Why treatment for severe mental illness looks radically different for rich and poor people

Earlier this month, the federal government released new data estimating that more than 650,000 people experienced homelessness in America on a single night in January 2023, an increase of 12 percent from 2022. More than 60 percent of the some 400 jurisdictions participating in the federal Continuum of Care program across the country reported growth in the number of people found sleeping outside or in cars, abandoned buildings, or anywhere not meant for humans to live. This rise in “unsheltered homelessness” was not limited to cities; more than half of rural and suburban communities saw their numbers go up too.

As street homelessness grows, so does a corresponding mental health and drug crisis. Many people experience trauma when they lose their homes and attempt to survive outside, and many turn to substances like methamphetamine to curb hunger or stay awake. While leaders maintain they’d like to clear their streets, they face barriers like a lack of affordable housing, a lack of stable medical and social services, and resistance from unhoused people to staying in cramped, congregate shelters.

Neil Gong, an assistant professor of sociology at the University of California San Diego, researches psychiatric services, homelessness, and how communities seek to maintain social order. His forthcoming book, Sons, Daughters, and Sidewalk Psychotics: Mental Illness and Homelessness in Los Angeles, explores inequality in mental health care, and specifically how divergent the psychiatric treatment options are for those living in poverty compared to those from wealthy families.

Gong’s book provides a new, critical lens through which to think about how cities respond to homeless tent encampments, deploy the Housing First model, and approach drug decriminalization. His work offers clarity for raging debates about whether governments should bring back forced treatment and asylums for those opposed to voluntary care.

Senior policy reporter Rachel Cohen talked with Gong about his research and what it means to have such bifurcated systems for treating mental health. Their conversation has been lightly edited and condensed for clarity.

Rachel Cohen

Your book finds that there are effectively two different systems for people dealing with serious mental illness. Can you explain how those systems have not only different treatment options, but also different measures of success?

Neil Gong

So we have a public safety net system, which has evolved since psychiatric institutionalization. And in many ways, it just sort of shuttles people through our broken welfare state. What that system is largely focused on is addressing all of these complex social problems like patient homelessness, incarceration, re-arrest. So if you look at the actual quantitative metrics they’re using, they’re things like how many days have people been homeless, how many days they’ve been hospitalized, how many days they’ve been incarcerated, and basically the goal is to lower those numbers. That makes a lot of sense, that’s a totally reasonable thing to do.

But then once you go and you look at what care is like for elite people, suddenly you realize there’s this whole different set of metrics you could be using for thinking about improving mental health care, with goals around not just symptom reduction but helping people rebuild their identities. So in many cases for affluent patients, this is about, for example, getting back to college after a psychotic break, getting a stable job, perhaps therapeutic work around complex family dynamics, helping people figure out what their dreams are, and trying to achieve them.

Rachel Cohen

When we think of bifurcated health care, we typically think about how a rich person might be able to access a certain type of effective cancer treatment that a low-income person might not be able to afford. In that scenario, the ideal therapeutic treatment is the same for both individuals, it’s just out of reach for the poorer patient. When you look at this landscape of psychiatric illness, do you feel like the ideal treatment for affluent people is the same ideal one for those you studied living in urban poverty?

Neil Gong

For the two treatment worlds I studied, each is able to achieve success, but only because they have such different ideas of what success is. So in the public safety net, because they’re dealing with patient homelessness and incarceration and people being stuck in this street-shelter-jail cycle, most of the focus gets devoted to that. And so in a sense, they are doing this more holistic treatment, not just using medication to focus on symptom reduction.

At the same time, it involves very different treatments than what you see for more privileged people, where you’re talking about a variety of types of psychotherapy because the goal is not generally about housing and basic stability, it’s about transforming the self. So you end up with different kinds of therapies that might range from psychoanalysis and behavioral therapies to family therapies. All of those things could certainly be positive for people who are living, say, in downtown LA and getting public safety net services, but in a sense, it doesn’t really make sense in that world because they’re so concerned with this other stuff.

Rachel Cohen

In your book, you coin this concept of “tolerant containment.” Can you explain what you mean by that?

