scorecardresearch Skip to main content

What our simulation models project will happen after Boston dismantles Mass. and Cass encampments

The results of our analysis are grim.

People pack up their belongings on Nov. 11 as the city works to clear out the encampment at Mass and Cass.Jessica Rinaldi/Globe Staff

There is a crisis at the intersection of Massachusetts Avenue and Melnea Cass Boulevard, also known as Mass. and Cass. City officials have responded with a cleanup of the area, offering temporary shelter and treatment services to those living in the encampment.

Dismantling such encampments seems appealing, since elected officials often believe that action, any action, is laudable. Given their intuitive appeal, there is a long history of “sweeps” in the United States, and this isn’t the first time that Boston has used this strategy in the name of public health. In 2019, then-mayor Marty Walsh and the Boston Police Department launched a similar plan in which dozens of people were arrested near the intersection of Massachusetts Avenue and Southampton Street while others were swept from their tents and forced to move away.


While little evidence exists that sweeps are effective, there is also a dearth of hard evidence that they are harmful. We are public health researchers who work on simulation models of drug use to investigate the impact of policy changes on the overdose crisis. Our models simulate the natural history of injection drug use and how individuals move through the health care system.

In these models, individuals escalate and de-escalate their drug use over the course of their lives, reflecting the reality that addiction is often a relapsing and remitting condition that has impact over the entire lifetime. Individuals may develop complications of drug use, such as overdose, endocarditis, and skin infections — all of which carry the risk of hospitalization and death. The models track those hospitalizations and deaths and also estimate the cost of those events.

People in the models can seek treatment for their opioid use and sterile injection supplies to reduce potential harms. When they do, the models track their clinical progress, simulating the benefits of treatment in terms of decreased overdoses, less criminal activity, and fewer hospitalizations. The models track all the costs of treating people for their addictions, but also the savings society accrues when people with treated substance use disorder do not present to hospitals and clinics with complications of drug use and ICU beds are not being filled by those who have overdosed.


We used our models to investigate the possible effects of the latest cleanup at Mass. and Cass. We sought to provide simulated evidence on critical questions in real time. What will dismantlement do to the expected overdose rate among those currently living in the encampment? What will happen to the hospitalizations for complications of injection drug use? How will retention on treatment for opioid use disorder change one month and one year after a clean sweep? What is the cost when all of the downstream consequences of the action are considered?

We modeled the dismantlement as a sudden disruption of every person living in the encampment. When they are swept up, some people in the simulation are mandated to court-ordered addiction treatment, while the rest are dispersed. We assume that some people who are dispersed will find emergency housing, but that the interruption will result in reduced access to harm reduction services and a higher chance of falling out of addiction care for all people.

The results of our analysis are grim. We project that the cleanup will lead to an immediate 30 percent increase in the overdose rate among those who had been living in the encampment (we found a 20 to 40 percent increase in various simulations depending on the assumptions we made in the models). Further, because sudden disruptions take a long time to equilibrate, the overdose rate will remain elevated for approximately nine months after the sweep. In total, we project about a 12 percent total increase in overdose mortality over the course of the year following a clean sweep. Additionally, our models predict that six months after the simulated sweep, fewer people will be seeking treatment for opioid use disorder in Boston. In the year following the sweep, hospitalizations for complications of drug use will probably rise by 11 to 15 percent for endocarditis, 4 percent for skin infections, and 46 percent for overdose.


We project that by the end of one year, total health care spending for people ensnared in a clean sweep will rise by approximately 13 percent. This at a time when, because of the ongoing COVID-19 pandemic, shelters and treatment centers are still forced to operate at reduced capacity, there are mounting worker shortages, medical examiners are months behind on determining cause of death, and public health departments remain short-staffed and under-resourced.

The core principle that elected officials need to recognize is that the cleanup operation rips some people who had been stabilized on medication-based treatment for opioid use disorder off of those medications, leaving them at particularly high risk of overdose. Even among those who are seeking or are forced into treatment, few remain in treatment when the mandate ends, and when people come off of medications for opioid use disorder, their tolerance for opioids is low and their risk of overdose is high. It also leaves them with limited access to vital harm reduction services, which are important in preventing infections and subsequent hospitalizations.


The city’s ongoing decampment and arrests are not proven public health responses to homelessness or the overdose crisis. Based on our models, creating a diaspora will probably generate more suffering among an already vulnerable population. Mayor-elect Michelle Wu needs to realize that the twin-demic of homelessness and overdose in Boston will be solved with public health solutions that are grounded in evidence-based solutions, including housing, accessible harm reduction services, and voluntary treatment for substance use disorders, intertwined with compassion and empathy for our fellow humans.

Dr. Benjamin Linas is a professor of medicine at Boston University School of Medicine and an infectious diseases physician at Boston Medical Center. Dr. Joshua Barocas is an associate professor of medicine at the University of Colorado School of Medicine.