skip to Main Content

In 1990 11% of homeless persons were older than 50.  Today half are over age 50.  Today we talk with Margot Kushel about how we got here, including:

  • That sense of powerlessness as a clinician when you “fix up” a patient in the hospital, only to discharge them to the street knowing things will fall apart.
  • Chronic vs acute homelessness
  • What is the major driver of homelessness in general?  
  • What is the major driver of the increase in older homeless persons?
  • Why do we say “over 50” is “older” for homeless persons, why not 65? 
  • To what extent is the rise of tech in San Francisco to blame for our local rise in homelessness? 
  • What are the structural factors and individual factors that contribute to homelessness?
  • How has the history of redlining and the federal tax subsidy of wealthy (mostly white) people in the form of a mortgage interest deduction contributed to racial inequalities in homelessness?
  • What can we do about it?  What are the highest yield interventions and policy changes?
  • What should we call it – homeless or unhoused?

We were fortunate to make it to the end of this podcast before Margot lost power.  It’s storming again in the Bay Area at the time we record this.  So much harder for the older homeless people on the streets with no power to lose.  A mad world out there (song hint).

Key references:

Margot Kushel’s UCSF Grand Rounds 

JAMA IM paper on mortality among older homeless persons

NEJM perspective arguing that interventions to address homelessness shouldn’t be evaluated on cost savings.


Eric: Welcome to the GeriPal podcast. This is Eric Widera.

Alex: This is Alex Smith.

Eric: Alex, who do we have with us today?

Alex: Today we are delighted to welcome Margot Kushel, who is a general internist and professor of medicine at UCSF, and she’s the director of the Benioff Homelessness and Housing Initiative. Welcome to the GeriPal podcast, Margot.

Margot: Thanks for having me.

Eric: We’re going to be talking with Margot about homelessness in later life, but before we do, Margot, I think you’ve got a song request for Alex.

Margot: Oh, yeah. I chose Curt Smith’s Mad World, because it kind of is a mad world. Then when I saw that acoustic version with his daughter at the beginning of the pandemic, it just really got me, and I’ve listened to it hundreds of times, much to my young adult children’s delight and hysteria.

Eric: I love this. We’ve had it play before, but it also comes from my favorite movie, Donnie Darko. You a Donnie Darko fan, Margot?

Margot: I am. I am. You just gave me the idea to show it to my kids, so I don’t think they’ve seen it.

Alex: I still haven’t seen it, despite it being plugged on this podcast multiple times. I’ll have to remedy that. Here’s a little bit of the song.


Eric: Love it.

Alex: Love it.

Eric: Every single time I hear that song, love it.

Alex: It’s a great song.

Margot: So great.

Eric: Well, let’s jump right into the subject. I’m going to start off by asking you, Margot, how did you get interested in homelessness from an academic perspective or even a personal perspective?

Margot: Yeah. Yeah, funny story. When I went through my training, I first thought I was going to work at a health department. That was my goal, and I thought the issues that I was most driven by were issues of the working poor. I had been involved in some union work and such, and then I did my residency primarily based at San Francisco General. I was in the part of the UCSF residency program that by choice spent most of our time at the General. It was in the mid-nineties, and about half of the patients that we cared for in the inpatient service were homeless. We would admit them to the hospital. We would drill down on the tenth differential diagnosis of medium vessel vasculitis, and then we would discharge them back out to the streets.

As an intern particularly, you were the one. You have to walk into the room and say to the patient, “You’ve got to go. They’re telling me you have to go,” and the patients would beg not to leave the hospital. Who doesn’t want to leave the hospital? But when you’re faced with going out into the cold and rainy … Then we would discharge folks, and then lo and behold, two or three days later, they would be back, in worse shape than they were in the first place.

Now I would have words for it with moral injury. It felt like a moral injury. What was I doing? Why was I spending all this time figuring out the detailed differential diagnosis of these obscure diseases when we weren’t actually attending to what people really needed? So I switched my focus. I guess I could say I’m not doing a very good job, because here it is. Homelessness is just increasing and increasing. But it really just became my life’s work to say, “We’ve got to as physicians step up and say, ‘We can’t do our job if people are homeless,'” and we need to have a voice in the solutions.

Eric: I want to get to the question about your role as a researcher versus an advocate, but before we do, what have you noticed, I guess both as a clinician and then also as a researcher, as far as the demographic changes that we’ve seen with homelessness in the time since you were a resident?

Margot: Yeah, I’ve been working on this issue. I feel like I’ve aged along with my work on this issue, and certainly when I started out really studying it and speaking about it, that was more the late ’90s, early 2000s, and I gave a ton of talks of, “Homelessness is a problem that happens to people who are young, middle-aged adults, like 25 to 44.” It was in my slide deck. I would talk about this, except I was a clinician, and I’m seeing these folks in the hospital and in clinic. It’s like, “Wait, they’re not 25 to 44. Everyone I’m seeing seems to be 50 and up.” I thought, “Am I missing it? Maybe because older people are sicker, maybe I’m seeing them, or is this real?”

