International Journal of Drug Policy

International Journal of Drug Policy 25 (2014) 556–561

Contents lists available at ScienceDirect

International Journal of Drug Policy

j ourna l h omepa ge: www.elsev ier .com/ locate /drugpo

esearch paper

ingle room occupancy (SRO) hotels as mental health risk nvironments among impoverished women: The intersection of olicy, drug use, trauma, and urban space

elly R. Knighta,∗, Andrea M. Lopezb,c, Megan Comfortc, Martha Shumwayd, ennifer Cohene, Elise D. Rileyb

Department of Anthropology, History and Social Medicine, University of California, San Francisco, United States Department of Medicine, University of California, San Francisco, United States Urban Health Program, Research Triangle Institute International, United States Department of Psychiatry, Trauma Recovery Center, University of California, San Francisco, United States Department of Clinical Pharmacy, University of California, San Francisco, United States

r t i c l e i n f o

rticle history: eceived 8 May 2013 eceived in revised form 8 September 2013 ccepted 30 October 2013

eywords: uilt environment RO hotels omen

rauma ental health rug use thnography

a b s t r a c t

Background: Due to the significantly high levels of comorbid substance use and mental health diagnosis among urban poor populations, examining the intersection of drug policy and place requires a consid- eration of the role of housing in drug user mental health. In San Francisco, geographic boundedness and progressive health and housing polices have coalesced to make single room occupancy hotels (SROs) a key urban built environment used to house poor populations with co-occurring drug use and mental health issues. Unstably housed women who use illicit drugs have high rates of lifetime and current trauma, which manifests in disproportionately high rates of post-traumatic stress disorder (PTSD), anxiety, and depression when compared to stably housed women. Methods: We report data from a qualitative interview study (n = 30) and four years of ethnography conducted with housing policy makers and unstably housed women who use drugs and live in SROs. Results: Women in the study lived in a range of SRO built environments, from publicly funded, newly built SROs to privately owned, dilapidated buildings, which presented a rich opportunity for ethno- graphic comparison. Applying Rhodes et al.’s framework of socio-structural vulnerability, we explore how SROs can operate as “mental health risk environments” in which macro-structural factors (housing policies shaping the built environment) interact with meso-level factors (social relations within SROs) and micro-level, behavioral coping strategies to impact women’s mental health. The degree to which SRO built environments were “trauma-sensitive” at the macro level significantly influenced women’s

mental health at meso- and micro-levels. Women who were living in SROs which exacerbated fear and anxiety attempted, with limited success, to deploy strategies on the meso- and micro-level to manage their mental health symptoms. Conclusion: Study findings underscore the importance of housing polices which consider substance use in the context of current and cumulative trauma experiences in order to improve quality of life and mentald wo

health for unstably housentroduction

In the United States, the comorbidity of substance use and men-

al illness is a widely recognized phenomenon at a national level Conway, Compton, Stinson, & Grant, 2006; NIDA, 2008; Volkow, 004), specifically among the urban poor (Bassuk, Buckner, Perloff,∗ Corresponding author at: Department of Anthropology, History and Social edicine, University of California, San Francisco, 3333 California Street, Suite 485,

an Francisco, CA 94143, United States. Tel.: +1 415 867 8405. E-mail address: (K.R. Knight).

955-3959/$ – see front matter © 2013 Elsevier B.V. All rights reserved. ttp://

men. © 2013 Elsevier B.V. All rights reserved.

Shari, & Bassuk, 1998; Hien, Zimberg, Weisman, First, & Ackerman, 1997). Epidemiological studies underscore significant gender dif- ferences in the presentation of comorbidity, with women more likely than men to be diagnosed with affective and anxiety-related mental health disorders (Diflorio & Jones, 2010; NIDA, 2008). Esti- mates of depression and Post-Traumatic Stress Disorder (PTSD) are disproportionately higher among substance-using, unstably housed women than cohorts of housed women (Coughlin, 2011;

El-Bassel, Gilbert, Vinocur, Chang, & Wu, 2011; Nyamathi, Leake, & Gelberg, 2000). While research has shown that access to hous- ing may contribute in a significant way to a number of individual mental health outcomes (Baker & Douglas, 1990; Earls & Nelson,

urnal o

1 T s a p ( t

t ( 2 b H b J W M P ( 2 M s

r e c m I N z v t 2 i t D e t w F e c t s i b w

o o s c i b a o h e p u


s (

K.R. Knight et al. / International Jo

988; Hanrahan, Luchins, Savage, & Goldman, 2001; Nagy, Fisher, & essler, 1998), there is a need to understand how housing policies hape specific built environments, which in turn impact women t risk for poor mental health outcomes and substance abuse. This aper analyzes the role of place, specifically single room occupancy SRO) hotel rooms, in exacerbating and ameliorating negative men- al health outcomes for substance using, urban poor women.

