Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Sep 29.
Published before final editing as: Subst Use Addctn J. 2026 Aug 25:29767342261476437. doi: 10.1177/29767342261476437

Health harms associated with forced displacements among people who use drugs in San Francisco, California

Esther O Chung a,*, Cariné E Megerian a, Leslie W Suen b, Lynn D Wenger a, Bradley Ray a, Barrot H Lambdin a,c, Alex H Kral a
PMCID: PMC13618012  NIHMSID: NIHMS2206862  PMID: 42639645

Abstract

Background:

Forced displacements may increase overdose risk among people who use drugs (PWUD) through disruptions to naloxone, safer drug use supplies, and substance use disorder (SUD) treatment access. We investigated whether forced displacements were associated with non-fatal overdose, naloxone access, and SUD treatment access among PWUD in San Francisco, California.

Methods:

We conducted a cross-sectional survey of 790 PWUD from September 2024 to September 2025. Participants reported number of forced displacements by police, security guards, or other officials in the past 3 months. Outcomes included past 3-month overdoses, naloxone access, and SUD treatment. We conducted separate multivariable modified Poisson models with robust standard errors for each outcome, adjusting for age, gender, race, housing status, past 3-month opioid use, and interviewer.

Results:

In the past 3 months, 70% of participants were forcibly displaced at least once and 36% were displaced 20 or more times. Of those forcibly displaced at least once, 65% had their belongings taken by those displacing them. Eleven percent experienced an overdose; 62% were provided with naloxone; 33% received any SUD treatment and 18% were currently receiving any treatment. Those forcibly displaced 20+ times had 1.99 times the prevalence of overdose of non-displaced participants (95% CI: 1.01 – 3.93). Participants who had their belongings taken had 1.59 times the prevalence of non-fatal overdose of those who did not (95% CI: 1.00 – 2.53). There was no association between forced displacements and receiving any SUD treatment, but participants who were forcibly displaced 20+ times had a lower prevalence of currently receiving any treatment (aPR=0.44, 95% CI: 0.27–0.72).

Conclusions:

Our findings demonstrate forced displacements have significant health harms and negatively affect treatment services. Forced displacement policies should be replaced with public health approaches that prioritize stability, housing, and supportive services.

MeSH Keywords: drug user, drug overdose, homelessness, drug treatment center

1. INTRODUCTION

Over one million people have died from a drug overdose in the past two decades1 and there were over 80,000 overdose deaths in 2024.2 In the same year, over 770,000 people experienced homelessness on a single night in the United States (U.S.), the highest number ever recorded.3 Of these people, one in three were unsheltered, defined as sleeping in places not meant for human habitation, including outdoors, vehicles, and abandoned buildings.3 People who use drugs (PWUD) are at increased risk of infectious diseases, overdoses, and severe mental illness,4–6 and unsheltered homelessness further exacerbates these risks. Unsheltered homelessness is associated with adverse health outcomes, including severe mental illness,7 substance use disorders,8 chronic and infectious diseases,9–11 and premature mortality.12,13

As the volume and visibility of unsheltered homelessness have increased, local governments have increasingly relied on forced displacements (e.g., encampment sweeps or removing unhoused people and their belongings en masse) as a primary strategy to respond to the public’s complaints about unsheltered homelessness.14–16 Under the threat of citation, arrest, and incarceration, people are forced to move locations, often without connection to viable housing options or supportive services.17,18 These practices have occurred alongside major legal and policy shifts that have expanded the authority of local governments to enforce laws that criminalize unsheltered homelessness. In June 2024, the U.S. Supreme Court ruling in City of Grants Pass v. Johnson allowed municipalities to enforce bans on camping or sleeping on public property including fines and potential jail time, even among people experiencing homelessness who have no access to shelter.19 Shortly after, state and local jurisdictions across the U.S. began implementing and enforcing such bans, primarily through forced displacements, including in San Francisco, California. The resulting legal and policy changes from the U.S. Supreme Court ruling have intensified the frequency of forced displacements, raising urgent questions about their health impacts on PWUD.

Prior quantitative studies have documented that forced displacements are associated with increased odds of non-fatal overdose, infectious disease, physical violence, and worsening mental health.20–22 Simulation studies further suggest that forced displacements are associated with increased non-fatal and fatal overdoses, serious infectious diseases, and all-cause mortality.23,24 Qualitative and mixed methods studies provide important context on how forced displacements shape overdose risk, including the loss of life-saving medications, supplies, and survival necessities, such as naloxone for overdose reversal, sterile syringes and other safer drug use supplies, identification cards, and telephones.14,25–28 Forced displacements also fragment social support networks and push people into more hidden, isolated, and hazardous places, increasing the risk of overdose deaths.14,25,26 However, research on forced displacements and substance use disorder (SUD) treatment engagement remains limited. To our knowledge, only one prior study has examined this association and found lower odds of SUD treatment in the past 6 months among those who were forcibly displaced.20

While existing research suggests adverse health outcomes stemming from forced displacements, these studies were conducted prior to the U.S. Supreme Court ruling in City of Grants Pass v. Johnson and the subsequent expansion of forced displacements. This new policy landscape warrants examination of whether forced displacements affect health outcomes among PWUD, as more widespread enforcement may amplify disruptions to overdose prevention resources and SUD treatment engagement. To address these gaps, this study aimed to investigate whether forced displacements were associated with non-fatal overdose, naloxone access, and SUD treatment among PWUD in San Francisco, California, following the Supreme Court’s Grants Pass v. Johnson ruling.

