assumptions about neuroscience into contradiction. Yes, malfunctioning brains can make us ill, but three years of death, uncertainty, and angst have demonstrated a homespun truth: the world can really mess you up.
The United States has the most funding for psychiatric research in the world, arguably the greatest array of professional talent, and significant private and public capacities. And yet the rising tide of mental illness after COVID will only highlight how our social contract with those patients has long been broken. The idea that shelter and humane care are human rights has dedicated advocates but little political power. In addition, although clinicians are armed with medications and therapies, they have long been pleading for new and better tools. Guidance about what this once-in-a-century pandemic might bring their way should come from our scientific and public-health leaders in Washington, but they are divided, with separate fiefdoms for psychiatric research, alcohol abuse, drug abuse, epidemiology, and the delivery of services. We suffer from systemic failures that seem to be no one’s responsibility.
A notable exception, Vivek Murthy, the Surgeon General, has called attention to the post-pandemic psychiatric crisis, citing burnout among frontline health workers, a spike in teen suicide, and an “epidemic of loneliness and isolation.” But who will take up his call? The National Institutes of Health has created the RECOVER initiative, which will support studies of the medical aftereffects of COVID, and Congress has allocated a small amount of money directly to the N.I.M.H. for COVID-targeted research. It’s not obvious what will happen when those funds run out.
To be clear, no one I spoke with advocated for calling off the Mars mission to understand the brain; exciting work is being done in optogenetics, in circuit dynamics, and in mapping the brain’s structural network (the “connectome”), to name just a few domains. It is critical that the U.S. invest in such basic research. Similarly, it would be unconscionable not to pursue solid, clinically relevant neuroscience. But there needs to be an adjustment. When I spoke to the present N.I.M.H. director, Joshua Gordon, he admitted that the introduction of RDoC had come at a cost. “It wasn’t communicated to the scientific community in a way that they understood. They took it as a kind of severing of the N.I.M.H. from diagnostic frameworks,” he said. “In my opinion, what should have been said was that it’s clear that there’s heterogeneity within our disorders, and there’s overlap across our disorders. The diagnostic labels are useful. But they have not proven tremendously useful in terms of uncovering biology.” Since Gordon’s appointment, in 2016, the N.I.M.H. has somewhat relaxed its focus on RDoC methodology. When I queried him about its value, he said, “Certainly I emphasize it less than my predecessor did.” Yet RDoC’s adoption reflected decades of at times quite strident belief that the causes of all “real” psychiatric illness could be captured by clinical neuroscience. These may not be commitments that can be easily undone.