The Hidden Side of Addiction: Allostatic Load, Health Disparities, and Opioids in the Time of COVID

For years, advocates had worked toward this moment. On November 30, 2021, the nation’s first medically supervised drug injection facility (officially “overdose prevention center”) opened in New York City. The public response—even in a bastion of progressive politics—was surprisingly negative. Opponents claimed that the centers would encourage drug use and moral decay. Why help people who are willfully making bad choices? Why waste public resources, especially when a much larger pandemic loomed?

By that same date, more than 750,000 people in the United States had died of COVID-19. For those who cared to notice, the pandemic further exposed the devastating impact of structural racism—how wealth inequalities and limited access to food, housing, and medical care cause significant health disparities. Rightfully, COVID was the center of virtually all public health conversations. And yet, at the same time a different health crisis, with similar structural roots, was unfolding.

Between April 2020 and April 2021, over 100,000 people in the United States died from overdoses, predominately driven by synthetic opioids. Much like COVID-19, the disparities were stark. While all groups showed an increase in overdose deaths during the pandemic, people identifying as Black and Native American showed a much greater increase (1). In fact, 2020 was the first year when overdose deaths occurred at a greater rate in Black compared to White individuals (1). How did the situation get so bad?

For much of the 20th century, addiction was cast as a matter of behavioral choice. Psychiatry, steeped in psychodynamic models, interpreted drug use as the result of early-life conflicts. Society as a whole, infused with a certain type of religious thinking, treated addiction as a moral failure. In a complex, bidirectional way, the two models fed off of each other and both contributed to an ongoing culture of blaming the user.

It wasn’t until the 1970s, with the discovery of the mu opioid receptor, that neuroscientists began to understand the biological basis of addiction. For the first time, scientists had a tangible target—something that could potentially lead to new treatments. But this turned out to be just the first step in a much bigger story.

Until this point, researchers had focused on the unique pharmacological properties of different addictive drugs. Then, in the 1980s, scientists made a surprising revelation: all of the drugs they were studying converged onto a common pathway—they all increased dopamine release from the ventral tegmental area to the nucleus accumbens (NAc). This wasn’t surprising for drugs like amphetamines and cocaine, which directly stimulate dopamine release. But it turns out that even opioids act through this pathway (albeit indirectly, by disinhibiting midbrain dopamine neurons). This shared circuitry led to the NAc being conceptualized as the “reward center” of the brain.

Over time, this model evolved into the more nuanced incentive salience theory. The basic idea is that the brain evolved to detect important things in the world. Smelling delicious food or hearing the voice of a friend triggers the release of dopamine into the NAc, which, in turn, helps us learn to associate events with reward. In the case of drug use, this means that over time dopamine is released when the individual experiences cues that predict the drug’s availability. Thus, a seemingly innocuous experience, like seeing a syringe on television, can trigger intense craving because an individual expects a reward. This is the insidious neurobiology of addiction: the same process that promotes natural rewards (like food and social relationships) gets hijacked and ensures that attention and motivation remain powerfully fixated on the drug.

While a more thoughtful understanding of biology might have helped dispel stigma toward addiction, paradoxically it seemed to fuel it. The problem came from how the model was covered in the media, with an overly reductive focus on the NAc as the reward center. Individuals who used drugs were cast as hedonists relentlessly seeking their next high.

It was in this landscape that a group of scientists discovered something very different. George Koob, then a neuroscientist at the Scripps Research Institute in San Diego, was studying the long-term impact of drug use on brain stress systems when he noticed something peculiar. Rodents injected with the stress hormone corticotropin-releasing factor (CRF) appeared more uneasy, more anxious, and less adventurous. In fact, they behaved remarkably similar to rodents who were withdrawing from alcohol (2). Koob knew that there were neurons in the amygdala that expressed high levels of CRF, so he asked a simple but radical question: could this same stress circuit be responsible for the anxiety seen in withdrawal? To test this, his team took rats that were withdrawing from alcohol and infused a CRF antagonist directly into the amygdala. Amazingly, the anxious phenotype disappeared and the rats behaved just like the control animals—as if they weren’t withdrawing at all (3).

The idea that the amygdala, a brain structure linked with fear and emotion, could be involved in withdrawal seemed strange. And yet, while much of the field remained focused on dopamine and reward circuits, other scientists conclusively demonstrated the amygdala’s role in addiction (4) (Figure 1A). Like the NAc, the amygdala sends important signals about how internal emotions relate to the environment—it helps form connections about people, places, and things associated with drug use. Consequently, lesioning the basolateral amygdala disrupts learning about environmental cues associated with opioids (5). In another fascinating study, animals with chronic pain showed not only increased opioid consumption but also synaptic and epigenetic alterations in the amygdala (6).