2. METHODS

2.1. Study sample

This study used data from a parent study evaluating the effectiveness and implementation of services designed to improve the health of PWUD in San Francisco, California. The parent study uses a serial cross-sectional survey design, conducting a community-based, cross-sectional sample of PWUD every year. Our study used the cross-sectional sample of PWUD (n=789) from September 2024 to September 2025. Prior studies using an earlier cross-section provide further details on the study design and data collection methods.29,30 Briefly, we used targeted sampling methods to recruit participants.31,32 Trained community outreach workers conducted direct outreach in street settings, homeless encampments, or other areas frequented by PWUD. Participants were recruited based on geographic area, demographic characteristics, common physical markers of drug use (e.g., swollen hands, visible drug use supplies), and ability to engage with the outreach workers. Eligibility determination, study participation decisions, and informed consent were then conducted at the research field sites. Participants were eligible if they were 18 years or older; used heroin, fentanyl, methamphetamine and/or cocaine in the past 30 days, as verified by a series of questions about their drug use patterns; spoke either English or Spanish; and provided competent verbal informed consent. Given that participants were recruited through direct community outreach in street settings using targeted sampling, a formal response rate cannot be calculated, as the number of individuals exposed to recruitment efforts prior to arriving at the field site is unknown. Among individuals screened for eligibility at the field site, there were no refusals to participate. Quantitative surveys were conducted face-to-face by trained research interviewers. Surveys were anonymous, confidential, and lasted 20–45 minutes. All respondents provided verbal consent prior to data collection and were remunerated $40 cash for their participation. All study procedures were approved by the Institutional Review Board at RTI International (#00022126). This study was conducted in accordance with the Declaration of Helsinki.

In the present study, we excluded participants missing data on the exposure variable, number of forced displacements (n=71), and other key covariates (n=18). To maximize statistical power, we then subset the data for analyses based on outcome domain: (1) overdose and naloxone and (2) SUD treatment. For overdose and naloxone analyses, an additional 8 participants were missing data on outcomes, resulting in a sample size of 692. For SUD treatment analyses, an additional 33 participants were missing data on outcomes, resulting in a sample size of 667.

2.2. Measures

2.2.1. Exposures: Forced displacement and belongings taken

Participants were asked “In the past 3 months, how many times did police, security guards, or other city or county officials force you to move where you were staying?” We defined displacement as participants reporting they had been forced to move one or more times. Displacement included both locations where participants may have been sleeping or residing and temporary spaces where they may have been standing or sitting. We categorized the number of forced displacements using the median value after excluding zeros, which resulted in the following categories: None (referent), 1–20 times, and 20 or more times.

Participants were also asked “In the past 3 months, how many times did police, security guards, or other city or county officials take away any of your belongings?” We generated a binary indicator for participants who had their belongings taken at least once in the past 3 months (yes/no). We used a binary indicator given the skewed distribution.

2.2.2. Outcomes

We had three self-reported outcomes of interest: non-fatal overdoses, naloxone access, and SUD treatment access. For non-fatal overdose, participants were asked “In the past 3 months, how many times have you had an opioid-related overdose?”. Given minimal heterogeneity in the number of overdoses, we created a binary indicator for any overdose in the past 3 months (yes/no). We used two questions to assess naloxone access. Participants were asked (1) “In the past 3 months, have you been given naloxone to carry with you (including injectable, intranasal, and/or autoinjector) for use on yourself or others?”; and (2) “Do you currently have naloxone with you (for example, in your bag or a pocket)?”. We used two questions to assess SUD treatment access. Participants were asked (1) “In the past 3 months, have you had any of the following kinds of drug treatment? Select all that apply.”; and (2) “Are you currently receiving any of the following kinds of substance use disorder treatment? Select all that apply.” Treatment options included medications (including buprenorphine, methadone, naltrexone for opioid use disorder, and mirtazapine, bupropion, methylphenidate or amphetamine/ dextroamphetamine for stimulant use disorder); outpatient (including outpatient programs, sober living or board and care, contingency management, working with a medical doctor for counseling, harm reduction counseling, and self-help groups like Narcotics Anonymous or Alcoholics Anonymous), and inpatient treatment (including partial hospitalization, residential or inpatient drug treatment, and withdrawal management).

2.2.3. Confounders

We selected the following confounders a priori based on potential differences between forced displacement and outcomes: age, gender, race-ethnicity, housing status, and opioid use. We categorized age into approximate thirds based on the data distribution: 18–34, 35–54, and 55 or older. Participants were asked to self-report their gender based on multiple choice questions and we created an indicator, comparing cisgender men to cisgender women and gender expansive participants. Participants self-reported their race and ethnicity based on multiple choice questions. Race and ethnicity are not indicators of biological differences, but rather sociopolitical experiences and histories.33,34 Given the high rates of overdose among Black San Franciscans,35 we grouped Black/African American participants (alone or in combination) to all other participants. Participants reported whether they considered themselves currently homeless or unstably housed. Participants self-reported whether they used the following substances in the past 3 months: fentanyl alone or mixed with stimulants (methamphetamine, powder or crack cocaine), heroin alone or mixed stimulants (methamphetamine, powder or crack cocaine), methamphetamine, powder or crack cocaine, prescription opioids not as prescribed, prescription stimulants not as prescribed, benzodiazepines, xylazine, nitazene, hallucinogens, and methadone and/or buprenorphine not as prescribed. We created an indicator for any opioid use in the past 3 months if participants selected fentanyl or heroin, either alone or mixed with stimulants.