Urban housing environments have received increasing atten- ion as sites that can both contribute to health and produce harm Freudenberg, Galea, & Vlahov, 2005; Northridge, Sclar, & Biswas, 003; Vlahov et al., 2007), and there is growing evidence linking the uilt environment to mental health (Evans, 2003; Frumkin, 2003; alpern, 1995; Parr, 2000). Contributing factors include neigh- orhood conditions (Cohen et al., 2003; Dalgard & Tambs, 1997; ohnson, Ladd, & Ludwig, 2002; Leventhal & Brooks-Gunn, 2000;

andersman & Nation, 1998), poor housing quality (Evans, Wells, & och, 2003; Freeman, 1984), crowding and lack of privacy (Baum &

aulus, 1987; Evans & Lepore, 1993; Wener & Keys, 1988), and noise Stansfeld, 1993), which negatively impact depression (Galea et al., 005; Weich et al., 2002), social support (Evans & Lepore, 1993; cCarthy & Saegert, 1979) and recovery from cognitive fatigue and

tress (Frumkin, 2001; Ulrich, 1991). Living in an SRO, when compared to living in other housing envi-

onments, has been associated with higher rates of HIV infection, mergency room use, recent incarceration, having been physi- ally assaulted, crack cocaine smoking, and cocaine, heroin, and ethamphetamine injection (Evans & Strathdee, 2006; Shannon,

shida, Lai, & Tyndall, 2006). Further, Lazarus, Chettiar, Deering, abess, and Shannon (2011) demonstrate that the specific organi- ation and management of SROs creates a gendered vulnerability to iolence and sexual risk taking among women. Political-economic heories which account for the role place (Bourgois & Schonberg, 009; Fullilove, 2013; Popay et al., 2003; Rabinow, 2003) have

ncluded an analysis of the structural-level policies responsible for he creation of built environments through the use of public funds. rawing from this example, we adapt Rhodes (2002, 2009) “risk nvironment” framework to argue that SROs can operate as “men- al health risk environments” for urban poor women. Consistent ith the risk environment framework (Rhodes, Singer, Bourgois,

riedman, & Strathdee, 2005; Rhodes et al., 2012), our analysis xamines the interplay between: (1) housing policies addressing omorbid substance use and mental illness as a macro-level fac- or shaping the built environments of SROs, (2) meso-level factors uch as the management of social relationships within SROs, includ- ng drug/sex economy involvement, and (3) micro-level individual ehaviors related to drug use and trauma management enacted ithin SROs.

Our application of the risk environment framework to SROs ffers potential contributions in the areas of theory, methodol- gy, and health policy. Theoretically, our analysis foregrounds how pecific constructions of urban space may exacerbate women’s o-occurring mental health issues and substance use. Methodolog- cally, we employ qualitative methods to examine the relationship etween space, drug use, and mental health to reveal the link- ges between housing policies, the socio-structural organization f urban built environments and everyday behaviors. In terms of ealth policy, our analysis highlights the importance of consid- ring comorbidity in housing policy for active substance users, articularly the role of trauma-sensitive housing environments for nstably housed women who use illicit drugs.


Our participants were recruited from a larger epidemiological tudy, the “Shelter, Health and Drug Outcomes among Women” SHADOW), a cohort study of homeless and unstably housed

f Drug Policy 25 (2014) 556–561 557

women living in San Francisco (Riley et al., 2007). A qualitative sub-sample (n = 30) was selected from the larger SHADOW cohort. Consistent with qualitative study designs, the sample was not representative of the larger cohort (Silverman & Marvasti, 2008). Rather, we purposefully sampled (Coyne, 1997; Higginbottom, 2004) women illustrative of a set of issues (recent physical and/or sexual victimization, unprotected sex, and needle sharing) previ- ously described in the epidemiological literature to be relevant to unstably housed women (Coughlin, 2011). Women in the sub- sample underwent a separate consent process and took part in approximately hour-long taped interviews with trained qualita- tive researchers (Knight, Lopez, and Cohen). During the interviews, participants were asked to describe their current and past living situations, current and past drug use, mental health (including experiences with diagnosis and psychiatric medications), sexual and friendship relationships, and experiences with violence and trauma. Participants completed a baseline, one-year, and 18-month follow up interview and were reimbursed $15 for each interview completed. All study procedures were approved by the Institu- tional Review Board at the University of California, San Francisco. In addition, the first author (Knight) conducted an independent, four-year (2007–2010) ethnographic study which included inter- views with housing and health policy-makers in San Francisco and a photo-ethnographic study of a variety of SRO hotel rooms. Over 500 photographs were taken during this timeframe in 25 different SRO hotels in San Francisco.

