Skip to main content
SearchLoginLogin or Signup

Shedding Light on the Dark Figure of Police Mental Health Calls for Service

Policing: A Journal of Policy and Practice: https://doi.org/10.1093/police/paac006.

Published onFeb 15, 2022
Shedding Light on the Dark Figure of Police Mental Health Calls for Service
·

Abstract:

Recent discussions around police reform have acquired a significant degree of traction. Within these discussions have been calls to remove the police as primary responders to calls involving persons with perceived mental illness (PwPMI). While previous research shows that ~1% of all calls for service involve PwPMI, limitations around police data recording practices likely mask the true proportion of PwPMI within and across calls for service. Accordingly, following manual review and text search of qualitative data appended to all calls for service made to a Canadian police service in 2019, we sought to: (1) identify the true proportion of calls for police service that involve PwPMI; and (2) predict the extent to which PwPMI are involved within and across different call classifications. Our findings reveal that while the ‘Mental Health’ call classification only comprised 0.9% (n = 397) of calls for service, PwPMI were in fact involved in 10.8% (n = 4,646) of calls. Further, logistic regression models reveal that PwPMI are more likely to be involved in certain call classifications relative to others. Implications for police practice and reform are discussed.

Keywords: Police; Mental Health; PwPMI; Police Reform; Dark Figure; Co-Response

Corresponding Author: Jacek Koziarski – [email protected]

This is a pre-copyedited, author-produced version of an article accepted for publication in Policing: A Journal of Policy and Practice, following peer review. The version of record, Koziarski, J., Ferguson, L., & Huey, L. (2021). Shedding Light on the Dark Figure of Police Mental Health Calls for Service. Policing: A Journal of Policy and Practice, is available online at: https://doi.org/10.1093/police/paac006. When citing, please cite the version of record.

Introduction

In recent decades, the police have been increasingly tasked with the role of ‘street corner psychiatrist’ as a result of growing interactions with persons with perceived mental illness (PwPMI)1 (Teplin & Pruett, 1992; Vaughan & Andresen, 2018). There is no single explanation for the increase in said interactions. Instead, a combination of numerous systemic and social factors exists, such as failures to establish adequate mental health care and intolerance of social disorder. These, in turn, have placed the onus on the police to become the de facto response to PwPMI in the community (Lamb & Weinberger, 1998; Markowitz, 2011). Indeed, having the police in this position has resulted in expressions of strong concern, not only from the police, but also policy-makers and the community writ-large (Livingston et al., 2014; Standing Committee on Public Safety and National Security, 2014), and especially so since the recent ‘defund the police’ movement wherein calls have been made to remove the police as primary responders to PwPMI and put civilian-based responses in their place (Koziarski & Huey, 2021; Lum et al., 2021; Watson et al., 2021).

Recent estimates typically suggest that approximately 1% of all calls for police service involve PwPMI (Hodgkinson & Andresen, 2019; Koziarski, 2020; Langton et al., 2021; Livingston, 2016; Lum et al., 2021; Vaughan et al., 2018; White & Goldberg, 2018). However, these are likely underestimates of the actuality of such interactions and the extent to which PwPMI are involved in calls for service, primarily due to two factors. First, call-takers/dispatchers and frontline officers do not possess the skills to reliably identify whether an individual does or does not have a mental illness (Bohrman et al., 2018; Livingston, 2016). Bohrman and colleagues (2018), for instance, identified that officers may inadvertently perceive symptoms of substance use disorder as symptoms of mental illness and vice versa. Consequently, calls for service that should be classified under the ‘Mental Health’ call classification within police data may be misclassified as another call-type.

Second, and most pertinent to the present study, how calls for service data are themselves recorded also likely contributes to an underestimation of PwPMI call proportion. More specifically, in many jurisdictions, calls for service are commonly classified based on the primary nature of a call. The nature is first established by the call-taker/dispatcher based on the information they receive during a call and then updated by the responding officers based on the information obtained at the scene (Ratcliffe, 2021; Simpson, 2020). Call classification solely based on primary nature can mask other, potentially important components of a call, such as whether the call involved PwPMI (Hodgkinson & Andresen, 2019; Koziarski, 2020; Shore & Lavoie, 2018; Vaughan et al., 2015). In light of this, fulsome police reform discussions, particularly as they relate to the role of the police in responding to PwPMI, cannot be had as our knowledge with respect to the true proportion of PwPMI involvement within police calls for service is likely an underestimation. Put differently, removing the police from responding to the ~1% of calls for service that fall under the ‘Mental Health’ call classification will likely only remove police from a small proportion of calls that indeed involve PwPMI and thus will do little in terms of uncoupling the police from responding to situations that involve PwPMI.