2.3. Statistical analysis

We used the overdose and naloxone sample (n=692) to provide summary statistics using means and standard deviations for continuous variables and percentages for categorical variables. We estimated adjusted prevalence ratios (aPR) and 95% confidence intervals (CI) using modified Poisson models with robust standard errors. All models adjusted for age, gender, race-ethnicity, housing status, and any opioid use in the past 3 months. We also included interviewer fixed effects given prior research suggesting that sensitive or complex questions can lead to greater variance across interviewers.36 To assess whether a finer categorization of the number of forced displacements affected results, we conducted a sensitivity analysis using four categories (None, 1–5, 6–20, 20 or more times). We did not find differences in demographics characteristics or multivariable model results between participants in the 1–5 and 6–20 groups. Therefore, we retained the 3-category exposure (None, 1–20, 20+ times). We used Stata 19 and R version 4.5.2 to conduct all analyses.37,38

3. RESULTS

3.1. Sample characteristics

In our community-based sample of PWUD, 62% were 35–54 years old, 70% identified as cisgender men, and 32% identified as Black/African American (Table 1). Over 70% of participants identified as currently homeless or unstably housed. Seventy percent were forcibly displaced by police, security guards, or other officials at least once in the past 3 months, and 36% were displaced 20 or more times. Of those forcibly displaced at least once, 65% had their belongings taken by those displacing them. Eleven percent experienced an overdose in the past 3 months; 62% were provided with naloxone and 45% were currently carrying a dose; 33% received any SUD treatment in the past 3 months and 18% were currently receiving treatment.

Table 1. Sample characteristics.

Number of forced displacements, past 3 months p-value
Total None 1–20 times 20+ times
N=692 206 (29.8%) 235 (34.0%) 251 (36.3%)
Age <0.001
 18–34 161 (23.3%) 29 (18.0%) 59 (36.6%) 73 (45.3%)
 35–54 431 (62.3%) 108 (25.1%) 157 (36.4%) 166 (38.5%)
 55 or older 100 (14.5%) 69 (69.0%) 19 (19.0%) 12 (12.0%)
Gender 0.372
 Cisgender man 479 (69.2%) 134 (28.0%) 160 (33.4%) 185 (38.6%)
 Cisgender woman 189 (27.3%) 64 (33.9%) 66 (34.9%) 59 (31.2%)
 Gender expansive 24 (3.5%) 8 (33.3%) 9 (37.5%) 7 (29.2%)
Race <0.001
 American Indian or Alaska Native 50 (7.2%) 20 (40.0%) 16 (32.0%) 14 (28.0%)
 Asian 16 (2.3%) 5 (31.2%) 7 (43.8%) 4 (25.0%)
 Black/African American 220 (31.8%) 99 (45.0%) 63 (28.6%) 58 (26.4%)
 Hispanic or Latinx 64 (9.2%) 12 (18.8%) 29 (45.3%) 23 (35.9%)
 Native Hawaiian or Pacific Islander 7 (1.0%) 1 (14.3%) 3 (42.9%) 3 (42.9%)
 White 335 (48.4%) 69 (20.6%) 117 (34.9%) 149 (44.5%)
Currently homeless or unstably housed <0.001
 No 195 (28.2%) 119 (61.0%) 44 (22.6%) 32 (16.4%)
 Yes 497 (71.8%) 87 (17.5%) 191 (38.4%) 219 (44.1%)
Any opioid use, past 3 months <0.001
 No 97 (14.0%) 44 (45.4%) 34 (35.1%) 19 (19.6%)
 Yes 595 (86.0%) 162 (27.2%) 201 (33.8%) 232 (39.0%)
Open in a new tab

3.2. Non-fatal overdose and naloxone access

The prevalence of non-fatal overdose among participants who were forcibly displaced 20 or more times was 1.99 times that among non-displaced participants (95% CI: 1.01 – 3.93) (Figure 1, Appendix Table 1). Associations between forced displacement frequency and naloxone outcomes were not statistically significant. Compared with non-displaced participants, those forcibly displaced 20 or more times had 1.11 times the prevalence of receiving naloxone (95% CI: 0.93 – 1.32) and 1.14 times the prevalence of currently having naloxone (95% CI: 0.89 – 1.47). Participants who were displaced 1–20 times had similar, but attenuated relationships with non-fatal overdose and naloxone outcomes.

Figure 1. Associations between forced displacements and belongings taken and overdose and naloxone outcomes (n=692).

Figure 1.

Open in a new tab

We conducted separate multivariable modified Poisson models for overdose and naloxone outcomes. Models adjusted for age, gender, race-ethnicity, housing status, opioid use, and interviewer.

Participants who had their belongings taken had 1.59 times the prevalence of non-fatal overdose compared to those who did not (95% CI: 1.00 – 2.53) (Figure 1, Appendix Table 1). Associations between having belongings taken and naloxone outcomes were not statistically significant. Compared to participants who did not have their belongings taken, those who had a similar prevalence of receiving naloxone (aPR=1.01, 95% CI: 0.90 – 1.15) and had 1.11 times the prevalence of currently having naloxone (95% CI: 0.93 – 1.32).

3.3. Substance use disorder treatment

Associations between forced displacement frequency and receiving any SUD treatment in the past 3 months were not statistically significant. Compared to non-displaced participants, those forcibly displaced 20 or more times had a similar prevalence of receiving any SUD (aPR=0.96, 95% CI: 0.70 – 1.32), and those displaced 1–20 times had 0.86 times the prevalence (95% CI: 0.64 – 1.14) (Figure 2, Appendix Table 2). However, we found a negative relationship between forced displacements and currently receiving any SUD treatment. Compared to non-displaced participants, those who were forcibly displaced 20 or more times had 0.44 times the prevalence of currently receiving any SUD treatment (95% CI: 0.27 – 0.72). We found similar results for participants forcibly displaced 1–20 times and currently on treatment (aPR=0.55, 95% CI: 0.38 – 0.80).

Figure 2. Associations between forced displacements and belongings taken and substance use disorder (SUD) treatment outcomes (n=667).

Figure 2.

Open in a new tab

We conducted separate multivariable modified Poisson models for SUD treatment outcomes. Models adjusted for age, gender, race-ethnicity, housing status, opioid use, and interviewer.