I was talking to a colleague, Judy Hahn. She’s an epidemiologist, and we were just riffing over coffee about this. She, I think, said, “We could look at that. We could look at whether this is real or not.” We had been part of a group that had done a series of studies in San Francisco since the early ’90s, since before, actually, either of us were involved, where they would recruit a bunch of homeless folks. Then they were actually trying to test them for HIV and see who got into a longitudinal cohort. But they would ask a bunch of questions. So we realized that we had a series of cross-sectional studies in San Francisco from the early 1990s to 2003. We looked at it and realized that in the early 1990s, 11% of folks were 50 and over, and by 2003, 37% were. I’ll tell you the punchline, that now almost half are, and we thought, “This is”-

Eric: Half are older than?

Margot: 50 and older. Yeah, half are older.

Eric: 11% and-

Margot: From 11%. Yeah, and the numbers are a little iffy. But we looked statewide recently in a big study that’s coming out, and I can tell you that 49% of people were 50 and older. So it’s a big issue, and we went to publish it. We thought, “This is huge.” We sent it to JAMA, and it got desk rejected from about ten journals. We sent it we to JGIM, which is a great journal, and they said, “Well, it’s not worth 3,000 words, if you can get it into a brief report,” and we’re like, “Oh, it’s been desk rejected from everyone.” So we got into a brief report, published it. Nine months later, it comes out, banner headline at the San Francisco Chronicle, really became flying to DC to brief Congress about it or whatever, from a paper that nobody wanted to accept. Yeah.

Eric: I’ve got a question.

Margot: Yeah.

Eric: What do you know about the older than 65 group?

Margot: Yeah, great question. So first of all, I think when we started, I’ll tell you it took me three tries over six years to get the National Institute of Aging to fund more research on this, which they then did, which I’ve learned so much from, and I’m super grateful. But it took me a lot of time, because they were like, “Well, 50 isn’t old.” But we had a hunch that turned out to be right that by the time these folks were in their fifties, they really had all the geriatric conditions and things we associate with much older.

65 and up was low, but it’s rising. It is the fastest rising group in the country, and some colleagues and demographers did work in three big cities around the country, New York, LA, and Boston. They found that between 2017 and 2030, they think the percentage of the population that’s 65 and up is going to triple in that 15-year period.

Eric: Oh, wow.

Margot: Yeah, and then it’s going to start coming down, because it’s a demographic effect. So it’s still overwhelmingly the older adults are overwhelmingly 50 to 64, unfortunately because of when folks die. But there is an increasing proportion who are 65 and older, and in our statewide study, the oldest person was 89. Yeah, not uncommon. We see people who are really old, old-old.

Eric: So you actually just published a … I think just. I can’t remember when you published the JAMA IM study that showed that median age of death for adults over the 50 and older who are experiencing homelessness is 64, 65 years of age.

Margot: 64. Exactly. Exactly. So if you look at homeless folks overall, the median age of death is before 50. But if you look at those who are 50 and older and homeless, we’ve been following this cohort now for about a decade, where everyone had to be 15 older, everyone had to be homeless when they were admitted, and now the deaths have continued since we published that paper. Now more than a third of the original cohort has died, and the median age of death is about 64.

Eric: Margot, the nice thing is we have an editor from JAGS on this podcast right now with us. Yeah, this is a geriatrics journal, and generally, if people aren’t over the age of 65, JAGS may not look at it as strongly. Would that be fair to say, Alex?

Alex: Our editor-in-chief, Joe Ouslander, likes to say that if the mean age in the study is less than his age, then he generally doesn’t consider it to be …

Eric: Margot, make a case why that’s a wrong policy here.

Margot: One of my favorite aphorisms in geriatrics … I’m not a geriatrician, but I’m a wannabe geriatrician, and one of my favorite aphorisms is age is just a number, right? I think that that is often said by geriatricians of what we should be looking at as people’s function, that there are 79-year-olds who are climbing Mount Everest, and they should be treated differently, right? Than someone who is perhaps dependent on all their ADLs. In terms of their risk, in terms of how we think of preventive health, as we think of screening, all of those things, I would turn around to the geriatricians and say, “Check your implicit bias. Check your Eurocentric, middle class thinking. Age is just a number,” and that for people who have lived their lives encountering structural oppression, racism, homophobia, who have lived their lives in deep poverty, they age earlier, and tragically, they don’t live long enough.

You can see all of the ways that our society is not geared towards the lived experience of people who’ve encountered structural oppression. One of the worst ways is you have a community of people who’ve paid into Social Security their whole life, paid into to support Medicare their whole life, and don’t live long enough to see the fruits of that. But I have had this fight, as it were, with the NIA and with JAGS over the time. You can’t say age is just a number when what you’re trying to say is not every 72-year-old is the same and then have an arbitrary cutoff. I think what we mean by geriatrics, what I love about geriatrics is that it is a side of medicine that focuses on function and people’s lived experience, and it is realistic about what is people’s life expectancy, what is people’s function, and how do we treat them according to that? Live your values. In people who face their lives faced by structural oppression, all of those things are shifted back 15 years earlier.

