Tim Hoff: Welcome to Ethics Talk, the American Medical Association Journal of Ethics podcast on ethics in health and health care. I'm your host, Tim Hoff.
We know that your interactions with law enforcement can influence your health status. As first responders, law enforcement officers are tasked with serving those in need and protecting those in danger. But a lack of consistent training and guidelines about whether, when, and how to serve and protect can end up harming individuals and communities in need.
Policing intersects with other social determinants of health by influencing access to health services and social support systems and can also directly influence morbidity and mortality, especially in encounters with persons experiencing a mental illness crisis. A report by the Treatment Advocacy Center suggests that people with untreated mental illness are 16 times more likely than other civilians to be killed by police officers. An analysis of police killings in the Annals of Epidemiology showed that police are more likely to shoot and kill unarmed Black men demonstrating symptoms or behaviors of mental illness than White men expressing similar symptoms or behaviors.
Integrating mental health professionals into law enforcement has potential to improve both disaster responses and to make an encounter with a person experiencing mental health crisis less likely to escalate to use lethal force. So, clinicians' integration into law enforcement response is part of many regions' tactical operations to community service and protection through law enforcement. But the methods for training and deploying tactical clinicians are numerous and sometimes competing. One prominent model is the Crisis Intervention Team: training that creates connections between law enforcement and mental health clinicians.
On this episode of the podcast, we'll be talking with Dr Amy Watson, a professor at Helen Bader School of Social Welfare at University of Wisconsin, Milwaukee, about how Crisis Intervention Teams can motivate efficiency and equity in tactical responses to 911 calls, and what community mental health intervention might look like when we think beyond the limits of law enforcement response. Dr Watson, thank you so much for being here. [music fades out]
Amy Watson, PhD: Great to be here.
Hoff: So, the debate on how to best integrate care for persons experiencing a mental health crisis is obviously ongoing and very much in the public eye at this point. And some propose that training police officers themselves to de-escalate crises is best, and others suggest that mental health clinicians in field work being key members of Crisis Intervention Teams is preferable. Can you talk us through what you see as the merits and drawbacks of these two approaches?
Watson: Sure. And I think there's value to both approaches, and they're not mutually exclusive.
Watson: I also think there's sort of a third option as well as having clinicians and other crisis responders that provide crisis response without law enforcement.
Watson: But certainly, we need to make sure that law enforcement is prepared to respond safely and effectively and compassionately when they are needed to respond to a mental health crisis event. So, even if we have clinicians that work with officers or mobile crisis teams that don't involve law enforcement, there's certainly times that crisis events have a safety issue that require police assistance. So, they certainly need to be trained, and we have several effective models of actually preparing officers to provide a safe and effective response.
In terms of having clinicians actually be part of crisis response teams, I think there's also quite a bit of value there as well. We've seen data that suggests that having a clinician provide response can reduce unnecessary emergency department runs and also can help better connect people to services in the follow-up period after a crisis event. So, certainly there's value of getting a clinician on scene to provide response.
What we're seeing now as well is that there are communities that are launching effective responses that don't always involve law enforcement. The most well-known example of that right now is the CAHOOTS [Crisis Assistance Helping Out On The Streets] model. And while there hasn't been formal research on it, their analysis of data suggests that they've been able to divert 5% to 8% of the calls for service that the Eugene Police Department gets and provide response, very rarely needing to ask for law enforcement assistance when they do respond.
Hoff: Can you briefly describe what that CAHOOTS model is for listeners who are unfamiliar?
Watson: Absolutely. So, the CAHOOTS model has actually been around for over 30 years in Eugene, Oregon, but it's a team that usually involves a crisis worker—oftentimes it's someone who's bachelor's level or equivalent experience that's had crisis response training—and a medic. So, that could be an EMT [emergency medical technician] or nurse, and they can provide response. They carry police radios, and they can be dispatched by the 911 center in Eugene to calls that are identified as a mental or behavioral health type of situation that doesn't have, you know, there's no evidence of a weapon or a significant safety issue. And then they can go out and provide response and kind of work with the person and determine what resources and supports that person needs.