Transcribed audio-recorded interviews from each study under- went a similar two-phase analysis, consistent with methods the authors have employed in several previous qualitative studies (Comfort, Grinstead, McCartney, Bourgois, & Knight, 2005; Knight et al., 1996, 2005). In phase one, the team of four analysts (three of whom were the interviewers) used grounded theory method- ologies (Strauss & Corbin, 1990) to construct memo summaries of each interview, which included basic background information, current circumstances, notable events and quotations, and analyst impressions and interpretations. Because previous research (Chan, Dennis, & Funk, 2008; Cohen et al., 2009; Hopper et al., 1997; Kushel et al., 2003; Luhrmann, 2008) indicated a potential relationship between lifetime histories of traumatic exposure, housing instabil- ity, current living situations, and sexual and drug use behaviors, we sought to keep narratives “intact” in the initial data analysis phase. The interview transcript and summaries were then discussed at a 2-h meeting devoted to analyzing each participant’s interview. The team identified each narrative’s micro, meso, and macro fac- tors for analysis. After the initial group analysis process, the team developed a preliminary codebook, which was amended through- out data collection. In phase two of analysis, interview transcripts were coded and entered into a qualitative data management soft- ware program (, to produce aggregate data for the entire qualitative sample. For the purposes of this analysis, memoed summaries and multiple aggregate sections of coded data (e.g., codes for housing, trauma, mental health, neighborhood) were analyzed. Photo-ethnographic data were coded by location, type of hotel, and date.


Macro-level factors: housing policies shape SRO built environments in San Francisco

The widespread implementation of mental health deinstitu- tionalization policies which took place in the 1970s and 1980s in

California was not accompanied by structured housing plans for the uptake of mentally ill persons now residing in the community (Lamb, 1984). Thus, community reintegration of adults with dis- abling mental illnesses created a housing need, which was largely

5 urnal of Drug Policy 25 (2014) 556–561

u a s

e t S l m S 2 a u t p p n o s m

i p P a o f h o w i e t “ s o c m i t

v t t a

t e a


The women in our sample had high rates of co-occurring men- tal health and substance use issues and extensive histories of

58 K.R. Knight et al. / International Jo

nmet. One policy maker outlined the statistics on co-morbidity mong the population in San Francisco, underscoring the relation- hip between drug use, place, and social policy in this setting:

Of the people in supportive housing in San Francisco, 93% have a major mental illness that we can name. That is very, very high. 80% use cocaine, speed, or heroin every thirty days, or get drunk to the point of unconsciousness. There are no more disabled people in this country.

Because of San Francisco’s small size and geographic bound- dness, it was expedient to use existing SROs as sites to house he burgeoning urban poor. To date, there are more than 500 ROs in San Francisco, providing homes for approximately 30,000 ow income individuals (CCSRO website). These built environ-

ents include both larger and smaller building stock, with some RO hotels housing up to 200 persons and others with only 5–30 rooms. The necessity of using existing SRO housing as sites to ccommodate the expanding population of impoverished individ- als created a trifurcated system. This system has led some women o find housing in older, privately run and managed SROs, some in reviously privately owned buildings whose master lease had been urchased by the City of San Francisco, and others to be housed in ew buildings built on the demolished cites of older SROs or in ther urban spaces.1 These three types of built environments pre- ented different challenges to women in the management of their ental health. The department of Housing and Urban Health (HUH), the first

n the country to formally integrate housing management with ublic health, was created within the San Francisco Department of ublic Health to develop and manage the publically funded older nd newer SRO buildings. The HUH discovered through the course f this progressive housing initiative that building new, publically unded SROs is more cost effective and produces better housing and ealth outcomes for the tenants, than converting exiting privately wned SROs. Even if rental payments could be deferred through elfare or subsidy payment mechanisms, simply placing adults

ndoors in older SRO buildings was not efficacious if the indoor nvironment was still chaotic, dangerous, and poorly managed. At he macro-level, the built environment needed to be responsive to trauma.”2 For a population of tenants with high rates of co-morbid ubstance use and mental health issues, the built environment – the rganization of the physical and social space – was construed as ritical to ensuring housing success. One health and housing policy aker compared the different levels of housing stability for tenants

n new SRO built environments to those in older SROs, to emphasize he interactive relation between the built environment and trauma:

