June 14 – Pairing a mental health expert with a paramedic to respond to low-level, nonviolent 911 calls decreased the number of criminal offenses in downtown Denver, a new study reveals.
These non-police response teams were trained to address problems related to mental health, depression, poverty, homelessness, and/or substance abuse. During 6 months in 2020, the project reduced crimes related to alcohol and drugs, disorderly conduct, and other crimes against people by 34%.
During the same pilot phase, overall crimes dropped 14% in the eight police precincts that took part in the Support Team Assisted Response (STAR) program, compared to precincts where police responded to all types of 911 calls.
The program trained dispatchers to recognize calls where traditional responses may not be needed, including situations without evidence of serious criminal activity such as threats, weapons, or violence.
Also, in about a third of cases, police called the teams themselves. "One of the striking details – a change in the broader respect – is that in some cases, police responded and then called in the mental health-EMT responders," says lead study author Thomas Dee, PhD.
The study was published June 10 in Science Advances.
These teams of mental health experts and paramedics may be less likely to report people as breaking the law, says Dee, a professor in the Graduate School of Education at Stanford University.
"We also found no evidence of negative effects on more serious or violent crimes."
The Broader Political Context
Although programs to give nonviolent 911 calls to non-police responders have been around for years, the idea is "getting more attention now because of the larger context," Dee says.
Sending a health care response team instead of police officers on some calls can appeal to both sides of the debate over policing in the United States, he says. "There is a broad political context, and this program should have broad appeal."
For example, "If your politics are 'back the blue,' you support this. Police officers often say they don't want to respond to these types of calls, and delegating them [to other responders] could raise police morale," he says.
On the other hand, he says, "if your politics are 'defund the police,' this program decreases the footprint of police officers and could decrease police budgets over time."
Police spend a substantial amount of their time responding to nonviolent emergency calls for assistance that often involve people having mental health or substance abuse crises, the researchers wrote.
But police are often not trained to help in these kinds of situations.
"As a result, emergency calls for assistance may be engaged as criminal violations, sometimes with unnecessarily violent or even tragic consequences, when they can be better addressed as health issues,” they wrote.
What About Safety?
Safety of the health care responders is a "perfectly legitimate concern," Dee says, adding that one concern is that if you don't have a police officer present, the situation could escalate.
But this “is not an off-the-shelf reform. You still need careful training, need to recruit the right people and have good protocols to keep them safe. And you still have to coordinate with the police at times," he says.
Researchers found that none of the 748 field calls handled by the health care responders ended in an arrest, resulting in 1,376 fewer criminal offenses reported over 6 months.
The STAR program cost $208,151 to run during that time, or $151 for every offense prevented, Dee says. The cost to arrest and process someone, meanwhile, is about four times more, or about $646 in 2021 dollars.
"Even if the program was not money-saving, it's a deeply humane way to respond,” Dee says.
Train, Co-Respond, or Replace?
The STAR program in Denver is one of three main options for responding to nonviolent emergency calls. Other approaches include training police officers on how to respond to someone in a mental health or substance abuse crisis as special crisis intervention teams or sending mental health practitioners along with police.
The study’s results "show that a community response [can be] effective to de-escalate crises and to improve the reference process," says Étienne Blais, PhD, who has studied similar programs in Canada.
There can be pros and cons to police presence on the co-responding teams, says Blais, a professor in the School of Criminology at the University of Montreal.
"For instance, the presence of a police officer allows mental health specialists to provide psychosocial consultation and apply de-escalation techniques even if the person in crisis shows signs of aggressiveness," he says. "On the other hand, some persons will not disclose sensitive but relevant information – such as drug-related issues or involvement in criminal offenses – in the presence of police officers."
The Right Team Can Make All the Difference
"It is one of the first studies to demonstrate the effectiveness of non-armed responses to behavioral health calls for service," says Bradley R. Ray, PhD, a senior researcher in the Division for Applied Justice Research of RTI International.
But it's likely not the last. "I would expect to see many more studies like this coming in the future that show the potential for non-police responses," he says.
Using the program in other communities might be challenging, given the different designs of dispatch systems nationwide, Ray says. For example, in some cases, a sheriff runs the system, while others are controlled by city governments that require fire, emergency medical services, and police to respond to all calls.
"Over the past couple of years, I have continually encouraged law enforcement agencies to carefully consider the calls where they need police officers, where they need co-responses, and where they need not respond and leave it other agencies," says Ray, who was also senior author on a study published in April that compared police-mental health co-response teams to traditional police responses in Indianapolis.
"In my own research on co-response teams, we have found minimal and often non-significant differences between the co-response and police responses as usual," he says. "When agencies are frustrated with those outcomes, I often tell them it is about selecting the right calls for service and sending the right team."