Watson: And their data suggest that they are able to provide response. And on very rare occasions do they get out there and determine that they need some assistance from law enforcement for a safety issue or if there's a need for an involuntary hospitalization and transport. Which they only provide voluntary services. So, their data suggest that it is possible for many of the situations that police are often called on to respond to could be handled by teams that don't involve law enforcement at all.
Watson: And I think what we're seeing is that all of these approaches have value, and ideally that we have a variety of options that we can triage and match to the appropriate call. So, when there is a safety issue, we need to make sure that the officers that go out there are appropriately prepared, that oftentimes, if agencies have CIT [crisis intervention team]-trained officers. There are situations, certainly, when a co-responder team with a clinician is useful. My biggest worry about that model is that we'll stop there and not develop some of the options that don't involve law enforcement at all.
Watson: And we're seeing growing evidence that there probably is a good portion of calls that could be handled without any involvement of law enforcement.
Hoff: It occurs to me that our listeners and I would benefit from a high-level overview of the kinds of responses that are perhaps most common. You've mentioned the CAHOOTS model, mobile crisis teams. I'm not sure if there's overlap there with crisis intervention teams. So, if you could call out a couple of those more prominent approaches, I think that'd be useful in this conversation.
Watson: Sure. So, the crisis intervention team model, many people have heard about it and think about it as a training model, which it includes training to prepare officers to respond to mental health calls. The original model has it that it's officers that are selected and self-select; they want to become CIT officers.
Watson: And then when a mental health call comes in, they're dispatched to the call. But they're prepared then to provide response and are more familiar with resources. So, that's one model.
What we're seeing in many places that are implementing co-responder models, that's typically an officer, oftentimes a CIT-trained officer, and a mental health clinician that ride together in a police car, and they may be dispatched to hot calls—so, as the first response—or they may be available if an officer goes out, determines that it's appropriate to call in the co-response team, they may be a secondary response. And in many places, the co-responder teams will provide follow-up to calls as well to make sure people are connected to care.
Watson: Mobile crisis teams typically are just clinicians. They may include peers as providers as well, so people with lived experience of mental illness that have gone through a certification and training process. And oftentimes they aren't dispatched via 911. They're dispatched via a behavioral health clinic crisis line in that capacity. And in many places that have mobile crisis teams, they're not funded to be able to provide 24-7 immediate response. So, they may not be quite as immediate as sort of a 911 first response. But there's some data that suggest they can be effective for getting out there, making determinations, and preventing unnecessary emergency department visits, but also then providing connection to care for people.
And then the CAHOOTS model is basically a response model that is 911 dispatched. And the idea is that it's a non-law enforcement response to calls that don't require law enforcement. It's not necessarily licensed, master's level licensed, clinicians that you might see on a mobile crisis team.
Watson: It's crisis workers and the medic that will go out and provide response and people to care.
So, we have sort of these different models, and in larger communities that are able to do it, we're starting to see them kind of have a continuum of options. So, they may have a CIT program and CIT officers, they may have some co-responder teams, and then many of these communities are also then looking at what can we develop similar to a CAHOOTS type model too, because many people are asking for something that doesn't involve law enforcement if there's not a specific need to have a police officer there.
Hoff: Sure. Yeah, that's very helpful. Thank you for giving that rundown.
In a 2019 paper that you and your coauthors published, you note that Crisis Intervention Teams' impact on safety outcomes is perhaps limited. What can you tell us about the data that currently exist and the current state of the research in this field?
Watson: So, currently, we have several studies that suggest that CIT officers may use less force. I did a study in Chicago where the CIT officers used less force with more resistant subjects. So, it was somewhat of an interaction effect. There's been other studies that didn't find a statistically significant effect on use of force.