When we look at our success in keeping people housed in our buildings, what we see is that places like the Marque,3 which has small, dirty rooms, case management, but shared bathrooms. The rate of people staying housed there for two years consec-

utively is 30%. That is horrible. The Zenith, a new building, has case management, same as the Marque. But it is beautiful; every room has its own bathroom. 70% of the tenants stay at least1 The payment structure for rent in these three types of SROs is complex and aries for tenants depending on whether they pay for SRO rooms out of pocket, or hrough welfare program linked subsidies, of which there are several. Discussion of he complex payment structures is beyond the scope of this paper, but is discussed t length in Knight, KR, Forthcoming with Duke University Press. 2 “Trauma” here is a colloquial (as opposed to clinical) term deployed to refer

o the complex array of affective symptoms many chronically-homeless persons, specially women, demonstrate in daily life as a result of historic experiences of buse and current vulnerabilities. 3 The names of SRO hotels are pseudonyms.

Fig. 1. Older SRO room compared to newly built SRO room.

two years.” The point is the good stuff is the better investment when it comes to supportive housing. The environment mat- ters. I think it is about trauma. People, who have had so much trauma cannot stabilize, cannot stay housed if they still living in a dump.

The following pictures draw a comparison between the physi- cal environment deemed to be “trauma-sensitive” and the standard situation in privately owned SRO. The physical layout of a typi- cal SRO is a single, 8 × 10′′ room with shared toilets and showers down the hallway. Newly built SROs were often clean, well-lit, less chaotic, well-managed, and safer. Newer SROs included individual bathrooms and sometimes small kitchens to prepare food. In con- trast, older and privately owned SROs often consisted of a double or single bed, a sink, a small chest of drawers, and a desk. The phys- ical conditions which routinely affected women’s mental health in our study included the presence of rats, mice, and bed bugs; graffitied walls and broken furniture; and, non-operating sinks, electricity, door locks, and TV sets. As demonstrated in the photos, the condition and functionality of the physical aspects of the built environment varied a great deal and this variation contributed in positive and negative ways to women’s mental health outcomes (Figs. 1–3).

Meso and micro-level factors: social relations and behavioral strategies intersect with the built environment to influence mental

Fig. 2. View of out of window newly built SRO compared to view out of older SRO window.

K.R. Knight et al. / International Journal o

c m w m

e o h r h w p

e h c g e t a m n

Fig. 3. Newly built SRO hallway compared to older SRO hallway.

hildhood and adult sexual and physical victimization, making the anagement of trauma symptoms an everyday life challenge. One oman described the impact her new calm, controlled environ- ent had on her risk for poor mental health:

I discovered that my environment had a lot to do with my men- tal state. So, when I had my own place, I was in control of the environment. You know, there was no drama, everything was nice and mellow, and so I was able to function. Everything was on an even keel; that was fine. It was when other people and situations were introduced into my environment that I couldn’t get away from, that would send me over the edge.

The physical and social organization of specific SRO housing nvironments made such a significant impact on the women in ur studies that many reported choosing street homelessness or omeless shelter stays if they could not secure a room in a monthly ate, clean, and safe SRO. Reinforcing the data provided from the ousing policy-maker, one woman described “shopping” for an SRO hich met her mental health needs, rather than accepting the first ublically subsidized built environment offered to her.

[The homeless shelter administrator] told me I would find a place [through a subsidized program] if I work with them. And they did find me a lot of places, but I didn’t want to go, because [those] SROs they have now are really nasty. Really tore up, tore down. Syringes in the bathroom. Blood on the toilet. Because you use the same toilet that everybody else uses. So it wasn’t sanitized. So I didn’t want to go. And I found the Martin Hotel and I went in and it was a really clean, nice place. So I went back to [the shelter] and I asked them ‘Can you please get me a place inside that hotel?’ They said that would be cool, they would work on it. And within two, three weeks I had a place at the Martin.