Substance use disorder (SUD)

Associations between having belongings taken and SUD treatment outcomes were not statistically significant. Compared to participants who did not have their belongings taken, those who did had a similar prevalence of receiving any SUD treatment (aPR=1.06, 95% CI: 0.84 – 1.35) and had 0.78 times the prevalence of currently receiving SUD treatment (95% CI: 0.55 – 1.11).

4. DISCUSSION

In our community-based sample of PWUD in San Francisco, 70% were forcibly displaced at least once in the past 3 months by police, security guards, or other officials, and over 36% of participants were displaced 20 or more times. This prevalence is in line with prior quantitative studies of PWUD that also report a high prevalence of forced displacements in San Francisco, Los Angeles, and Denver ranging from 44% to 75%.20–22 Participants who were forcibly displaced 20 or more times had twice the prevalence of non-fatal overdose compared to those who were not displaced. Furthermore, participants who had their belongings taken by police, security guards, or other officials had a 59% higher prevalence of non-fatal overdose compared to those who did not, suggesting property confiscation may be a particularly harmful component of displacements. Our findings support prior observational and simulation studies that found increased risk of overdose stemming from forced displacements.20,23 Several mechanisms can explain how forced displacements may increase overdose risk. First, forced displacements create destabilizing and disruptive conditions that can lead PWUD to use drugs more quickly or consume larger or unpredictable doses to cope with the psychological distress and emotional trauma resulting from displacement or loss of belongings. Qualitative work from Baltimore, Maryland highlights how unsheltered PWUD navigate their drug use in public and private settings, often at the expense of drug use safety.39 Additional qualitative studies in Los Angeles, California and Vancouver, Canada found encampment displacements and constant surveillance from law enforcement disrupted individuals’ routines of daily survival (e.g., food, money) and drug use and increased feelings of anxiety, paranoia, and strain.25,28 Second, displacements can result in loss of sterile equipment for drug use and medications for opioid use disorder,25 including through direct confiscation of belongings by law enforcement and other officials, which increases the risk for infectious disease transmission and opioid withdrawal. Recent work has identified increased sharing of syringes and injection-related materials associated with forced displacements.40 Furthermore, the loss of belongings may directly increase overdose risk by removing safer drug use supplies, such as naloxone, fentanyl test strips, and sterile equipment that PWUD rely on to reduce drug-related harms, consistent with our finding that participants who had their belongings taken had a 59% higher prevalence of overdose. Third, displacements can sever critical social ties, moving people away from their peers who are often the first responders to an overdose and best positioned to recognize its signs and administer naloxone and rescue breathing.25,41 Finally, forced displacements often push people to more remote or isolated areas,14 further amplifying the loss of social connections and increasing the likelihood of using alone or using drugs from new or unfamiliar sources, both of which are known risk factors for overdose.42–44 Our study suggests that forced displacements may create preventable health harms for an already vulnerable population.

Although we did not observe differences in receiving SUD treatment in the past 3 months between PWUD who were and were not forcibly displaced, PWUD who were displaced 20 or more times had a 56% lower prevalence of currently receiving treatment. These results suggest that while forced displacements may not affect whether people had engaged in treatment within 3 months, exposure to high-frequency displacements interferes with the continuity of care, including the ability to adhere to medical or behavioral health visits and/or refill prescriptions. Our findings align with prior work showing that displaced PWUD face greater barriers to initiating and maintaining treatment care.20,25 Furthermore, forced displacements can fracture relationships with trusted outreach workers, street medicine teams, and social service providers, leading to fragmented care, loss to follow-up, and less effective management of chronic health conditions.45 During our study period, San Francisco actively expanded access to treatment, which included investing in a nighttime telehealth program to prescribe medications for opioid use disorder for people experiencing homelessness and increasing the hours and reach of existing programs and clinics.46,47 At the same time, the city continued and then aggressively escalated displacements in August 2024.48–50 Our findings highlight contradictory co-occurring policies in San Francisco in which the city implemented initiatives to expand SUD treatment, while simultaneously implementing punitive displacement policies that make it more difficult for PWUD to access and maintain treatment.

Contrary to our expectations, we found participants who experienced a high frequency of forced displacements had a higher prevalence of receiving naloxone and currently having naloxone on them, though these findings were not statistically significant. While prior studies have documented loss of naloxone as a result of forced displacements,25 it may be that in a densely populated city like San Francisco, PWUD who lose their naloxone are able to access more doses when they need it. The city has a long history of syringe services programs that offer outreach services to PWUD and our previous work has documented that over half of adult residents in the city live within a 20-minute walk of such programs.29 This estimate was based on the general adult population; however, many of these programs are concentrated in neighborhoods with the highest rates of unsheltered homelessness, such as the Tenderloin and South of Market, suggesting that geographic access to such programs may be reasonable for many unhoused residents. Therefore, although PWUD may lose naloxone during forced displacements, some may have the ability to more readily replenish their supply given their geographic access to syringe services programs. San Francisco’s Healthy Street Operations Center, which coordinated the city’s response to encampments during the study period, included outreach teams tasked with connecting people to shelter, medical care, and behavioral health resources during encampment closures.51 However, a prior report, drawing on city records and systematic observations of operations, found that encampment closures were largely conducted by police and public works employees rather than trained health professionals, and that naloxone and other safer drug use supplies were more often lost or seized during sweeps than distributed.52 To the extent that outreach teams provided the support described in the city’s policies, our findings may underestimate the true harm of forced displacements, as participants who experienced more forced displacements had a higher prevalence of non-fatal overdose despite a stated policy of supportive outreach. Additionally, it may be that PWUD who experience high-frequency displacements by police, security guards, or city/county officials prioritize holding onto essential and portable supplies such as naloxone, despite the risk of ongoing loss. Repeated moves and the threat of property loss shape how belongings are managed among people experiencing homelessness.25,53 In the present study, we did not measure whether participants lost naloxone during a displacement. Further studies could explore how PWUD adapt to the frequency of displacements, including loss of naloxone and whether and how they strategically maintain essential supplies like naloxone for themselves and their peers.