And really, as a researcher, this has been a difficult question to tackle, looking at uses of force for a couple of reasons. One is that while very important outcomes to look at, uses of force are statistically rare events. So, you need sort of a larger sample to really look at whether or not you're going to have, if there's a difference related to a CIT response versus a non-CIT response. If it's statistically significant, you need a larger sample of calls.
Watson: The other issue is, is just how force is documented. So, to get that larger sample, we probably want to use police department administrative data. And departments vary in terms of what their policies are and what types of force need to be documented and whether or not that documentation also identifies whether it was a mental health-related call. So, I might be able to get all of the data on all the uses of force, but I might not know which of those calls actually were mental health related. So, then it becomes difficult to look at.
To further kind of confound this, there was Michael Compton, who is a psychiatrist and researcher at Columbia University. He did a study where he did look at uses of force. He collected data from officers. And one of the count-ons in there is that they counted use of handcuffs as a use of force; however, many agencies have the policy that if someone is going to be transported to the hospital or transported for care, they have to be handcuffed. So, if a CIT officer is more likely to actually take the step to get the person to care, they're also then more likely to have to handcuff the person.
Watson: And that gets counted as a use of force. So, he did not find that CIT officers were using less force, but he had that in the mix. What he did find was that when force was used, including those handcuffing situations, CIT officers were less likely to also arrest the person.
Watson: So, again, it's been a really kind of difficult nut to crack based on the data that we would need to look at it and the level of variation in terms of how that data's recorded.
Hoff: So, in addition to making interactions between persons in crisis and law enforcement safer, Crisis Intervention Team training, as you've noted in your own research, has the promise for reducing stigma toward people experiencing mental illness. Can you talk a little bit about that and tell us if the data gives us reason to hope that Crisis Intervention Teams actually do that?
Watson: Absolutely. So, part of the training that officers go through to become a CIT officer includes opportunities to interact with people with lived experience of mental illness. So, CIT trainings will bring in a number of people with lived experience to talk to officers and talk about their experiences. And oftentimes, they'll do like a lunch where officers then can more informally interact with people. They also bring in family members that are caretakers of people with lived experience of mental illness, and so officers get to meet with them as well. CIT trainings also have officers do site visits to mental health provider agencies, and there, they get to interact with staff, but also with people who are patients as well.
And what we know about stigma is that one of the best ways to reduce stigma is through contact, is getting to know people and learn about their story and their experiences. And so, that is built into the training, and we do. We have good evidence from a number of different studies that have looked at this, that show that CIT training actually does reduce stigma and improve knowledge about mental illness. And Michael Compton has done a number of studies where he's looked at that, and he's looked at CIT-trained officers sometime after they completed the training. And his findings suggest that it's a durable effect that lasts over time.
Watson: There's also been research that's been done by other researchers in different places that support the same finding, that it really is effective for reducing stigma and improving knowledge and attitudes. So, that's a pretty robust finding.
Hoff: That is. That, [chuckles] that does give me hope.
Hoff: Do you know if other intervention models have shown similar kinds of benefits outside of sort of immediate safety outcomes?
Watson: I haven't seen as much research. I mean, part of the issue is just what's been researched.
Watson: So, certainly, I don't know that anybody's looked specifically at stigma with co-responder models. There's a number of agencies that are having all of their officers go through mental health first aid training, and there's a version of it for first responders. And I have not seen research specific to that version of the training, but there is evidence in the initial model of mental health first aid training that suggests it can reduce stigma.
Watson: But I haven't seen a lot of research on other trainings in terms of looking at impacts on stigma.
So, you end this 2019 article by considering a drawback that you've touched on already in a couple of your responses, and you suggest that CIT is better framed as an intervention that can reduce the need for law enforcement involvement entirely in some cases. Can you talk a little bit more about that and why that might be sort of a better goal than simply training officers to de-escalate or respond better to mental health crises?