Eric: This sounds like a great JAGS editorial.

Margot: Oh, yeah, I’m all over it. I’d be happy to write it. It’s funny, because I would say JAGS is one of those journals I love publishing in. We get a lot of desk rejects like, “They’re not old enough,” and other people are like, “Oh my God. This is what we’re about. You’re talking about a group of people, functional disabilities, cognitive impairment, urinary incontinence, the whole shebang.”

Alex: Right, and I would say that for those listening and considering sending your work to JAGS, strongly consider if your work is on, say, incarcerated older adults, homeless older adults, we have an ethno-geriatric subsection.

Margot: Yes.

Alex: But the longer title’s Ethno-Geriatrics and Special Populations. That section is edited by people who understand this, not that other editors absolutely don’t understand it, but just that you probably will have a more consistent review, so I urge you to submit to that section. The journal section’s requirements are the same as for a clinical investigation. That was a stirring defense of that approach. Really appreciate that. You did mention structural factors, and we saw you give a terrific grand rounds at UCSF, which is like, “Oh, we need to get her on the podcast.” On that grand rounds, you differentiated structural factors from … I forget what the word was, individual-

Margot: Individual risk factors. Yeah.

Alex: I wonder if you could talk about that conceptually and then maybe in more specifics, too.

Margot: Yeah, It’s interesting that the general public … and I would say actually, there’s been polling done. I would say particularly white voters or white folks will say this. When they think about homelessness, their mind goes to what I would call the individual risk factors, things like having a substance use disability or mental health disability. I think about homelessness as really an interaction, and this is a framework put out in, gosh, the early to mid-eighties by Marty Burt and others, interaction between structural factors. So structural factors are like the availability of affordable housing across a population, the wages for low wage workers, racism. Those are structural factors. So it’s really an interaction between the structural factors in a society, the individual risk factors, so things like whether people have substance use, mental health problems, incarceration histories, et cetera, and the presence or absence of a safety net. The worse the structural factors are, the fewer individual factors you need to become homeless.

So you go to Northern Europe, there are certainly people who are homeless, but those are relatively flat income structures, lots of social safety nets and things. But the folks who are homeless are much more likely to have a really high burden of these individual risk factors. In the United States in 2023, the truth of the matter is you don’t need very many individual risk factors to become homeless. Now, if you have those, you’re at higher risk, and this is the other confusion that people have. There’s a really popular and completely wrong ideological thing that, “Oh, California has so many homeless people because we’re libertine about drugs and alcohol,” or whatever people think about us. Not true, actually. I don’t know what you think about we’re the libertine about those things, but if you look at the rates of homelessness in a community, the rates of homelessness in a community in the United States, the difference in rates is entirely explainable.

It’s a beautiful, straight, linear straight line by the disconnect between the number of extremely low income households. Those are households that make less than 30% of the median income of the area, and the availability of housing that those folks afford. It is completely disconnected, like a scatter plot, between things like rates of substance use and mental health. So if you look across the country, parts of the country that actually perform the worst on tracking that we use on mental health or substance use rates tend to be the former Rust Belt, Appalachia, and things. They actually have much higher rates of substance use and mental health problems than California, which actually scores low, but their housing is cheap. So yes, they have a lot of substance use and mental health problems, but folks aren’t homeless. The places where homelessness is high, California and New York, et cetera, Hawaii, those are the places that have the highest housing costs. They also generally have the lowest rates of substance use and mental health problems.

Eric: Is there a difference, though, if you’re chronically homeless and you age into late life homelessness versus your first instance of homelessness is during late life?

Margot: After 50. Yeah, for sure. So the other part of that thing I would say is all of those things are true about the difference in rates between a community. But for an individual, if you are an individual with those risk factors, your personal chance of being homeless is obviously much higher. You’re going to have a harder time competing in a tight housing market. We see this playing out in older adults. In our study, which we call the HOPE HOME Study, which is funded by the NIA longitudinal cohort, which is entering its third five-year period right now, you had to be 50 and older and homeless when you were enrolled. We found that about half, it was 44%, had never been homeless, not even one time in their life before they were 50. So everyone in the study was 50 and homeless, but almost half of them were homeless for the first time in their life after the age of 50.

So the first thing that we found, and a lot of this work was done with the fabulous Rebecca Brown, a geriatrician who used to be at UCSF and now at Penn, the first thing that we found is that those whose first homelessness was after the age of 50 were very different from those whose first homelessness was before the age of 50. Obviously, this is in aggregate individual. Individual may vary, but basically, the folks who were first homeless before 50 and remained homeless after 50 or became homeless before 50 and then became homeless again after 50, those were folks who had tons of individual risk factors, lots of them. We did a lot of life course analysis. Lots of really early life trauma, lots of substance use and mental health problems that started early in life. Lots of entanglement long-term with the criminal justice system. Their lives had never really gotten off the grounds after really terrible childhoods, and their lives never really came back.