Our study has important policy implications given the expansion of displacement policies following the City of Grants Pass vs. Johnson ruling. One month after the ruling, Governor Gavin Newsom issued an executive order directing state agencies to remove homeless encampments throughout California, and in San Francisco, Mayor Breed launched a “very aggressive” campaign to remove encampments, authorizing law enforcement to cite and incarcerate individuals who refused shelter.48–50,54 In a state where 90% of people experiencing homelessness spent at least one night in an unsheltered location in the past 6 months,55 the intensification of encampment removals at both the state and local level raises serious public health concerns. Our findings suggest that enforcement and displacement responses to visible homelessness may undermine public health goals. Alternative approaches to managing unsheltered homelessness are urgently needed. Public health strategies that prioritize housing first approaches, including low-barrier, non-punitive housing paired with voluntary, supportive services, can improve health, stability, and safety for PWUD.26

Our results should be interpreted with some potential limitations in mind. First, this was a cross-sectional study and causality cannot be established. Future research using longitudinal designs would help with establishing temporality and additional qualitative research would help describe mechanisms. Second, data were self-reported and recall and social desirability bias may be present; however, the relatively short 3-month recall period and anonymity and confidentiality of the survey help minimize these biases. Additionally, underlying differences between groups, such as history of housing instability, SUD severity, and mental health comorbidities, may partially explain differences in observed outcomes. Third, while all participants reported using illicit drugs, we did not ask participants to report whether they were diagnosed with SUD. Fourth, our sample was limited to individuals who were alive at the time of the study; therefore, fatal overdoses could not be examined. If forced displacements are associated with fatal overdose, our estimates of the association between forced displacements and non-fatal overdose may underestimate the true burden of overdose in this population. Fifth, we did not collect data on whether harm reduction supplies or services were provided to participants during displacement events, which may have influenced our outcomes. We also did not collect data on where displacement events occurred. Finally, our study was conducted in San Francisco, California and results may not be generalizable to other settings with different political contexts, health and social services systems, and law enforcement practices.

5. CONCLUSION

Our findings suggest that punitive policies that forcibly displace people are associated with both increased overdoses and reduced engagement in SUD treatment. To minimize serious and preventable harms, forced displacement policies should be replaced with health-oriented approaches that prioritize stability, housing, and supportive services for PWUD rather than punishment.

Acknowledgements

We are deeply grateful to the people who participated in our study. We would also like to thank the following people for important contributions to the collection of data: Shelby Bergeron, Adrienne Leddy, Ramon Lopez, Vero Majano, Terry Morris, and Jason Storm. We also thank our Community Advisory Board for providing expertise and feedback on our study.

Funding Statement

This work was supported by the National Institute on Drug Abuse (R01DA057613), which did not have any role in data collection, analysis and interpretation of data, the writing of the report, or the decision to submit the article for publication.

Appendix

Appendix Table 1. Associations between forced displacements and overdose and naloxone outcomes (n=692).

Any non-fatal overdose, past 3 months Received naloxone, past 3 months Currently have naloxone
N No
N (%)		 Yes
N (%)		 aPR
(95% CI)		 No
N (%)		 Yes
N (%)		 aPR
(95% CI)		 No
N (%)		 Yes
N (%)		 aPR
(95% CI)
Number of forced displacements, past 3 months
 None 208 194
(93.3%)		 14
(6.7%)		 ref 71
(34.1%)		 137
(65.9%)		 ref 125
(60.1%)		 83
(39.9%)		 ref
 1–20 times 235 215
(91.5%)		 20
(8.5%)		 1.45
(0.70–2.97)		 93
(39.6%)		 142
(60.4%)		 0.90
(0.77–1.04)		 127
(54.0%)		 108
(46.0%)		 1.03
(0.82–1.29)
 20+ times 252 208
(82.5%)		 44
(17.5%)		 1.94
(1.01–3.93)		 103
(40.9%)		 149
(49.2%)		 1.11
(0.93–1.32)		 128
(50.8%)		 124
(49.2%)		 1.14
(0.89 – 1.47)
Times belongings taken, past 3 months
 None 368 334
(90.8%)		 34
(9.2%)		 ref 152
(41.3%)		 216
(58.7%)		 ref 217
(59.0%)		 151
(41.0%)		 ref
 At least once 324 281
(86.7%)		 43
(13.3%)		 1.59
(1.00–2.53)		 115
(35.5%)		 209
(64.5%)		 1.01
(0.90–1.15)		 160
(49.4%)		 164
(50.6%)		 1.11
(0.93–1.32)
Open in a new tab

We conducted separate modified Poisson models for each outcome. Models adjusted for age, gender, race-ethnicity, housing status, opioid use, and interviewer.

Adjusted prevalence ratio (aPR); Confidence interval (CI); Standard deviation (SD)

Appendix Table 2. Associations between forced displacements and substance use disorder (SUD) treatment outcomes (n=667).

Received SUD treatment, past 3 months Currently receiving SUD treatment
N No
N (%)		 Yes
N (%)		 aPR
(95% CI)		 No
N (%)		 Yes
N (%)		 aPR
(95% CI)
Number of forced displacements, past 3 months
 None 202 132
(65.3%)		 70
(34.7%)		 ref 143
(70.8%)		 59
(29.2%)		 ref
 1–20 times 336 165
(70.2%)		 70
(29.8%)		 0.86
(0.64–1.14)		 198
(84.3%)		 37
(15.7%)		 0.55
(0.38–0.80)
 20+ times 134 155
(66.5%)		 78
(33.5%)		 0.96
(0.70–1.32)		 209
(89.7%)		 24
(10.3%)		 0.44
(0.27–0.72)
Times belongings taken, past 3 months
 None 355 242
(68.2%)		 113
(31.8%)		 ref 283
(79.7%)		 72
(20.3%)		 ref
 At least once 312 207
(66.4%)		 105
(33.7%)		 1.06
(0.84–1.35)		 264
(84.6%)		 48
(15.4%)		 0.78
(0.55–1.11)
Open in a new tab

We conducted separate modified Poisson models for the SUD treatment outcomes. Models adjusted for age, gender, race-ethnicity, housing status, opioid use, and interviewer.