Fortunately, scholars have begun thinking about how to potentially identify PwPMI that may be ‘hidden’ within police data. Hartford and colleagues (2005), for instance, developed an algorithm for this purpose. Their algorithm is based on identifying PwPMI through individual-level flags (e.g., 'suicidal tendencies,' ‘mental instability’) and mental health-related addresses (e.g., psychiatric facility). If either of these conditions are met, a text search is conducted across the data to look for terms that suggest the presence of PwPMI. While this particular algorithm has been employed in subsequent work (see, for example, Crocker et al., 2009; Hoch et al., 2009), we note that it does encompass some significant limitations that can equally contribute to an underestimation of PwPMI within police data (also see, for example, Huey, Ferguson, & Vaughn, 2021).

First, and as the authors themselves acknowledge, the partial dependence of the algorithm on mental health-related addresses effectively omits individuals who reside at home or are experiencing homelessness (Hartford et al., 2005). This is a critical limitation given that not only have PwPMI calls for service been found to disproportionally occur at residential addresses (Charette et al., 2011; Shore & Lavoie, 2018) but also because individuals experiencing homelessness disproportionally experience mental health problems and are at greater risk of coming into contact with the police (Koegel et al., 1988; Kouyoumdjian et al., 2019).

Second, and specifically with respect to calls for service data, not all police services have the ability to 'flag' a particular call or individual. While this may be the case for an occurrence report—which, as Hartford et al. (2005) explain, is a more formal report that may stem from a call for service—officers may use their discretion to not generate an occurrence report after a call because it may, for example, have been informally resolved. Indeed, previous research has shown that calls for service generally, including most PwPMI calls for service, conclude with no arrest, apprehension of PwPMI under mental health legislation for assessment, or production of an occurrence report (Charette et al., 2011; Lum et al., 2021; Watson & Wood, 2017). Therefore, if an officer were to attend a call involving PwPMI but could not set a 'flag' for the call and used their discretion not to generate an occurrence report, such police-PwPMI interactions would be omitted by the algorithm. Combined with the address dependency, both limitations of the algorithm developed by Hartford and colleagues (2005) likely contribute to an undercounting of PwPMI in police data.

The final limitation of this particular algorithm is that the text search only occurs after an individual has already been identified via a flag or mental health-related address (Hartford et al., 2005). This, therefore, precludes searching for calls involving PwPMI that do not meet either of these criteria, thus again contributing to an underestimation in the proportion of PwPMI calls. More recently, however, scholars have drawn upon less restrictive methods of PwPMI identification within police data to arrive at a more thorough understanding around the proportion of calls for service that involve PwPMI. For instance, Langton and colleagues (2021) examined 3.6 million calls for service made to the Greater Manchester Police in England between January 2014 and September 2017 and found that PwPMI calls—identified through a specific call classification—only comprised 1.9% of all calls during this time, which is largely in-line with previous research (Hodgkinson & Andresen, 2019; Koziarski, 2020; Langton et al., 2021; Lum et al., 2021; Vaughan et al., 2018; White & Goldberg, 2018). However, by using an artificial intelligence approach to mine calls for service text logs, ignoring flags or addresses associated with any particular call, they instead found that a striking 9.9% of calls for service involved PwPMI (Langton et al., 2021). In other words, their analysis revealed that PwPMI are involved in approximately five times more calls for service than are identified through the ‘Mental Health’ call classification alone.

Coalescing the above data and research issues, along with the recent findings of Langton and colleagues (2021), it is clear that the true proportion of calls for police service involving PwPMI is hidden within police data, thereby representing the dark figure of occurrences. Indeed, as discussed earlier, not knowing the true extent to which the police and PwPMI are intertwined within day-to-day police activity can stifle informed discussions around disentangling mental health from the broader police mandate. As such, further efforts in identifying PwPMI within police calls for service data are required as they can not only contribute to generating a more accurate figure on the proportion of calls that involve PwPMI but also can shed light on which specific call classifications have a higher likelihood of encompassing a PwPMI. This is what we seek to achieve herein.