The folks who were first homeless after 50 had incredibly different life stories. They generally led what could be called “typical lives,” deeply poor, but typical. They were working usually more than one job. These were usually very physically demanding, low paid, often usually non-union work, that they worked their whole lives and were hanging on by their fingernails. They didn’t have particularly high rates of substance use or mental health problems. They were married, all of the things. They went to church on Sunday, all of the things. But they were poor, really poor. Sometime after the age of 50, we could identify an event where the folks who were first homeless before 50, there wasn’t a clear crisis that precipitated it. Folks who were first homeless after 50, there was an event, and usually those events could be described in one of four categories.

They got sick. Their spouse or partner got sick. Their marriage broke up. They had a job loss, or their spouse or partner died or their mom died. Lots of folks who they were 52, living with Mom, who was 78. Mom passes. Their name isn’t on the lease or the mortgage. They lose their house, and they wind up outdoors. I think one of the really horrifying things that we’ve found, there’s a lot that’s horrifying, is even though those folks had a much lower burden of mental health or substance use problems, when we looked at mortality over the study, and the mortality rates are shocking and high, they had about twice the odds of dying than the folks who were first homeless after 50.

The other thing that we found was … because we followed people longitudinally, so we found them in and out of homelessness. I like to say homelessness is a state and not a trait. It’s not an intrinsic quality of a person. It is an experience that they go through. But what we found was that people who became housed, no matter how they became housed, had a dramatic decrease in their risk of death. So when we looked at the things that increased your risk of death, being homeless first time after 50, associated with a dramatically higher risk of death during the study period, and becoming housed at any point in the study period dramatically lowered your risk of death. Now, this might seem obvious, but we actually thought we might find the opposite, because one of the ways that homeless folks get housing is they get prioritized through a system that is designed to prioritize the highest-risk folk. So we could imagine that the highest-risk folks were getting housed, and then they might die because they were high-risk. But even with that, even with that system, being housed seemed to be extremely protective.

Eric: Is that in both groups, the chronically homeless and the-

Margot: Yes, in both groups. Yeah, that was adjusting for the age of first homeless. So really interesting. The folks who became homeless late in life became chronically homeless, because chronic homelessness is by definition being homeless for more than a year, more or less, and having a disabled condition. They generally had a disability. Their disability was geriatric conditions. But I thought first of all, the fact that so many folks are falling into homelessness for the first time in late life is really important policy implications for society as a whole, for healthcare providers, because so many of the crises were things that we would know about, but also that they are not being prioritized for housing because they don’t have all of these other burdens, and yet they’re dying even faster.

Eric: Do you feel like the interventions that we do for homelessness should take that into account, these potentially two different groups?

Margot: Yeah, absolutely. What I like to say is that everybody needs a home that they can afford, particularly in these high-cost regions and particularly for older adults who are not going to dramatically increase their income, right? They’re not going to get retrained and get a job that’s going to pay a lot more. They’re on fixed incomes. That housing is going to need to be subsidized. But for someone who has first homeless after 50, first of all, we should be identifying them and preventing the onset of homelessness. But the second thing is they don’t need a million behavioral health treatments, right? They don’t need substance use treatment and mental health treatment. That’s not why they’re homeless. They need housing. They need help accessing housing. They need a voucher to pay for it. They need help navigating a system that is biased against people, discriminates against [inaudible 00:24:27] vouchers, et cetera, and then they need mainstream care.

I would love to get the geriatrics community to say, “Age is just a number.” I take care of a lot of patients where I’m like, “Gosh, this person needs a geriatrician, but they’re 52.” But they’ve got all the geriatric conditions. So they need mainstream healthcare, and they need housing support. But they don’t need all the other support. For someone who had horrendous childhood trauma, was in prison for 30 years, has a really bad substance use problem, there is an intervention that works incredibly well for this population. It’s called permanent supportive housing, but it’s based on housing supports plus voluntary, but robust social supports to help keep that person housed.

Eric: I want to talk about permanent supportive housing. I also want to-

Alex: So before we go on with interventions, can I just go take a step back?

Margot: Yeah.

Eric: I was going to take a step back, too. Go ahead. We’re now taking a step back.

Margot: We’re all stepping back.

Alex: We all want to step back. My step back is I want to know, just going back to the rise in the proportion of homeless who are over age 50, from 11% to 50%-

Margot: Yeah. Yeah, what’s going on?

Alex: … when you think about the factors that are leading into that, what is the explanation? Is this 90% rising cost of housing, as far as those structural individual factors? [inaudible 00:25:53] it’s almost all structural and a huge proportion of it’s housing, but tell me.