Adjusted prevalence ratio (aPR); Confidence interval (CI); Standard deviation (SD); Substance use disorder (SUD)

Footnotes

Author Contributions: CRediT

All authors assisted in study design and played a role in conducting the study. Esther O. Chung: Conceptualization, Data Curation, Formal Analysis, Investigation, Methodology, Software, Visualization, Writing – Original Draft. Cariné E. Megerian: Investigation, Methodology, Project Administration, Writing – Review and Editing. Leslie W. Suen: Funding Acquisition, Methodology, Writing – Review and Editing. Lynn D. Wenger: Investigation, Methodology, Writing – Review and Editing. Bradley Ray: Conceptualization, Methodology, Writing – Review and Editing. Barrot H. Lambdin: Funding Acquisition, Methodology, Supervision, Writing – Review and Editing. Alex H. Kral: Conceptualization, Funding Acquisition, Investigation, Methodology, Supervision, Writing – Review and Editing.

Statements and Declarations

Ethical Considerations

All respondents provided competent verbal informed consent prior to data collection. All study procedures were approved by the Institutional Review Board at RTI International (#00022126). This study was conducted in accordance with the Declaration of Helsinki.

Consent for Publication

Not applicable.

Declaration of Conflicting Interest

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability

Data from this study will be stored in a public data repository at the completion of the research grant.