Figure 1.

(A) Brain regions involved in the negative affective states associated with addiction. Koob and Volkow (4) have proposed a 3-stage model as a framework to understand the underlying brain mechanisms of addiction: 1) the binge/intoxication stage; 2) the withdrawal/negative affective stage; and 3) the preoccupation/anticipation stage. The areas outlined in red are canonically involved in the negative affective states associated with addiction, including the extended amygdala complex comprised of the amygdala, the bed nucleus of the stria terminalis (BNST), and a transitional zone of the nucleus accumbens shell. The areas outlined in pink are cortical regions that are involved in the preoccupation/anticipation aspects of addiction and likely also contribute to negative affect. These areas include the anterior cingulate cortex (ACC), orbitofrontal cortex (OFC), ventromedial prefrontal cortex (vmPFC), ventrolateral prefrontal cortex (vlPFC), and dorsolateral prefrontal cortex (dlPFC). Brain areas are approximate and superimposed onto a 2-dimensional sagittal human brain section, as some structures are in a different location on the medial-lateral plane. (B) Allostatic load and learned relief of negative affective states. Here we present a case adapted from Evans and Cahill (8) in which a fictional patient experiences a stressor, such as housing instability, leading to worsening negative affect as represented on the x-axis. The person initially uses prescription opioids to relieve this negative affect, but repeated stressors including COVID-19 and food insecurity compound with opioid use (with the hypothetical patient eventually moving to heroin and/or synthetic opioid use) to worsen negative affective states, leading to an escalating allostatic load.

And it wasn’t just about rodents—compelling clinical studies told the same story. David Epstein, a researcher at the National Institute on Drug Abuse, applied real-time assessment of behavior and mood in people with opioid use disorder (7). This study and others clearly demonstrated that people use opioids to blunt feelings of sadness and anger (8).

This was the exact opposite of the common perception. For years, psychiatry and society had conceptualized addiction as a disease of reward. But a critical aspect of the illness was there all along, hidden in plain sight. The process of negative reinforcement—learning that the drug relieves pain, anxiety, and stress—is a major driver of addiction. This idea forced a massive conceptual shift in the field and set the stage for a critical connection between addiction and society at large.

Just 200 miles north of the National Institutes of Health along the I-95 corridor, at Rockefeller University, Bruce McEwen was advancing a new way to think about stress. A central concept in biology is homeostasis—the idea that if a system is perturbed, counterbalancing forces will help return it to baseline. McEwen recognized that this wasn’t always true—if the brain experiences the same stressor over and over, the baseline itself may shift (9). To explain this concept, he coopted and refined the term allostasis—a set of adaptive processes in neural circuits, synaptic systems, and gene expression that create new set points to prepare for future stressors.

And this is how the two pandemics became hopelessly intertwined. The past 3 years have created unprecedented levels of stress for everyone. The situation has been even worse for historically oppressed populations who are disproportionately impacted by social forces—including housing instability, lack of access to medical care, and racial trauma (Figure 1B). McEwen’s work illustrates how the cumulative impact of this stress can create an allostatic load that shifts the body’s set point to a persistently negative affective state, similar to what would be induced by withdrawal. If such an individual is exposed to a substance, there may be an initial reward, but this aspect of the experience rapidly diminishes. The crucial process is that the brain quickly learns that the drug relieves distress. When the drug wears off, the negative state returns and the individual may be even more strongly driven to use.

It’s time to move past the conventional reward-based model of addiction. The disease is far more complicated, and context clearly matters. Treatment must include a comprehensive approach that firmly situates the patient within their lived environment and accounts for the causal role of negative affective states—both internally generated (as in chronic pain and withdrawal) and externally driven (as by the systemic inequities that affect the lives of our patients). We also can’t ignore fundamentals of evidence-based medicine—including that medications like buprenorphine and methadone protect patients from the cycle of withdrawal and make them far less likely to die.

The data are also clear for supervised consumption sites: they save lives (10). Opponents of these centers claim a moral imperative. But a true moral approach would be to address the deep inequities in our society that are fueling the crisis and not stand idly by while people die of preventable overdoses. In the meantime, the centers remain open in New York City and offer a brief respite for individuals trying to escape the grips of pain, stress, and anxiety exacerbated by an unjust society. And having a safe place to escape just may save their life.