Margot: Yeah, yeah, yeah. So first of all, there’s this generational effect. There seems to be this, that folks who were born in the second half of the Baby Boom, I just missed it by about two years, but 55 to 65, those folks have had bad luck through their whole life. We know people born in the second half of population booms often do worse in job markets, because a lot of the jobs are taken by the big population bubble in the first half. So that is a truism, but also, they had particular bad luck. This was a group of people who entered adulthood during a recession. We know people who enter adulthood during a recession never make up the lost income. They entered adulthood in a time where unions and the power of unions decreased, and so they were much less likely to get pensions and other supports.

They entered adulthood during the beginning of a period of mass incarceration, so a lot of them, particularly people of color, got … We should talk about that and the way racism plays into all of this, but got caught up in the criminal justice system, and most importantly, but most quietly, they entered adulthood right at the cusp in the early eighties, when there was a massive, very quiet, but massive change in federal policy around supporting housing, around housing subsidies that is at the root of our entire homelessness problem.

So why that population? So actually, if you look at it over time, when that group of people were in their twenties and thirties, the big group of people who were homeless in their twenties and thirties, it’s like following the mouse through the snake, right? There’s like a bubble that you can see, and as that population ages, that homeless population ages. When they start to die off, we’re going to see a decrease in the age again. But the answer to homelessness is it’s all about this disconnect between incomes and housing costs and the lack of subsidized housing. That explains the whole story. There’s nothing else. People are like, “Oh, the closure of the mental hospitals, this or that.” It’s all about that.

Eric: I’ve seen you quote, “The fundamental cause of homelessness is poverty.”

Margot: Is poverty, but more so than poverty. It is poverty, but it is the lack of housing, because, for instance, when you look in regions in the country that are very impoverished, but their housing costs are similar, they have less homelessness. The reason homelessness is so bad in these blue states, it’s like Hawaii, New York, and California have the highest rates, per capita rates of homelessness. It’s because of this disconnect, because of the inequality and the lack of housing.

Eric: Where does race fit into all of this?

Margot: Yeah. So everyone screams and yells about mental health, but the real increased risk of homelessness is exposure to racism. So remember, I saw a talk by Nikole Hannah-Jones who said the North and the West didn’t need Jim Crow. They had housing policy, that one of the ways, one of the many ways that racism has played out in our country has been through housing policy. If you look, for instance, at the main way Americans build intergenerational wealth, it’s through home ownership. That is the main way that we pass money on to the next generation. In the post-World War II era, there was this huge explosion of home ownership. That was by design. That was by federal policy, where they wanted to have people own a home as a way to stabilize communities.

Black Americans, Indigenous Americans were literally shut out of this, not just because of individual people being racist, but because of the federal policy. It was unethical, immoral, racist, but totally legal. So for instance, areas had racial covenants. So you could say in your neighborhood, “You can only sell to someone who’s white. You can’t sell to someone who’s Black,” and so areas became basically racialized, right? Where there were some communities that Black folks had to live and other communities where only white folks could live in. Then the banks with federal support for mortgages got into the picture and basically said, “We will not guarantee mortgages. We will not give you mortgages if you live in these areas that the Black folks have to live in.” So Black folks and Indigenous folks were literally shut out of the post-World War II increase in home ownerships.

So first of all, we still basically live according to … If you didn’t see a map … That became illegal with the fair housing laws, but we still live in those neighborhoods. If you had to guess where those areas were redlined, you could look at where people live today and guess with 98% accuracy, because our areas are still effectively redlined, because we still are living in those. But this did several things, and I’ll just use the Black community as an example. Black folks that face racism in so many aspects of their life, so paid less for the same work, et cetera, et cetera. But one of the biggest things is that they have much less intergenerational wealth. Their parents are less able to pass down wealth to their children. So there’s a difference in income, but an even bigger difference in wealth. A huge portion of that difference in wealth you can trace back to the fact that their families were not able to get into the housing market.

But the other thing is if you own a house, you have a tremendous ability to protect your beloveds from homelessness, because if you own a house and someone’s going through a hard time, they can come and sleep in your living room. No one’s going to care or know or have anything to say. If you are a renter, you actually don’t have that ability. So not only does generational wealth help, right? You can loan someone money. You can help them pay for college. You can do all these things, but also quite literally, you can’t bring people into their homes. So there’s this myth out there like, “Oh, homeless folks are just disconnected from their friends or family, and if only we connected them by Facebook, all would be good.” That is a myth.

White folks who are homeless are often folks who have cut ties with their families for a variety of reasons. But if you look, for instance, in the Black community, when we look at Black folks who are homeless, they are very much in touch with their families, very, very much. It’s just that their families are fighting the same upstream forces. The families are hanging on by dear life to their apartments that they’re being gentrified out of, that their landlords are looking for an excuse to kick them out of, and they can’t just take in a family member, or if they do, they’ve taken in a bunch, and they don’t have it.