REFERENCES

  • 1.Monnat SM. Demographic and Geographic Variation in Fatal Drug Overdoses in the United States, 1999–2020. Ann Am Acad Pol Soc Sci. 2022;703(1):50–78. doi: 10.1177/00027162231154348 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Ahmad FB, Cisewski JA, Rosen L, Sutton P. Provisional drug overdose death counts. Published online 2026. Accessed January 14, 2026. 10.15620/cdc/20250305008 [DOI] [Google Scholar]
  • 3.de Sousa T, Henry M. The 2024 Annual Homelessness Assessment Report (AHAR to Congress) Part 1: Point-In-Time Estimates of Homelessness, December 2024. Department of Housing and Urban Development, Office of Community Planning and Development; 2024. https://www.huduser.gov/portal/datasets/ahar/2024-ahar-part-1-pit-estimates-of-homelessness-in-the-us.html [Google Scholar]
  • 4.Degenhardt L, Webb P, Colledge-Frisby S, et al. Epidemiology of injecting drug use, prevalence of injecting-related harm, and exposure to behavioural and environmental risks among people who inject drugs: a systematic review. Lancet Glob Health. 2023;11(5):e659–e672. doi: 10.1016/S2214-109X(23)00057-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Dinwiddie AT, Gupta S, Mattson CL, O’Donnell J, Seth P. Reported Non–Substance-Related Mental Health Disorders Among Persons Who Died of Drug Overdose — United States, 2022. MMWR Morb Mortal Wkly Rep. 2024;73(34):747–753. doi: 10.15585/mmwr.mm7334a3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Tanz LJ, Stewart A, Gladden RM, Ko JY, Owens L, O’Donnell J. Detection of Illegally Manufactured Fentanyls and Carfentanil in Drug Overdose Deaths — United States, 2021–2024. MMWR Morb Mortal Wkly Rep. 2024;73(48):1099–1105. doi: 10.15585/mmwr.mm7348a2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Gutwinski S, Schreiter S, Deutscher K, Fazel S. The prevalence of mental disorders among homeless people in high-income countries: An updated systematic review and meta-regression analysis. PLOS Med. 2021;18(8):e1003750. doi: 10.1371/journal.pmed.1003750 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Richards J, Kuhn R. Unsheltered Homelessness and Health: A Literature Review. AJPM Focus. 2023;2(1):100043. doi: 10.1016/j.focus.2022.100043 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Beijer U, Wolf A, Fazel S. Prevalence of tuberculosis, hepatitis C virus, and HIV in homeless people: a systematic review and meta-analysis. Lancet Infect Dis. 2012;12(11):859–870. doi: 10.1016/S1473-3099(12)70177-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Ly TDA, Castaneda S, Hoang VT, Dao TL, Gautret P. Vaccine-preventable diseases other than tuberculosis, and homelessness: A scoping review of the published literature, 1980 to 2020. Vaccine. 2021;39(8):1205–1224. doi: 10.1016/j.vaccine.2021.01.035 [DOI] [PubMed] [Google Scholar]
  • 11.Waddell CJ, Saldana CS, Schoonveld MM, et al. Infectious Diseases Among People Experiencing Homelessness: A Systematic Review of the Literature in the United States and Canada, 2003–2022. Public Health Reports®. 2024;139(5):532–548. doi: 10.1177/00333549241228525 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Montgomery AE, Szymkowiak D, Marcus J, Howard P, Culhane DP. Homelessness, Unsheltered Status, and Risk Factors for Mortality: Findings From the 100    000 Homes Campaign. Public Health Rep. 2016;131(6):765–772. doi: 10.1177/0033354916667501 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Roncarati JS, Baggett TP, O’Connell JJ, et al. Mortality Among Unsheltered Homeless Adults in Boston, Massachusetts, 2000–2009. JAMA Intern Med. 2018;178(9):1242–1248. doi: 10.1001/jamainternmed.2018.2924 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Chang JS, Riley PB, Aguirre RJ, et al. Harms of encampment abatements on the health of unhoused people. SSM - Qual Res Health. 2022;2:100064. doi: 10.1016/j.ssmqr.2022.100064 [DOI] [Google Scholar]
  • 15.Craven K, Sapra S, Harmon J, Hyde M. “I’m No Criminal, I’m Just Homeless”: The Greensboro Homeless Union’s efforts to address the criminalization of homelessness. J Community Psychol. 2022;50(4):1875–1892. doi: 10.1002/jcop.22671 [DOI] [PubMed] [Google Scholar]
  • 16.Herring C. Complaint-Oriented Policing: Regulating Homelessness in Public Space. Am Sociol Rev. 2019;84(5):769–800. doi: 10.1177/0003122419872671 [DOI] [Google Scholar]
  • 17.Herring C, Yarbrough D, Marie Alatorre L. Pervasive Penality: How the Criminalization of Poverty Perpetuates Homelessness. Soc Probl. 2020;67(1):131–149. doi: 10.1093/socpro/spz004 [DOI] [Google Scholar]
  • 18.Robinson T. No Right to Rest: Police Enforcement Patterns and Quality of Life Consequences of the Criminalization of Homelessness. Urban Aff Rev. 2019;55(1):41–73. doi: 10.1177/1078087417690833 [DOI] [Google Scholar]
  • 19.City of Grants Pass v. Johnson. 603 (U.S. Supreme Court 2024). Accessed December 16, 2025. https://www.supremecourt.gov/opinions/23pdf/23-175_19m2.pdf
  • 20.Chiang JC, Bluthenthal RN, Wenger LD, Auerswald CL, Henwood BF, Kral AH. Health risk associated with residential relocation among people who inject drugs in Los Angeles and San Francisco, CA: a cross sectional study. BMC Public Health. 2022;22(1):823. doi: 10.1186/s12889-022-13227-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Goldshear JL, Ganesh SS, Borquez A, Gelberg L, Corsi KF, Bluthenthal RN. Material hardship, forced displacement, and negative health outcomes among unhoused people who use drugs in Los Angeles, California and Denver, Colorado: a latent class analysis. BMC Public Health. 2025;25(1):591. doi: 10.1186/s12889-025-21626-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Meehan AA, Milazzo KE, Bien M, et al. Involuntary displacement and self-reported health in a cross-sectional survey of people experiencing homelessness in Denver, Colorado, 2018–2019. BMC Public Health. 2024;24(1):1159. doi: 10.1186/s12889-024-18681-w [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Barocas JA, Nall SK, Axelrath S, et al. Population-Level Health Effects of Involuntary Displacement of People Experiencing Unsheltered Homelessness Who Inject Drugs in US Cities. JAMA. 2023;329(17):1478. doi: 10.1001/jama.2023.4800 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Zwick H, O’Dea R, Barocas JA, et al. Health and Economic Outcomes of Addressing Encampments of Individuals Using Opioids. JAMA Netw Open. 2025;8(6):e2517095. doi: 10.1001/jamanetworkopen.2025.17095 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Goldshear JL, Kitonga N, Angelo N, Cowan A, Henwood BF, Bluthenthal RN. “Notice of major cleaning”: A qualitative study of the negative impact of encampment sweeps on the ontological security of unhoused people who use drugs. Soc Sci Med. 2023;339:116408. doi: 10.1016/j.socscimed.2023.116408 [DOI] [PubMed] [Google Scholar]
  • 26.Mayer M, Mejia Urieta Y, Martinez LS, Komaromy M, Hughes U, Chatterjee A. Encampment Clearings And Transitional Housing: A Qualitative Analysis Of Resident Perspectives. Health Aff (Millwood). 2024;43(2):218–225. doi: 10.1377/hlthaff.2023.01040 [DOI] [PubMed] [Google Scholar]