Methods

Data and Variables

To explore the extent to which PwPMI are involved within and across police calls for service, we used data obtained from the record management system (RMS) of one Canadian municipal police service. These data contain particulars on each call, such as the dispatcher's comments, event synopses, occurrence details, type of call initially determined by the dispatcher, the final call classification determined after police investigation, final Uniform Crime Report (UCR) categorization, times and locations, and other additional information acquired by police (e.g., event address, time cleared). Data were anonymized by the crime analyst at the police agency, who extracted all calls for service for the year 2019 and generated an Excel spreadsheet documenting these calls for service split by crime-related calls and social-related calls. This provided a total of 42,996 calls for service (12,910 crime-related call records and 30,086 social-related) for this one year.

Independent Variable

The independent variable used in the analyses is 'Mental Health', a binary dummy variable created to represent whether each call involved PwPMI or not (0 = did not involve PwPMI, 1 = involved PwPMI). To create this variable, we manually coded all the calls for service within Excel. This was performed by first executing an initial manual scan of all qualitative information—dispatcher comments, event synopsis, and/or occurrence details—to reveal any words within the call records representing possible PwPMI involvement. This preliminary data examining occurred to generate an extensive list of keywords to search through the qualitative data from each call for service. These keywords included, for example, ‘ADHD’, ‘OCD’, ‘Bipolar’, ‘Depressed/Depression’, ‘Suicidal/Suicidal Ideation’, and ‘Anxiety/Anxious’, among others2. Then, we manually searched for these keywords throughout all the calls for service.

After identifying calls for service through keyword search, we read the qualitative data associated with the call in full before coding to ensure PwPMI involvement instead of relying only on the keyword search to code. This process occurred as some calls had one or several keywords in the qualitative information but did not explicitly state or show that a PwPMI was involved in the call for service. Therefore, it was necessary to manually read the data prior to coding to ensure coding accuracy and that codes were not formed on the basis of assumptions about the meaning of the comments. Once it could be verified that the call for service did, in fact, involve PwPMI, the record was subsequently coded as '1' to indicate this involvement. All calls for service that did not include qualitative information were coded as 'N/A' to represent that it is unknown whether or not they involved PwPMI.

Dependent Variable

The dependent variable used in the analyses are all of the police calls for service classifications. These include: 'Assault', 'Disturbance', 'Domestic', 'Medical/Check Welfare', 'Missing Persons', 'Property Damage', 'Theft', 'Trouble with Persons', and 'Weapons'. Other call classifications documented within the data involving low cell counts (i.e., 1.0 per cent or less) were collapsed into a variable labelled 'All Other Calls'. This grouping encompasses, for example, 'Robbery' (n = 196), 'Impaired' (n = 234), 'Break and Enter' (n = 276), and 'Trespassing' (n = 296) (see Table 1 for full list).

Table 1. Descriptive Overview of all Police Calls for Service Classifications

Calls for Service Classifications

Total Calls (%)

Medical/Check Welfare

9436 (21.9)

Theft

7031 (16.4)

Assault/Battery

2136 (5.0)

Missing Person

2033 (4.7)

Disturbance

1925 (4.5)

Property Damage

1237 (2.9)

Domestic

991 (2.3)

Weapons

892 (2.1)

All Other Calls:

Section 11: Mental Health Act

2,940 (6.8)

397 (0.9)

Other Complaints/Criminal Code3

393 (0.9)

Threats

316 (0.7)

Trespassing

296 (0.7)

MVC

290 (0.7)

Break and Enter

276 (0.6)

Suspicious Person

273 (0.6)

Shoplifters

269 (0.6)

Impaired

234 (0.5)

Robbery

196 (0.5)

Total

42996 (100)

Analyses

First, basic descriptive analysis by way of cross-tabulation was employed to show how often PwPMI are involved across all police calls for service categories based on the 'Mental Health' variable. Then, logistic regression models are used to predict the extent to which PwPMI are involved in police calls for service and across the different call classifications. The use of this model was selected as, given that the dependent variable in the analysis is binary (PwPMI involvement; 0 = no and 1 = yes) and involves bounded categories, the application of a linear probability model violates the standard Ordinary Least Squares (OLS) assumptions. For police calls for service i, the full model estimating the probability of PwPMI involvement is:

log π(PwPMI involvement)i 1 π(PwPMI involvement)i \text{log\ }\frac{\pi{(PwPMI\ involvement)}_{i}\ }{1 - \ \pi{(PwPMI\ involvement)}_{i}\ } =β0+ βxi + εi= \beta_{0} + \ \beta\mathbf{x'}_{i}\ + \ \varepsilon_{i},

where βxi\beta\mathbf{x'}_{\mathbf{i}} represents the set of call classifications added, and εi\varepsilon_{i} captures residual variation. Below, we analyze two types of models: first, a model predicting PwPMI involvement across all calls for service, and then, in Model 2, we analyze PwPMI involvement across the different calls for service classifications. The purpose of this is to first provide a baseline estimate on the extent to which PwPMI are involved across all police calls for service more generally, then to show how each call classification differentially involves PwPMI.