So you add to all of the housing discrimination. By the way, now there are fair housing laws, but they’re not well enforced. So there’s really good evidence to show that if a Black person shows up to rent an apartment with the same housing resume as a white person, they’re less likely to get that apartment. So they wind up paying more for lower-quality housing. You add to that employment discrimination, criminal justice discrimination, and of course educational system discrimination, and of course, we pay for our public schools based on local taxes. So all these formerly redlined areas, like white areas, have much “better schools.” They’ve got a lot more local tax money to pay for their schools. It’s not a surprise that Black Americans are about three to fourfold overrepresented in the homeless population. Indigenous folks are eightfold overrepresented.

Alex: Yeah. We have to get to interventions. This is such [inaudible 00:33:19].

Margot: Yes. Otherwise it’s too depressing. Yeah.

Alex: We should’ve scheduled two hours, and maybe we’ll do a followup. Maybe we’ll have Rebecca Brown on the followup. We’ll figure this out. But one more question that maybe is a way to lead into it. You are the director of the Benioff Homelessness and Housing Initiative, and you work at Chan Zuckerberg, San Francisco General Hospital. These are people who started Salesforce, Mark Zuckerberg, and-

Margot: Yeah. Salesforce is Marc Benioff. Yeah.

Alex: Oh, yeah. Salesforce, Marc Benioff. Sorry. Yes, and Facebook, Mark Zuckerberg. I think his wife may work at San Francisco General. But to what extent are they addressing problems of their own creation? To what extent should we lay the blame of this, a lack of affordable housing, on gentrification due to explosion of tech in the Bay Area as a case example?

Margot: Yeah, I wouldn’t personalize it, maybe, although obviously, the fact that we have tax structures and stuff that allow us to have such wealth inequality is a big problem here, is the problem here, and the fact that the Bay Area, there was about a decade period where they added about ten new jobs for every new housing start. California is notoriously a very hard place to create housing, any housing. To solve the homelessness problem, we need housing that is deeply, deeply, deeply subsidized. But in the Bay Area, there’s no housing for anybody, right? So it’s really interesting. I was giving a community talk last week, and everyone has this myth. Not everyone, but some people have this myth that homeless folks are rushing into California. Why would you rush into a place where there’s … Why? No, it’s not true. It’s literally not true. It’s just false. I just want to put it out there, false.

But people have this mentality people are rushing here to be homeless. People are not rushing here to be homeless. But for a very long time, people were rushing here to live here, and we didn’t build the housing. That creates compression at every level. So you have doctors living in quality of housing that someone who makes less than they might make would live in another community, let’s say public servants, teachers, and you have teachers barely able to afford any housing. So they’re living in housing that really low income workers would live, and the really low income folks are forced out into the street. We have no housing. We need both a huge increase in subsidized housing, but frankly, we need housing. We need housing.

I was giving this community talk a week or two ago, and there’s a really poignant question from a guy who was a tech worker who said, “I didn’t grow up here. I came here to be in tech. Am I the problem? Am I the reason?” I was like, “Okay, dude, the fact that you recognize that it’s not you literally, but yes, the people rushing in, what did the Bay Area think when they added ten new jobs for every new unit of housing?”

Eric: Well, is it tech that’s the problem, or is it the fact that we just don’t-

Margot: Can’t build housing.

Eric: … build enough housing-

Margot: Right. Right.

Eric: … for multiple reasons?

Margot: Exactly. I think it’s great that we’re a place that people want to live. That’s great. That’s why we all want to live here, right? It’s great that we have jobs. That’s great. Having jobs is a good thing, right? Having jobs that pay well is not the problem. There’s a problem with tax structure, which isn’t … I will give Marc Benioff preps for Prop C, where he basically came out really in favor of taxing himself and his company a lot, that proposition to support. It was for those not from the Bay Area, a local proposition to support homeless services that was designed to be a tax on tech companies. It was literally designed that way, and he came out and basically came out, much to the displeasure of many of his colleagues, strongly in favor of it and threw a lot of resources into the campaign and got it passed to basically tax his company a lot.

I’m a big fan of taxes, right? We need taxes. But I think that there’s a very specific problem within the general national context of rising inequality because of really messed-up tax policies. But also in California, we have not kept up with the housing. I also like to say I often get these talks and say, “I live in subsidized housing,” which is true. Now, you guys can see me. The viewers probably can’t see me. But I live in a beautiful home, whatever. They’re like, “Well, that’s weird.” But my husband’s an attorney. We own our home, and anyone who owns their home in this country has a massive federal subsidy in the form of the mortgage interest deduction. That was done because it was viewed that home ownership was a good thing that would stabilize communities. In what moral universe is a federal government deeply subsidizing my housing and not subsidizing that of my patients? I want to just add the racialized angle in here that it’s awfully coincidental, right? I say with sarcasm, but that the government decided to subsidize home ownership and simultaneously designed a system where Black folks couldn’t be homeowners.

Eric: Well, let’s talk about some targeted intervention for late life homelessness. What are your tops?

