"Sweeps kill!" exclaimed a chorus of homelessness activists in response to a new study published in early May 2023 by the Journal of the American Medical Association. The paper concluded that encampment sweeps, or cleanups, in the language preferred by Denver city officials, "decrease life expectancy" of unhoused individuals who inject drugs.

These findings received heightened attention in Denver during a mayoral race where the city's homelessness crisis was a leading campaign issue. The paper’s lead author, a professor at the University of Colorado Anschutz Medical Campus, Dr. Joshua Barocas, is an unknown to many, but he is a fierce public critic of the outgoing Hancock administration's policies on encampments and an adherent of Housing First and Harm Reduction dogmas.

Most recently, incoming Mayor Mike Johnston announced the members of his transition team, dubbed Vibrant Denver. Barocas was granted a high-profile assignment as Vibrant Denver’s Co-Chair of the Public Health and Environment Committee.

From the credulous media coverage of the Barocas study, casual readers could be forgiven for believing that researchers must have analyzed past deaths and determined they were directly caused by encampment cleanups. Denverite tugged on heartstrings, invoking the "We Will Remember" annual report on street mortality by Colorado Coalition for the Homeless: "The new study suggests some of those recognized over the years died because of ongoing encampment sweeps." Barocas pulled no punches in a quote for Denverite: “[The study] means our states and our cities are literally killing people.”

But does the paper prove that sweeps kill?

Barocas and his team created a model to simulate the lives of "unsheltered people experiencing homelessness who inject drugs" over 10 years. After analyzing the simulation, the team concluded that a "continual displacement strategy" may increase mortality among unsheltered homeless people who inject drugs by as much as 25%.

This may seem definitive, but a review of the study reveals a troubling foundation, blatantly rooted in the team’s biases.

Barocas and his team used longitudinal survey data from the CDC to develop the simulation. While this CDC dataset segments unsheltered homeless people and injection drug users, it does not, however, track whether a person experienced "involuntary displacement". Absent this data, the team simply inserted their own opinions! They instructed their model that, when encountering a sweep, their simulated people would suffer three negative effects: greater probability of overdose, diminished access to medication for opioid use disorder, and increased syringe sharing. As the Denver Post summarized, "those three [effects] stand in as an avatar for sweeps."

What a shamefully audacious way of baking in their assumptions! Barocas and his team associated a bunch of nasty outcomes with sweeps. What else could their poor model do but predict death for its simulated drug users?

One can easily imagine a competing model created by researchers with an opposing point of view. There’s no doubt that an encampment clean-up is a disruptive event for its denizens, but aren’t there good potential outcomes, too? What if this alternate model was informed that unsheltered people who experience a sweep are more likely to be connected with the city’s generously funded services ($250-million a year, according to the Hancock administration) and are more likely to enter treatment and recovery plans? Such a model would conclude that encampment clean-ups save lives!

The bottom line is that Barocas and his team produced a biased model, fed with biased assumptions, that output a biased result.

And there can be no question about their biases. Perusing their Twitter posts reveals an endless stream of activism for their causes and vitriol for their perceived opponents. These examples are all from April 2023 alone:

            • Barocas castigated Colorado state legislators who didn't vote to authorize supervised injection sites, which he supports, saying "we hold grudges." He elaborated, "I heard the same tired arguments from opponents and the same tired questions from skeptics. These are a charade and everyone knows it. You're not curious. You don't care. Just say it: You don't value the life of someone who uses drugs."

            • On the same topic, Sarah Axelrath said "As more Coloradoans die, we’ll remember who refused to listen to public health research and expert testimony... I am the damn queen of political grudge holding."

            • Axelrath showed her unwillingness to consider alternative viewpoints: "I don’t listen to the opposition testimony anymore. It’s too frustrating to hear the same old stigmatizing, non-evidence based, and sometimes frankly nonsensical arguments against proven harm reduction policies... I understand *you* might be uncomfortable with the idea of [supervised injection sites]. I don’t care."

            • Axelrath also presented slides from her lecture on Harm Reduction where she says, "It is not my job to stop people from using drugs." She described how she follows that image with her laugh line: "that’s actually the job of the police, who have been famously successful at it."

This team believes they have the only opinions that matter, and anyone who doesn't blindly follow them is beneath contempt. Can we be surprised they created a model that spat out the very conclusions they support?

And now Barocas and fellow travelers, such as Lisa Raville of Harm Reduction Action Center and Dr. Sarah Rowan, have a direct line to the incoming Denver Mayor, a line which they’ll use to advocate for the same enabling policies that have been failing in West Coast cities for decades.

Make your voices heard! Tell the new Johnston administration that in Denver, we don’t want to encourage criminality and public drug abuse. We want a new approach to our crisis of addiction and mental illness that focuses instead on treatment and recovery!