CAHOOTS in Denver: Experimenting with Alternatives to Police

A particular challenge for police is that they often end up as the first responders to all sorts of emergency calls, not just those related to an obvious crime.

In some cases, might there be a more effective alternative?

In a paper in the Fall 2021 Journal of Economic Perspectives, Monica Bell makes the case that academic research on policing has focused heavily on the number of police and on methods of deploying police, but much less on alternatives to a police response or on alternative ways of reducing crime (“Next-Generation Policing Research: Three Propositions,” 35:4, 29-48). Here’s one of the examples that she mentions:

But there are a number of other community-based programs or alternatives to traditional policing that remain largely unstudied, even though some of them are becoming models for other jurisdictions across the nation. For example, CAHOOTS (Crisis Assistance Helping Out on the Street) started in Eugene, Oregon, to send two-person clinical response teams to aid people in mental health crises, without relying on armed police officers. Although the program has existed for more than three decades, in summer 2020 it gained national attention and became the model for numerous pilot programs—in San Francisco, Denver, Rochester, Toronto, and more. Eugene’s CAHOOTS program is funded and overseen by the police department, but some other emerging programs are funded and managed separately from police. Despite its long duration—even longer than the violence interruption programs mentioned above—CAHOOTS has never been rigorously evaluated. There are also rich debates over, among other things, how to measure its diversion rate (Gerety 2020). There is a dearth of information and modeling of police-free crisis response, though one hopes that will change as more cities embrace these approaches.

Well, now there is at least one piece of evidence on a CAHOOTS-style program. Thomas Dee and Jaymes Pyne write about the Denver experience in “A community response approach to mental health and substance abuse crises reduced crime” (Science Advances, June 8, 2022. 8: 23). Here’s some background (citations omitted):

Support Team Assistance Response (STAR) program in Denver provides a mobile crisis response for community members experiencing problems related to mental health, depression, poverty, homelessness, and/or substance abuse issues. The STAR response consists of two health care staff (i.e., a mental health clinician and a paramedic in a specially equipped van) who provide rapid, on-site support to individuals in crisis and direct them to further appropriate care including requesting police involvement, if necessary. The design of the STAR program is based on the Crisis Assistance Helping Out On The Streets program developed in Eugene, Oregon.

STAR began operations on 1 June 2020 for a designated 6-month pilot period. During this period, STAR limited its operations to selected 911 calls for assistance in eight purposefully chosen police precincts (i.e., out of the city’s 36 precincts), where the need for STAR services was anticipated to be the greatest. … [A]ll but one of the neighborhoods in the STAR pilot service area are also designated by the city as “displacement-vulnerable” areas, rapidly gentrifying city spaces where poor and otherwise at-risk residents are being pushed out. …

Operators responding to 911 calls for assistance dispatched STAR staff to eligible incidents that were located in the designated police precincts and during the program’s hours of operation (Monday to Friday, 10 a.m. to 6 p.m.). The identification of emergency calls eligible for STAR services relied on two specific screening criteria. First, the incident had to designate at least one of several codes: calls for assistance, intoxication, suicidal series, welfare checks, indecent exposure, trespass of an unwanted person, and syringe disposal. Second, to dispatch the STAR van, there needed to be no evidence that the incident involved serious criminal activity, such as weapons, threats, or violence, or serious medical needs. The STAR team also responded to calls from uniformed police to engage with community members in crisis and initiated engagement in the field on their own. Over the 6-month pilot period, the STAR team responded to 748 incidents or nearly 6 incidents per 8-hour shift. Roughly a third of calls to STAR occurred at the request of responding police, while the rest were due to a direct 911 dispatch or to the STAR team responding independently to a field observation—none of which required a call to police for assistance or for a response to a criminal offense.

Because the program was rolled out only in certain precincts, and only at certain times and days of the week, it becomes possible to compare trends and patterns, looking both at patterns within the precincts and in comparison with the contiguous areas. (Those who want details on how these “difference-in-differences” comparisons are done can scan through the paper itself.) In particular, the authors looked at certain low-level crimes that seemed likely to be interrelated with the kinds of situations to which the STAR team was responding, like disorderly conduct, trespassing, alcohol, and drug use. Again, the STAR team was not being dispatched for serious crimes or medical emergencies.

The authors find that “the service reduced the number of STAR-related offenses in treated precincts by 34% over the 6 months of the pilot phase. … This impact estimate implies that the STAR pilot program prevented nearly 1400 criminal offenses within the eight participating precincts and the 6 months of operation … This program-induced reduction in measured offenses is broadly consistent with the scale of STAR operations. Specifically, the STAR team responded to 748 calls during our study window. At baseline (i.e., during the pretreatment period), each STAR-related incident resulted in an average of 1.4 recorded offenses in treated precincts. This suggests that we should expect 748 field calls by STAR staff to result directly in just over 1000 fewer recorded offenses (i.e., 748 × 1.4 = 1047).”

It’s interesting to note that some of the decline in the number of criminal offenses happened outside the 10-6 timeframe when the STAR program was actually operating, which is consistent with the idea that the STAR interventions didn’t just deal with the immediate issue, but improved the broad situation. Moreover, this approach seems potentially cost-effective:

The total cost of the 6-month STAR pilot program was $208,141 (50). One useful way to frame this public outlay is to note that the corresponding reduction of 1376 offenses implies a cost of $151 per offense reduced. To put this in perspective, the available estimates (8) suggest that the direct criminal justice cost for a minor criminal offense (e.g., imprisonment and prosecuting) averages $646 (in 2021 dollars). In other words, the direct costs of having police as the first responders to individuals in mental health and substance abuse crises are over four times as large as those associated with a community response model. A fuller reckoning of the costs and benefits associated with community response models would also include the costs and benefits associated with any health care brokered by the first responders. For example, police officers may be more likely than community responders to direct individuals in crisis to comparatively expensive emergency room care or to no care at all. Nonetheless, the results presented here suggest that community response models merit careful consideration as a highly cost-effective way to reduce police engagement with nonviolent individuals in crisis and to instead respond with appropriate health care.

Of course, the study has some obvious limitations. For example, one would want to be hesitant about extrapolating from a program that operated from 10-6 on weekdays to what might happen late on a Friday or a Saturday night. In addition, the focus on the number of low-level crimes reported is interesting, but surely incomplete. Yes, it’s good for people not to build up a record of low-level criminal offences. But was the problem that led to the STAR team being called resolved in a way that helped the people directly involved? Did other community members who were affected feel that the problem had been addressed?

With these and other caveats duly noted, the partial results are clearly encouraging. As the authors write:

Nonetheless, the results presented here suggest that community response models merit careful consideration as a highly cost-effective way to reduce police engagement with nonviolent individuals in crisis and to instead respond with appropriate health care.