Margot: My tops are, one, housing subsidies. Well, I’ll start from the real top is prevention, right? We know that there’s a huge inflow into homelessness, and we know that folks are entering homelessness in huge numbers in late life. Prevention is hard. For it to work, it needs to be targeted, meaning there are a lot of people who have the general risk factors, but only some of them become homeless. But we’ve learned a lot about what those risk factors are. We need to be identifying those folks at highest risk and getting money, services to them. Sometimes it’s a one-time thing, right? Sometimes you need to get people to pay back their back rent and deal with their fight with their landlord to keep them where they are. Sometimes it’s more of an ongoing subsidy.

The nice thing about older folks is we often know what their incomes are. There’s often both a floor and a ceiling, because let’s say they have a pension or they have government benefits. We know what their income is. We need to make up the difference between that so that they’re only paying 30% of their income on rent. We need to identify who’s at highest risk of losing housing and get to them quickly before they lose it, because once they lose it, particularly in parts of the country like we have with rent control, once they lose it, they can’t reenter the market, right? Because it’s much more expensive to get back into the market. So one, prevention.

Two, subsidies. There’s no way around it. If you make $925 a month, $1,000 a month, you can’t pay $1,100 a month for your housing. You literally can’t, right? You can generally … if you’re that poor can pay about a third of it. So we need subsidies, and then there are the issues of supports. So for folks, let’s say, with the early life homelessness who are still homeless, who have been out there, living homeless for many years with lots of behavioral health problems, there’s this incredible intervention called permanent supportive housing. That’s a fancy way, and it’s done on what’s called Housing First. All housing should be on Housing First, because the choice is housing first or housing last. We all know as physicians if someone’s homeless, trying to talk to them about any other problem doesn’t really get very far, right? You’ve got to get them housed first, and then you talk about everything else.

So permanent supportive housing basically just says subsidized housing. So it’s subsidized with either onsite or closely linked supportive services that meet the need of the individual. So someone who’s got, let’s say, a psychotic disorder and a substance use disorder, those services are going to be pretty intense case management to help them just manage their life. It’s going to be having easy availability of mental health supports and substance use supports and the like. What I think we’re starting to see, though, is a lot of the older folks, they use less substances, but even whether they’ve been homeless early or late, they’re actually dealing with geriatric conditions. A lot of them need what we would call in the middle class world assisted living, right? They might need a couple hours a day of personal care. They need their units to be accessible. They need them to have grab bars and good lighting and community and efforts to address loneliness and social isolation. It’s not rocket science here, but we just need to match the services. So they might need less substance use and mental health things, but they might need assisted living level services in permanent supportive housing.

Eric: You’re a researcher, Margot. What do we know about the outcomes of permanent supportive housing?

Margot: Permanent supportive housing works, hands down. There’s been so much … It’s actually organized disagreement. It’s not just random. It looks like it’s random people. Actually, now they’ve found who’s funding these campaigns against it. I will say first thing, permanent supportive housing, all props to the radical lefty administration of George W. Bush, which was actually the administration who discovered, trumpeted-

Eric: Yeah. Well-know progressive.

Margot: Well-known progressive radical leftist George W. Bush. But I’m a believer in giving credit where credit’s due, and it was a Republican, deeply Republican, deeply conservative administration that was like, “Wait. This is cheaper. This is better. This works. We’re all about outcomes,” and they trumpeted it. Then Obama administration wrote it into law that it became the federal policy that it should be done. But we know that even in populations with the worst-case scenario, permanent supportive housing works really well. We did this really interesting study in Santa Clara. Median age was 48, because it wasn’t oriented only towards older folks, but still, where we basically found the 400 most chaotic folks in Santa Clara, highest rates of jail use, psychiatric emergency department use, psychosis, all the things, right? We basically met them. We didn’t even wait for them to come in. We met them at the steps of the jailhouse or in the ER after being up all night and said, “Hey, do you want to fill out a million pieces of paper for a 50/50 chance of getting housing?,” because we only had a certain number of housing things.

We basically approached 425 people, and three people said no. Remember, this is like … I wouldn’t have signed 30 pieces of paper after being up all night in jail, right? Everyone was like, “50/50 chance? I’ll take it.” These are the people that the public would believe don’t want housing. They want housing, right? They said yes. As soon as they said yes, we randomized them on the spot. If they were in, we counted them as in. They were counted as in housing from that moment on. We found, and we followed them for seven years, that by the time we finished, over 90% of these folks who we met at the jailhouse doors or the door of the psych ER got housed, and they stayed housed over the seven years for well over 90% of their nights on average. So they basically stayed housed.

Eric: So I’ve got a question for you. I remember deep in the recesses, 2016, I think it was, a New England Journal piece came out saying that we need to reframe the debate about cost.

Margot: Yeah. That was me. Yeah.

Eric: That was you?

Margot: That was me.

Eric: You’re the author of it. Got that.

Margot: Yeah, yeah, yeah.

Eric: Maybe permanent supportive housing doesn’t actually-

Margot: Save health dollars.

Eric: … save health dollars.

Margot: Yeah. Right.