Results

Descriptive Analysis

Table 2 presents the results of the cross-tabulation analysis showing the calls for service classifications tabulated by whether or not PwPMI involvement was suggested within the dispatcher's comments, event synopsis, and/or occurrence details. As can be seen, 10.8% or 4,646 of all calls for service involved PwPMI, whereas 87.2% or 37,488 did not. Notably, this proportion differs quite significantly from the ‘Mental Health’ call classification in our data—Section 11: Mental Health Act—which comprised only 0.9% (n = 397) all calls for service in 2019 (see Table 1). Further, there were missing values across each call classification, with 'Theft' (n = 261; 3.7%) and 'Trouble with Persons' (n = 205; 1.4%) being the categories with the highest amount of missing qualitative data. Thus, due to the absence of information, it could not be determined if PwPMI were involved in 862 calls (2%). Generally, the overall trend and majority of calls for service did not involve PwPMI. Nonetheless, the amount of calls for service involving PwPMI warrants attention, particularly in terms of how this phenomenon is dispersed across the range of calls for service classifications.

Table 2. Crosstabulation of Police Calls for Service Involving PwPMI

Calls for Service Classifications

PwPMI Involvement (%)*

Total

Calls (%)ª

Yes

No

Unknown

Trouble with Persons

959 (6.7)

13211 (91.9)

205 (1.4)

14375 (33.4)

Medical/Check Welfare

1881 (19.9)

7462 (79.1)

93 (1.0)

9436 (21.9)

Theft

61 (0.9)

6709 (95.4)

261 (3.7)

7031 (16.4)

All Other Calls

1056 (35.9)

1776 (60.4)

108 (3.7)

2940 (6.8)

Assault/Battery

101 (4.7)

2024 (94.8)

11 (0.5)

2136 (5.0)

Missing Person

328 (16.1)

1659 (81.6)

46 (2.3)

2033 (4.7)

Disturbance

47 (2.4)

1856 (96.4)

22 (1.1)

1925 (4.5)

Property Damage

92 (7.4)

1122 (90.7)

23 (1.9)

1237 (2.9)

Domestic

39 (3.9)

885 (89.3)

67 (6.8)

991 (2.3)

Weapons

82 (9.2)

784 (87.9)

26 (2.9)

892 (2.1)

Total

4646 (10.8)

37488 (87.2)

862 (2.0)

42996 (100)

* Percentages refer to row total.

ª Percentages refer to column total.

To break this down further, the call classification 'Medical/Check Welfare' had the highest number of calls for service involving PwPMI as suggested within the event synopses, dispatcher comments, and occurrence details of these calls. Notably, this call classification had 1,881 reports involving PwPMI. The next call classification with the highest number of PwPMI involvement emerged as 'All Other Calls.' This includes 'Section 1: Mental Health Act' apprehensions, 'Assist Other Agency,' and 'Suspicious Person, Vehicle, or Noise,' for example. 'All Other Calls' brought about 1,056 calls for service with PwPMI involvement. Regarding the total calls for service for this call classification, this made up 35.9% of calls classified as 'All Other Calls.' Next, 'Trouble with Persons' had 959 calls with PwPMI involvement, which made up 6.7% of all calls in this type of call for service. The other category that emerged with a higher number of calls with PwPMI involvement was 'Missing Person', which documented PwPMI involvement in 16.1% of calls for service within this call-type.

Notably, most of the calls for service classifications with less apparent criminal implications—in other words, calls for service least evidently involving crime—such as police called for checking on the wellbeing of persons, emerged as having the greatest involvement of PwPMI. In contrast, those with more explicit crime components, such as 'Theft' and 'Assault/Battery,' had less PwPMI involvement. That is, social-related calls for service—which comprised most calls for service (n = 30,086)—appeared as significant drivers of PwPMI involvement within police calls for service in contrast to crime-related calls for service.