Watson: Sure. So, the CIT model really is a model to bring partners together, so law enforcement, the advocacy community, including people with lived experience and family members—so, often NAMI [National Alliance on Mental Illness] groups are involved—and then the mental health system. And really partnering to not only prepare police officers to better respond when they're needed, but to look at sort of the crisis response system and look at where the barriers and gaps are and work together to actually improve that system. So, now that we're starting to see a lot more attention on developing crisis response systems overall, communities that have preexisting strong CIT programs already have CIT steering committees and councils that are stakeholders that have been working together to look at this. That's really a foundation of the model.
So, I think in that way, CIT programs are well positioned to really work towards making sure that officers are prepared to respond when necessary and that the partnerships are there as well. But also, just to start developing services on the health systems side, stronger services for first response so that law enforcement can take the assist role as opposed to the primary role assisted by the mental health system. So, I see that as being an important piece.
And I've had this kind of realization in my own work is that I've really focused on the police response side of it. And what I've learned from that is that police are responding to a lot of things that really should be more the domain of the mental health system. And they've worked really hard because they've had to respond to improve how they do that. And I see oftentimes agencies that are now hiring clinicians, it's because that's part of what's within their control to develop, to provide better response. But I think a CIT program and partners can also really work together to start developing things on the mental health systems side so that when a call comes in that there's options besides always sending a police officer.
Watson: Because we know that there's a lot of situations that don't require it. But we also know that there will always be some situations that do. And a CIT program and similar partnerships can really work together to make sure that law enforcement and the health system can partner when it's necessary.
Hoff: Mmhmm. That's very interesting. Thank you.
Our upcoming March issue is on abolition medicine, and it seems like there might be a fair amount of overlap in the work being done around law enforcement responses to health crises and the examination of carceral logic in health care. Do you find that to be the case, or are those conversations not quite happening yet?
Watson: And I mean, I think that's…. One of the things that I think makes the CAHOOTS model particularly exciting because their approach is really more of a voluntary approach. It's not quite as medical model as something that's more reliant on master's level-licensed clinicians and more really medically centered.
Watson: And so, I think that if we can take a step back and really go back to what do communities, what are they asking for, what types of responses, and really working to develop that as opposed to kind of having it entirely based on what experts think we should provide for people because we know better.
Watson: Because we don't always know better.
Watson: So, I do see, I mean, and I think sort of that movement really needs to be part of the discussions around developing what types of responses we can develop because I have seen that too. I've talked to communities that are really kind of working to look at what they can do that doesn't involve law enforcement. And they're not, they're also not entirely comfortable just leaving it within the mental health system because of prior experiences.
Watson: So then, thinking about what that means, and then how can we build something that really addresses that concern but can bring in that more formal mental health side when it's appropriate and when it's useful to the people that we're responding to.
Hoff: Mmhmm. Yeah, that's very interesting. Thank you for laying that out. It seems like for someone who may have had a bad experience with mental health care in the past, it's essentially the risk of interacting with law enforcement who might not be properly trained to respond to mental health crises or dealing with mental health professionals who rely on those coercive measures or who aren't operating out of that model of community engagement and involvement that you're talking about.
Watson: Yeah, and I've seen police officers treat people much better than the clinician that they hand the person off to.
Watson: And so, there are police officers that really do a good job. And I mean, there are certainly many clinicians that do a good job too.
Watson: But the mental health system has a lot of work to do as well.
Watson: I mean, it's perpetuated inequities. I think some of it, it's too easy to take the shortcut of using coercive measures when you know you're short on resources, but we really need to figure out how to address that.
Watson: Otherwise, people will continue to stay away, and we miss the opportunity to provide meaningful supports and services.
Hoff: Mm, mmhmm. It might be of interest to you and to our listeners that our April 2021 issue was on so-called compassionate uses of force, which is very much part of this conversation. Our upcoming—well, it's upcoming as of this conversation, but it will have been published by the time this podcast comes out—November issue on health care for people experiencing homelessness and specifically, the podcast is on street psychiatry: the provision of mental health services to people where they are. And then as I said earlier, our upcoming March issue is on abolition medicine. So, it seems like all of these topics are adjacent, and people are more interested in having these conversations.