Eric: But we should still do it.

Margot: Right. Let’s be real here. What medical intervention saves health dollars, costs less than not doing it? Maybe vaccines, maybe. HPV vaccine. It’s hard, actually, right? To think of things that actually, you pull money out of the system.

Eric: In truth, the amount of money that you’re seeing with these new drugs coming out, lecanemab, amylin antibodies-

Margot: Right. You could house-

Eric: … $2.3 million for some therapies coming out … I think the most expensive one is $3.6 million.

Margot: Right, right, right. Even in the high-cost region, permanent supportive housing, 25, 30 grand a year, right? Even in the highest-cost region. God knows giving people a housing subsidy … By the way, there are other costs that aren’t health costs that it would save. You know the amount of money we spend on shelter? Shelters, terrible shelters, 100 bucks a night. Good shelters, 300 bucks a night. It’s cheaper than that. How much are we spending cleaning up homeless encampments? How much are our businesses losing?

Eric: I forgot the amount San Francisco spends on a tent, but it’s-

Margot: On a tent. It’s insane, right? It’s really terrible. But if you’re looking at health costs, a few of us were sitting around at a conference, talking about this, and Stefan Kertesz wound up … I landed the plane. I went home to see my daughter’s track meet, and by the time I landed, he’s like, “I’ve transcribed the conversation you guys were having, and I think we should turn it in to the New England Journal.” Literally, that’s genesis of that. But I would ask you, why is everyone so obsessed with pulling money out of the healthcare system for a population that is mostly older people of color …

Eric: Because it’s not a drug. If it was a drug …

Margot: Right. If we had a drug that had these outcomes, these mortality outcomes, these health outcomes, we’d spend a lot of money on it. But it’s not a drug. But nothing in healthcare. We said, “Reframe the debate,” because what we saw, for instance, in this permanent housing trial is that people started to use primary care. They started to use outpatient mental health services. They started to use outpatient substance use services. This is awesome, right? They didn’t go to the psych ER again. Now, they didn’t shut down the psych ER, so did they really save money, because they didn’t go? I don’t know. The psych ER was still open, and they spent money. But they spent money on the good things. We spend money in healthcare. Yeah. It’s okay.

Eric: All right. For the viewers, Margot just lost power, but she’s back. We’ve edited everything out. I am going to ask my question. Margot, real quick, you gave a great UCSF grand rounds, and one of the questions at the Q&A is how do you balance being an advocate, being a neutral researcher, and how do you think about that?

Margot: Yeah, yeah. Look, I will admit that I’m an unabashed advocate for ending homelessness. I think that homelessness is a scourge in our society, and I’m happy to freely say I advocate for that. But I’m a researcher in that I have no horse in the race of how we do it. What I’m interested as a researcher is actually to go where the evidence takes us and to be a truth teller, to say, “This is what the truth says without politics, without advocacy.” I work often with advocacy organizations who will want to use the research to advocate for certain policies. I spend a lot of time, I spend pretty much every week now in Sacramento giving testimony to Senate and the assembly hearings. But I don’t go in as, “I’ve got an agenda” as much as, “Let me educate you. Let me tell you what we found,” and I believe that the truth will set us free, right? I do advocate for the end of homelessness.

Eric: We just lost Margot again. We are going to have links to Margot’s stuff on our GeriPal website, including her UCSF grand rounds. Really encourage everybody to read it. If she joins us, I’ve got one more question for her. But Alex, while we wait, do you want to give us a little bit more of Mad World? Because for those who are in the Bay Area right now, it is a mad world outside with the wind blowing and power going out.

Alex: The wind howling was coming through my microphone here. Here’s a little bit more of Mad World.


Eric: Ain’t that the truth, Margot?

Margot: That is the truth. Now I’m on my phone. Sorry about that.

Eric: Margot, I’ve got one last super fast lightning question. It’s probably not fast or lightning, because it probably brings up a thorny issue. We’re using the term homelessness. Is that the term we should be using? Is it unhoused? How do you think about that?

Margot: Yeah, I think language is evolving, and we constantly have this conversation. I tend to lean into the true experts on this, who are the people who are experiencing homelessness, and I will say that in our last polling, people are split. Some people say homeless doesn’t make sense because they have a home. Their home is the San Francisco Bay Area or California or whatever. What they don’t have is a house, and they would prefer houselessness. Other people feel like that’s just getting around the truth. It’s trying to take an ugly problem and put a pretty name on it. So I would say that this is going to change over time. Right now, I’m primarily using homelessness unless my community experts tell me they prefer houseless or houselessness.

Eric: I love going to the community, and I really love this entire podcast. Thank you very much for being on with us and for the amazing work that you’re doing. We’ll have all of the articles we talked about on our show notes. So thank you, Margot.

Margot: Thank you for having me, and sorry for all the technical problems.

Eric: Good luck with the power out there.

Margot: Thank you.

Eric: Thanks to all of our listeners. Have a great night.

Back To Top