Neil Gong

So tolerant containment is this approach to social deviance, which is basically neither trying to really correct problem behaviors nor trying to get at the root causes of an issue, just tolerating it so long as we can keep it out of the way. I see this as emerging from separate developments, with civil libertarian court decisions around social disorder on the one hand and then fiscal austerity on the other. In many ways, tolerant containment is this response to the social and economic costs of things like mass incarceration, or the asylum system, but we end up with situations where cities are told by the courts that you can’t police encampments or arrest drug users or hospitalize people in psychosis against their will — but we also don’t have the kind of resources we need to actually help people. You kind of end up just tolerating things in public space.

Rachel Cohen

Is there a difference between “tolerant containment” and harm reduction?

Neil Gong

I think of tolerant containment as a governance strategy focused on managing social disorder, whereas harm reduction comes from a philosophy of protecting individual and public health and respecting agency. In many cases, harm reduction did come out of crises where people were essentially abandoned, like during the AIDS crisis or the opioid crisis, leaving people to do the best they could with what they had. But I see tolerant containment as something cities have come to do because they have to since they are hemmed in by civil liberties law and fiscal austerity. The two overlap, and a city agency may use harm reduction practices in its pursuit of tolerant containment. Or, on the flip side, a well-intentioned harm reduction practice like Housing First may descend into mere tolerant containment when implemented poorly.

Rachel Cohen

In your book, you look at what emerged after the closing of asylums for treating low-income people with serious mental illness. What does that look like today?

Neil Gong

So in California, what emerged are called Board and Care homes, which are essentially these psychiatric group homes, which is this new business model that came about because people now had these federal disability checks. And these places are characterized by minimal oversight. They technically have rules about taking medication and not drinking, but the SSI-derived economic base means there’s really just not a lot of resources, and they essentially become these flop houses.

Rachel Cohen

Some of these places sounded really bleak. What do you see as the difference between these and the old asylums?

Neil Gong

The huge difference is they’re usually not locked and in a community setting. So people will go outside, and they can sort of do as they please, during the day, because again, there’s not enough staff to surveil residents. And although there might be rules against drinking and drugs, there’s often no one checking up on you. There’s certainly no therapeutic activities, because again, there’s not enough staff. And so it ends up being what I see as this kind of de facto harm reduction model. It’s basically tolerant containment: They tolerate a certain amount of drug use and social disorder so long as it’s not too disruptive to other residents within the building.

Rachel Cohen

There’s a lot of evidence for the Housing First model for ending homelessness, but reading your book I did start to think about the approach through the lens of tolerant containment — and moving people into housing primarily to satisfy the objectives of getting off the streets, staying out of jail, and out of the ER.

Neil Gong

Yeah, there’s a way in which the client empowerment and civil libertarian impulse behind harm reduction activism and Housing First can easily dovetail with the reality of austerity. In other words, because social workers do not want to force someone into being clean or to take psychiatric meds, there’s this sense of, “Well, we got someone inside and they said they’re fine. And they want to be left alone, so we must have achieved our goal, and we’ve respected their rights.” And there’s something to that logic, but there are also times when people are saying no to treatment or to care and contact for a variety of complex reasons, including having been burned before and trauma.

Rachel Cohen

Something that came through clearly in your book I had never really considered before is how private mental health providers and affluent families would never consider Housing First a successful treatment for severe psychiatric illness.

Neil Gong

This was articulated very well by a case manager who himself was a former patient of one of these elite clinics and then had done his social work internship at a Housing First agency. And he said, paraphrased, that the Housing First model was great in terms of keeping people out of prison and saving the county money, but you’d want more for your own loved one. And then I started to see that everywhere, which is that for wealthy people, the idea that their loved one might be living alone in an apartment yelling at a wall was hardly a success. Success for them is defined much more around these upper-middle-class norms around work, school, friends, and family.

I think Housing First should certainly be part of our system as a baseline for getting people into stable housing, but if that’s where you stop it’s as if we’re acting as if poor people with mental illness have no future or don’t deserve one.