Watson: Yeah. And I do think we're actually in a really exciting time that the discussions that are happening now, probably we needed to start having a long time ago, but we're having them now. And I think there's more attention and potentially some resources to move things forward.
Hoff: Absolutely. So, given the increasing number of health professionals working with law enforcement in the field through whatever model, what do you think are three things that health professions students and trainees should know if they're interested in moving into this tactical health care space?
Watson: Yeah. I have a hard time. I bristle a little bit with the term “tactical health care.”
Hoff: I was wondering how sort of common that is. Can you explain a little bit your thoughts on it?
Watson: Yeah, I haven't heard it used a lot. But one of the things that I see, and I actually have some pictures I use in presentations with some of the co-responder teams where police agencies will hire a clinician, is that the clinician starts to dress in tactical type clothing and starts to kind of morph into looking like a police officer.
Watson: And they are working oftentimes for a police agency under those policies.
Watson: And I think I find that problematic because I think the idea of having the clinician there is to have a different approach available. But it's so easy, I mean, to some extent. I think some people just think it feels cool. I know lots of people who work with law enforcement that aren't actually officers that like to buy tactical pants and start to dress that way.
Watson: But I think it is problematic if we're really trying to remember that we're providing a health care response to people, that the clinician's role is really the health care response side. And if you're going out to situations that may have some risk, obviously, they have to be aware of safety. But in some places, you can't tell who the clinician is and who the officer is, and I find that difficult. So, what I would want health care providers that are getting into this space is to kind of remember that role.
Watson: And remember just how your presentation to the person who's in crisis has a big impact on they feel safe to engage.
Watson: And if you look really tactical, and they have an issue with law enforcement, or having law enforcement there makes them feel more vulnerable, if you also look like law enforcement, you missed an opportunity to actually provide them with a sense of safety from that perspective.
Watson: So, I think that's a really important thing to remember.
Watson: And again, I would just encourage people getting into this space also is to make sure that you take opportunities to get to know people with serious mental illness when they're not in crisis. And making sure really to kind of remember the person that you're responding to when they're in their worst moment and in crisis is still that person that could be your child, your spouse, your parent. It could, that person is somebody's family member, does have a whole kind of identity outside of that crisis.
Watson: And I think that's really important, too. And by kind of making sure that people stay grounded in that can be really important as well. And I mean, I see that also with police officers, that many that become CIT officers, they become CIT officers because they already have that experience, and they feel that that's an effective piece of kind of who they are that they can bring to better respond to people.
Watson: So, I would definitely encourage people in the health care space that are getting into that.
The other thing is, I work with, I'm in the College of Social Work—I train social workers—is to also kind of take some time to understand if you are working with law enforcement, get to understand a little bit about their culture as well.
Watson: Because it's very different than the health care setting too, and you can more effectively work with them if you take some time to understand that as well. [theme music returns]
Hoff: Great. Well, Dr Watson, I really appreciate you coming onto the show and this conversation. Thank you very much for being here.
Watson: Thank you for having me. I enjoyed talking with you.
Hoff: That's our episode for the month. Thanks to Dr Amy Watson for being with us. Music was by the Blue Dot Sessions. To read the full issue, visit our site, JournalOfEthics.org. And for all of our latest news and updates, follow us on Twitter and Facebook @Journal of Ethics. We'll be back next month with an episode on abolition medicine. Talk to you then.
Credit Designation Statement: The American Medical Association designates this enduring material activity for a maximum of 0.50 AMA PRA Category 1 Credit(s)™. Physicians should claim only the credit commensurate with the extent of their participation in the activity.
CME Disclosure Statement: Unless noted, all individuals in control of content reported no relevant financial relationships.
If applicable, all relevant financial relationships have been mitigated.