SRO environments where women felt unsafe exacerbated sev- ral physical and emotional symptoms associated with poor mental ealth. The physical organization of SROs frequently consisted of rowding people with addiction and mental health issues into a sin- le space. Crowding, in combination with chaos related to drug/sex conomy interactions, and rapid cycling of new tenants contributed o stress-related sleeping problems, hyper-vigilance, and drug and lcohol use. Many women described needing prescription sleep edication to rest in chaotic hotel settings and avoid conflicts with


When I go in [to the SRO hotel] and shut my door, I just try to shut my eyes and block it out. Sometimes they [neighbors]

f Drug Policy 25 (2014) 556–561 559

have their TVs on and I want to say something. I’m thinking, ‘You know, [says her own name], just be quiet! Just go to sleep.’ Once I take my [sleeping] pills, I’m good.

Women commonly adopted a strategy of deliberate social isola- tion to shield themselves from risk for victimization within unsafe SRO environments. For some women, isolation in the hotel room was an emotionally self-protective response to daily living in a trau- matized state. One woman provided an example of isolation linked to on-going fears of being attacked:

So I started using back in 2009, which I have been using drugs for a year now. I got raped last year. I got raped, I got kidnapped. I was tortured for days. My best friend died, as I told you. It’s just everything fell apart and I have been tore up since then. . .Since I moved to [my SRO], I basically stay in my room all day.

For others, isolation served as a strategy to avoid being “caught up” in unpredictable violence and social disorder associated with the drug–sex economy:

So, now I’m here, you know, just trying to deal with a lot of different things, you know. Adjustment of being back [in my SRO room] which I’m getting more adjusted to it, but I don’t like the space that I’m in because it’s small. Of course, I don’t mingle with my neighbors either. . .I just tend to stay to myself because I see trouble there and I avoid that because I don’t need that in my life, you know. So, that’s another thing I deal with on a day-to-day basis you know.

In contrast to the women above who described deliberate social isolation as a mental health survival strategy, another woman positively described increased safety and independence in built environments which were perceived as safe and non-chaotic. For example, one positively described her highly structured SRO hous- ing environment, a place specifically designed to reduce her fear and anxiety due to repeated victimization and to enhance her abil- ity to manage her mental health symptoms despite years of trauma and housing instability:

Oh, it’s [my room’s] beautiful, it’s comfortable and it’s quiet and it’s clean! I mean the manager there is up on it. He’s got secu- rity cameras now. It’s secure, I’m high up. The only way you can get into my window is if you try to do it. And if you try to do it and you fall, you’re going to die. It’s out of the way [out of the neighborhood], yeah. And so the [public] bus takes me to school. Takes me straight to school, straight home. Boom, no chaos. Wal- green’s right there. Boom, psych meds, boom right there, boom. Bus pass, Walgreen’s right there, boom. Everything’s right there. You know [the bank] is right on the corner, boom. I’m just – McDonald’s everything, grocery store, laundromat, everything is just right there in my commute. I don’t have to go a block to go to the laundromat. I don’t have to go through a block to go to grocery shopping. So, everything is just perfect for me.

In terms of localized drug policy and housing, the adjudication of in-building drug use was not prioritized by the women in our study to the same extent as other measures taken to ensure the built environment was spatially and socially organized to reduce fear, anxiety, and conflict. While women acknowledged the risks that the drug–sex economy posed to their mental health, many also actively participated in those economies as drug users and (inter-

mittently) as sex workers. Even women who were seeking to reduce or eliminate their own drug use, or who were abstinent, did not suggest that drug use or sex work should be outlawed within their hotels to promote safety. Opinions veered towards a “closed-door”

5 urnal

p a o b w i a D i d c f p d l i o c t

h l a m


t c c t h o p u s a O v r

c p H i c S F r a t w t S s r c

60 K.R. Knight et al. / International Jo

olicy, particularly about drug use. Women expressed that, ide- lly, open-air drug markets and the street-level chaos and violence ften associated with the drug–sex economy should be mitigated y the hotel management, thereby promoting safety and control ithin the housing environments. In one example, several women

n our study positively described an active campaign by SRO man- gement, which evicted drug-dealing tenants from the building. rug-using tenants were not targeted; however, those participat-

ng in the economy that brought associated violence and social isturbance were systematically removed. In another example, a rack and heroin-using woman described her building as safe, had riendships with neighbors in the hotel, and could list several exam- les of how her hotel manager helped her and other tenants. “We on’t have an open-air drug market here,” she noted. At the macro-