Rachel Cohen

I recently wrote about tiny homes where advocates are excited about giving people who are living on the streets a private room with a door that locks. Did that come up in your research at all with regard to tolerant containment?

Neil Gong

From a safety and dignity perspective, a private space with a locked door can be important. But I think the danger that advocates see is also real, which is that municipalities may invest in these solely as a means of circumventing laws, or rulings like Martin v. Boise.

The risk identified is that these can be used primarily as a means to hit certain metrics so cities can then sweep encampments without actually helping people. That’s a real risk, and I hope we’re able to build up our infrastructure of new housing development, treatment centers, tiny homes, all of these things all at once.

Rachel Cohen

It didn’t seem like the elite mental health treatment centers could necessarily take unhoused or low-income patients, even if they somehow had subsidies to fund that kind of care.

Neil Gong

Yeah, I think the clearest way we can see this is that at one point an insurance company had a test case where they sent a homeless woman diagnosed with schizophrenia and opioid addiction to one of these elite clinics, and the center just couldn’t figure out what to do with her. They couldn’t figure out how to house her because the insurance companies won’t pay for non-medical housing. They couldn’t do their therapeutic procedures because they had to take her to court appointments, which ate up all their time. And while the woman’s family was paying for insurance, they weren’t involved in the broader sense of coordinating care or having home-based interventions. The kind of model these elite clinics employ doesn’t really work if you don’t have that baseline of stable housing and involved family.

Rachel Cohen

Your book looks at some of the pitfalls of the affluent private care model, too. What did you find there?

Neil Gong

One of my findings is that in the public safety net, they’ve ended up with tolerant containment because they essentially don’t have the capacity to surveil and control people. And actually where you do find this kind of surveillance and control is in these centers for privileged people. They don’t necessarily think of it as surveillance or control in this bad way. It’s more like, well, families are paying all this money to have their loved ones looked after. But from the patient perspective, it can be kind of overbearing, and you know, it’s ironic because a lot of social theorists and critical theorists working in the tradition of Michel Foucault predict that the state is going to micromanage these poor people who are social deviants. But again, as I’ve said, they basically can’t, and don’t have the resources to do so. And so where you see the micromanaging is in these elite centers, and sometimes that treatment looks really good and people appreciate it, but in other cases, they feel dominated, frankly.

Rachel Cohen

We are seeing this resurgent debate around forced treatment and whether there should be a return to institutionalization for people with severe mental illness. In your book, you push back on the premise and say there are things we need to answer first before we can get to the question about returning to asylums.

Neil Gong

I think it can be true that there’s some small sliver of people who will require long-term inpatient care or perhaps even life-long care. I think that can be true at the same time that a lot of people we currently think need that actually don’t. And so I’d say the first step is addressing housing needs and access to high-quality community care. If we build all of that out, there will be a lot of people who we probably thought needed a conservatorship but with all these high-quality voluntary services, actually will be able to get the care they need without giving up their rights.

At the same time, after we’ve done that, I think there will still be a sliver of people who are going to need this kind of long-term involuntary care, but instead of kind of disappearing a mass of people via old school asylum tactics, we will really have winnowed it down until there’s only a very small percentage of folks who we truly have figured out we cannot serve well through voluntary community services. I think there is a role for the asylum after we’ve done all those other things right.

Rachel Cohen

Do you have any ideas on how to avoid the pitfalls of the past when it comes to asylums?

Neil Gong

I think one really important move is to bring more care workers who have personal experience with mental illness and have been through experiences like conservatorships so they can help identify what parts work better than others. Could we redesign psychiatric wards with the input of architects who have themselves been hospitalized?

And then there’s the procedural justice question, which is like, even if people are going to have to go to court and lose their case and be conserved, there’s still a need to make sure that their voices are heard so that it’s actually legitimate and not a kind of kangaroo court situation. Because when people are treated with dignity, it makes a huge difference. So I think that ideally we’ll have far fewer people who even seem to be candidates for asylum or institutionalization. But among those who do, there’s both a design issue and procedural justice issues where we can make huge improvements.