Statistical Analysis

Table 3 presents the results of the logistic regression models predicting PwPMI involvement across the calls for service classifications. Model 1 shows that police calls for service generally are significantly less likely to involve PwPMI. Specifically, police calls for service from the police service sampled are 0.988 times less likely to involve PwPMI (p < .001). Therefore, the likelihood of police calls for service not involving PwPMI 50.3%. Despite this, Model 2—which predicts PwPMI involvement in police calls for service across the different call classifications—highlighted several call categories that emerged as either more or less likely to involve PwPMI.

The logistic regression model predicting PwPMI involvement across the different call classifications is shown in Model 2. This reveals that 'Trouble with Persons,' 'Medical/Check Welfare,' 'Missing Person,' 'Domestic,' and 'Weapon' calls for service are significantly more likely to involve PwPMI. In contrast, 'Assault/Battery' emerged as significantly less likely to involve PwPMI. To expand on the former, the odds of PwPMI involvement increased by a factor of 1.751 if the call classification was 'Trouble with Persons' compared to all other police calls for service (p < .01). ‘Missing Persons’ reports appeared as 2.734 times more likely to involve PwPMI (p < .001). The likelihood of PwPMI involvement increased by a factor of 3.492 if the call classification was ‘Medical/Check Welfare’ compared to all other calls for services (p < .001). Next, ‘Domestic’ calls for service are over one and a half times more likely to involve PwPMI (i.e., increased by a factor of 1.640) (p < .01). Finally, concerning call classifications that emerged as positively statistically significant, the likelihood of a call for service involving PwPMI increased by a factor of 1.454 if the call was classified as 'Weapons' in comparison to all other calls for service classifications. Turning to the call-type that emerged as negatively statistically significantly associated with PwPMI involvement, the odds of 'Assault/Battery' involving PwPMI decreased by a factor of 0.227 compared to all other police calls for service classifications (p < .05).

Table 3. Logistic Regression Models (Odds Ratios) Predicting PwPMI Involvement in Calls for Service and Calls for Service Classifications

Model 1

Model 2

Calls for Service

0.988***

-

(0.004)

-

Calls for Service Classifications (=All Other)

Trouble with Persons

1.751**

(0.801)

Medical/Check Welfare

3.492***

(0.148)

Theft

0.571

(0.261)

Assault/Battery

0.227*

(0.094)

Missing Person

2.734***

(0.189)

Disturbance

0.332

(0.051)

Property Damage

0.703

(0.417)

Domestic

1.640**

(0.107)

Weapons

1.454***

(0.486)

Constant

0.072

Log-likelihood

-129.478

Pseudo R-squared

0.711

Number of Observations

42134

Standard errors are in parentheses below parameter estimates.

*p <.05; **p <.01; *** p <.001 (two-tailed tests).

Weighted estimates [95% CIs].

Discussion

The recent ‘defund the police’ movement has sparked much-needed national and international conversation on police reform. As part of these conversations, reformers have called for the removal of the police from their long-standing role as primary responders to PwPMI calls. While studies estimate that approximately 1% of all calls for service involve PwPMI, we note that police data recording practices—notably, the recording of calls for service based on the primary nature of a given call—may mask the involvement of PwPMI within and across calls for service thus leading to an underestimation of the true proportion of calls that involve PwPMI. As such, the purpose of the present study was two-fold: (1) identify the proportion of calls for police service—beyond calls explicitly classified as ‘Mental Health’—that involve PwPMI; and (2) predict the extent to which PwPMI are present within and across calls for service. In order to achieve these study objectives, we—via manual review and text search of qualitative data appended to calls for service—examined all calls made to a police service in Southern Ontario, Canada for the year 2019.

