Encampment Sweeps Were Linked to Fewer Transitions Out of Homelessness
Encampment sweeps did not behave like a neutral street-management tactic in this Vancouver cohort. Among unhoused people who use drugs, being subjected to municipal confiscation of personal belongings was associated with a lower likelihood of no longer reporting homelessness during follow-up, and forced displacement showed an even stronger negative association.
The findings land at the intersection of housing policy, harm reduction, and survival. Personal belongings are not incidental in this setting; they often include identification, medications, clothing, food, sleeping materials, and the basic structures that make life outdoors minimally stable. Removing those items can sever the small logistical supports that help people stay connected to services, keep documents intact, or mobilize toward a housing placement when one becomes available.
Prospective Cohort Data From Vancouver Followed Through 2025
Investigators drew from two harmonized prospective cohort studies of people who use drugs in Vancouver, with semi-annual interviews conducted between December 1, 2021 and May 31, 2025. The analytic sample included 235 participants. Baseline data showed that 59 participants, or 25.1%, reported confiscation of belongings. In the secondary analysis covering a later period, 50 of 154 participants, or 32.5%, reported forced displacement.
The primary outcome was whether participants stopped reporting homelessness in the prior six months during follow-up. Overall, 128 participants, or 54.5%, no longer reported homelessness during follow-up, corresponding to an incidence rate of 55.6 per 100 person-years, with a 95% confidence interval of 46.4 to 66.1. The analysis used proportional hazard marginal structural modeling to adjust for time-varying confounding, including demographic characteristics, drug use patterns, health status, and other social and structural exposures.
A single statistic carried the central message: confiscation of belongings was associated with a substantially lower probability of leaving homelessness.
Confiscation and Displacement Both Pointed in the Same Direction
In the proportional hazard marginal structural model, confiscation of belongings was associated with an adjusted hazard ratio of 0.35, with a 95% confidence interval of 0.13 to 0.94 and a p value of 0.036. Forced displacement was associated with an adjusted hazard ratio of 0.07, with a 95% confidence interval of 0.02 to 0.20 and a p value below 0.001. Both exposures were negatively associated with transitioning away from homelessness.
The separation between those hazard ratios is clinically and socially meaningful. Forced displacement, even when it did not necessarily involve confiscation, likely disrupts location stability, peer networks, outreach contact, and the repeated routines that allow people to access shelters, case management, or other housing pathways. Confiscation adds a second layer of damage by stripping away documents, medication, clothing, and makeshift shelter materials that are often hard to replace quickly. In a population already exposed to high drug-related risk and unstable access to care, those losses can trap people in a cycle of repeated movement without real housing gain.
Why the Mechanism Matters at Bedside and Policy Levels
Housing transitions rarely occur through one dramatic event. They usually depend on cumulative readiness: a phone that works, identification papers, transport, medication continuity, an intact connection to a health worker, and enough stability to complete forms or attend appointments. Encampment sweeps can interrupt every one of those steps at once. When belongings are seized, people may lose the very materials that let them prove identity, protect medications, or maintain a familiar sleeping arrangement while waiting for shelter or housing placement.
That mechanism is consistent with the study’s conclusion that sweeps may undermine pathways to housing rather than accelerate them. For clinicians and outreach workers, the implication is immediate. A patient who has recently been displaced may have a sudden worsening in medication access, clinic attendance, personal safety, and administrative readiness for housing applications. A housing system that depends on paperwork, continuity, and repeated contact becomes harder to navigate after a sweep, not easier.
Observational Design Limits How Far the Results Can Be Extended
The investigators were careful about what the data can and cannot prove. This was an observational study, so causal inference remains limited despite the use of marginal structural modeling. Time-varying confounding was addressed analytically, but the design cannot fully rule out unmeasured differences between people who experienced sweeps and those who did not.
A second limitation cuts toward overestimation of housing transition. The cohort may have included unhoused people who were more connected with services than the broader population, which could make rates of leaving homelessness look higher than they are in the community at large. The study also did not capture the quality of the housing people moved into, which matters because unstable or unsafe housing can lead to return to homelessness. Without that information, a transition out of homelessness cannot be assumed to mean durable or adequate housing.
Even with those caveats, the direction of effect is hard to ignore. Both confiscation and forced displacement moved in the wrong direction for housing stability.
Policy Pressure Shifts Toward Dignified Housing, Not Disruption
The data support a straightforward policy reading: practices that remove people from public space while confiscating their possessions are unlikely to create durable exits from homelessness. They may instead deepen the same instability they purport to manage. In a city where homelessness continues to rise and point-in-time counts likely understate the true burden, the housing response has to be judged by whether it produces stable housing, not by whether it clears an encampment temporarily.
For public health systems, this study also reframes encampment sweeps as more than a policing or sanitation issue. They operate as a structural exposure with downstream effects on housing trajectory, health service contact, and survival. The outcome is not simply displacement from one place to another. It is the possibility of being pushed further from the conditions required to secure a home.
Future work needs to examine the quality of housing placements and whether certain forms of housing are more likely to prevent return to homelessness after a sweep. The current study leaves that question open, along with the broader issue of what happens after people are moved but not meaningfully housed.