  • 27.Westbrook M, Robinson T. Unhealthy by design: health & safety consequences of the criminalization of homelessness. J Soc Distress Homelessness. 2021;30(2):107–115. doi: 10.1080/10530789.2020.1763573 [DOI] [Google Scholar]
  • 28.Chayama KL, Narayan P, McNeil R, Boyd J, McDougall P, Knight R. Displacement and the disruption of drug use patterns and risks: Experiences of people living with HIV in two urban neighbourhoods. Int J Drug Policy. 2026;152:105279. doi: 10.1016/j.drugpo.2026.105279 [DOI] [PubMed] [Google Scholar]
  • 29.Chung EO, Ray B, Humphrey JL, et al. Sociodemographic, behavioral, and geospatial factors associated with syringe services program use in San Francisco, California. Drug Alcohol Depend. Published online January 13, 2026:113046. doi: 10.1016/j.drugalcdep.2026.113046 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Suen LW, Megerian CE, Browne EN, et al. Linkage to substance use disorder treatment through syringe services programs and the criminal legal system: A Cross-Sectional Study. Am J Prev Med. 2025;0(0). doi: 10.1016/j.amepre.2025.108150 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Kral AH, Malekinejad M, Vaudrey J, et al. Comparing Respondent-Driven Sampling and Targeted Sampling Methods of Recruiting Injection Drug Users in San Francisco. J Urban Health. 2010;87(5):839–850. doi: 10.1007/s11524-010-9486-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Watters JK, Biernacki P. Targeted Sampling: Options for the Study of Hidden Populations. Soc Probl. 1989;36(4):416–430. doi: 10.2307/800824 [DOI] [Google Scholar]
  • 33.Bailey ZD, Krieger N, Agénor M, Graves J, Linos N, Bassett MT. Structural racism and health inequities in the USA: evidence and interventions. The Lancet. 2017;389(10077):1453–1463. doi: 10.1016/S0140-6736(17)30569-X [DOI] [PubMed] [Google Scholar]
  • 34.Phelan JC, Link BG. Is Racism a Fundamental Cause of Inequalities in Health? Annu Rev Sociol. 2015;41(Volume 41, 2015):311–330. doi: 10.1146/annurev-soc-073014-112305 [DOI] [Google Scholar]
  • 35.City and County of San Francisco. Unintentional drug overdose death rate by race or ethnicity and year. 2024. Accessed September 17, 2025. https://www.sf.gov/data--unintentional-drug-overdose-death-rate-race-or-ethnicity
  • 36.West BT, Blom AG. Explaining Interviewer Effects: A Research Synthesis. J Surv Stat Methodol. 2017;5(2):175–211. doi: 10.1093/jssam/smw024 [DOI] [Google Scholar]
  • 37.R Core Team. R: A Language and Environment for Statistical Computing. Published online 2025. https://www.R-project.org/
  • 38.StataCorp. Stata Statistical Software: Release 19. Published online 2025. [Google Scholar]
  • 39.Truong AQ, Duffy B, Saleem HT, et al. Navigating public, semi-public, and private drug use environments: A qualitative examination of unsheltered homelessness and drug use in Baltimore City. PLOS ONE. 2025;20(12):e0338107. doi: 10.1371/journal.pone.0338107 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Goldshear JL, Corsi KF, Ceasar RC, et al. Housing and displacement as risk factors for negative health outcomes among people who inject drugs in Los Angeles, CA, and Denver, CO, USA. BMC Public Health. 2026;26(1):753. doi: 10.1186/s12889-026-26200-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Mamdani Z, Loyal JP, Xavier J, et al. ‘We are the first responders’: overdose response experiences and perspectives among peers in British Columbia. Drugs Educ Prev Policy. 2024;31(1):91–104. doi: 10.1080/09687637.2022.2132134 [DOI] [Google Scholar]
  • 42.Carroll JJ, Rich JD, Green TC. The protective effect of trusted dealers against opioid overdose in the U.S. Int J Drug Policy. 2020;78:102695. doi: 10.1016/j.drugpo.2020.102695 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Riback L, Pérez-Correa AE, Ghiroli MM, López-Castro T, Fox AD. Injecting Alone: Practices and Preferences among People Who Inject Drugs in New York City. Subst Use Misuse. 2022;57(13):1988–1996. doi: 10.1080/10826084.2022.2125273 [DOI] [PubMed] [Google Scholar]
  • 44.Rosen JG, Glick JL, Zhang L, et al. Safety in solitude? Competing risks and drivers of solitary drug use among women who inject drugs and implications for overdose detection. Addiction. 2023;118(5):847–854. doi: 10.1111/add.16103 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Qi D, Abri K, Mukherjee MR, et al. Health Impact of Street Sweeps from the Perspective of Healthcare Providers. J Gen Intern Med. 2022;37(14):3707–3714. doi: 10.1007/s11606-022-07471-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.San Francisco Department of Public Health. New San Francisco Program Uses Nighttime Telehealth to Provide Immediate Access to Prescription Medication to Start Recovery. May 22, 2024. Accessed January 16, 2026. https://www.sf.gov/news--new-san-francisco-program-uses-nighttime-telehealth-provide-immediate-access-prescription
  • 47.San Francisco Department of Public Health. Overdose Prevention Plan 2024. Published online December 19, 2024. Accessed January 16, 2026. https://www.sf.gov/reports--december-2024--overdose-prevention-plan-2024
  • 48.Guevarra EC, Hossaini S, Montecillo A. SF Promises to Make Life “Uncomfortable” for People Sleeping Outside | KQED. August 7, 2024. Accessed December 16, 2025. https://www.kqed.org/news/11999056/sf-promises-to-make-life-uncomfortable-for-people-sleeping-outside [Google Scholar]
  • 49.Knight H. San Francisco Takes Harder Line Against Homeless Camps, Defying Its Reputation. N Y Times. August 3, 2024. Accessed December 16, 2025. https://www.nytimes.com/2024/08/03/us/san-francisco-homeless-london-breed.html [Google Scholar]
  • 50.Breed ML. Executive Directive 24–02 “Journey Home: Providing Relocation Assistance Services.” Office of the Mayor, San Francisco; 2024. Accessed December 16, 2025. https://www.sf.gov/sites/default/files/2024-08/Executive%20Directive%2024-02_Journey_Home_0.pdf [Google Scholar]
  • 51.Healthy Street Operations Center. HSOC - Encampment Resolution Policies. City and County of San Francisco. Accessed May 21, 2026. https://www.sf.gov/sites/default/files/2024-10/HSOC%20Encampment%20Resolution%20Policy.pdf [Google Scholar]
  • 52.Coalition on Homelessness. Behind the Healthy Street Operation Curtain. Coalition on Homelessness; 2021. Accessed May 21, 2026. https://www.cohsf.org/behind-the-healthy-street-operation-curtain/ [Google Scholar]
  • 53.Chavez M, Evans T, Shapton C, Flynn A, Sylvestre ME, Blomley N. Life in the ‘Hum’: Belongings and Everyday Dispossession. Int J Homelessness. Published online August 26, 2024:1–17. doi: 10.5206/ijoh.2023.3.16988 [DOI] [Google Scholar]
  • 54.Newsom GG. Executive Order N-1–24 “Addressing Encampments.” Governor of California; 2024. https://www.gov.ca.gov/wp-content/uploads/2024/07/2024-Encampments-EO-7-24.pdf [Google Scholar]
  • 55.Kushel M, Moore T, Birkmeyer J, et al. Toward a New Understanding: The California Statewide Study of People Experiencing Homelessness. UCSF Benioff Homelessness and Housing Initiative; 2023. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Data from this study will be stored in a public data repository at the completion of the research grant.

RESOURCES