We find that while the proportion of calls for service in our data under the ‘Mental Health’ call classification (n = 397, 0.9%) is consistent with the ~1% rate identified in previous research (Hodgkinson & Andresen, 2019; Koziarski, 2020; Langton et al., 2021; Livingston, 2016; Lum et al., 2021; Vaughan et al., 2018; White & Goldberg, 2018), our results reveal that PwPMI were, in fact, present in 10.8% (n = 4,646) of all calls in 2019. In other words, there are over ten times as many calls for service in our study jurisdiction that involve PwPMI than if one were to look at the ‘Mental Health’ call classification alone. This finding is in line with those of Langton et al. (2021) who found that 9.9% of calls for service in Manchester, England involved PwPMI even though the ‘Mental Health’ call classification only comprised 1.9% of all calls for service. Together, these findings show that police call classification practices based on primary nature do indeed mask a significant proportion of calls for service that involve PwPMI. We, therefore, suggest that PwPMI comprise a far more significant proportion of calls for service than the frequently cited 1% estimate and that, as such, there is a strong need to replicate this work across other jurisdictions to arrive at a more accurate understanding of the proportion of PwPMI in police calls for service. Doing so is also especially crucial for informing future discussions on police reform when it comes to responses to PwPMI. This is because if changes in responding to such calls are made with a narrow focus on the ~1% of calls for service under the ‘Mental Health’ call classification, our findings suggest that the reform effort will overlook a considerable proportion of calls that involve PwPMI to begin with.

This then begs the question of how the net of reform efforts can be widened in order to encompass the full scope of calls for police service that involve PwPMI. Indeed, while our results show that calls for service generally are less likely to involve PwPMI, certain call classifications—'Trouble with Persons,' 'Domestic,' 'Weapons,' 'Medical/Check Welfare,' and 'Missing Person' calls—exhibit higher probabilities of PwPMI involvement. Unfortunately, given this wide scope of calls for service that involve PwPMI, the solution to broadening reform efforts may not be all too straightforward nor conform to the civilian-based responses that reformers have been calling for. Take, for example, the ~1% of calls that fall under the ‘Mental Health’ call classification. Previous research has shown that such calls most often conclude with no action on behalf of the responding officer(s) because there is no legally justifiable reason to either arrest the individual for a criminal offence or apprehend them under involuntary civil commitment criteria (Charette et al., 2011; Teplin & Pruett, 1992; Watson & Wood, 2017). Calls in this so-called 'grey zone' of police-PwPMI interactions are, as reformers have called for, suitable for being dealt with by a civilian-based mental health response (Watson et al., 2021). Further, and in addition to the police largely having a peacekeeping role in such calls (Bittner, 1967), the possible alternative to replace the police as responders to such calls is clear: mental health practitioners. The remaining call classifications, on the other hand, are unquestionably a police responsibility—such as ‘Domestic’ or ‘Weapons’ calls due to the threat or possibility of bodily harm—or, as Lum and colleagues (2021) point out in their analysis of over four million calls for service, are not clearly attributable to any particular organization or social service. ‘Medical/Check Welfare’ calls as well as ‘Trouble with Persons’ calls, for example, are so-called ‘umbrella’ call classifications which commonly involve a myriad of possible issues that lead to a call for service being generated. ‘Missing Person’ calls, on the other hand, are more specific in nature, but they are also more resource-intensive and can at times require multi-jurisdictional co-operation. In the case of these call classifications, despite them having a higher probability of involving PwPMI, they cannot be addressed by mental health practitioners as they encompass far more than just mental health considerations. However, outside of the police themselves, no social service or organization currently exists with a broad enough mandate, suitable resourcing, and around-the-clock availability that can replace the police as responders to these issues (Lum et al., 2021; Koziarski & Huey, 2021).

Until such a day arrives—if at all—a possible solution to reforming responses to all calls for service involving PwPMI may not be to look for civilian actors who can replace the police in these instances, but to work towards improving how the police respond to these calls to begin with. Efforts broadly referred to as co-response teams whereby a trained officer and mental health practitioner jointly respond to calls involving PwPMI have been immensely successful in this regard. More specifically, research on co-response teams has shown that such efforts are capable of reducing unnecessary PwPMI transfers to hospital, time spent at calls, and use of force against PwPMI, and well as increasing PwPMI satisfaction when interacting with police, referrals to community-based mental health resources, and engagement with outpatient mental health services (Blais et al., 2020; Fahim et al., 2016; Kirst et al., 2015; Kisely et al., 2010; Lamanna et al., 2018; Semple et al., 2021). While co-response efforts are currently found to endure a lack of funding, under-staffing, and other practical challenges (Bailey et al., 2018; Koziarski et al., 2020), they may be a suitable reform-oriented alternative to a frontline police response particularly to calls for service that have a higher likelihood of PwPMI involvement. In such circumstances an individual with extensive mental health training is always on-scene and the most appropriate side of the co-response team can take the lead depending on the context of a given call. Co-response teams may therefore provide the opportunity for the police to reduce their footprint in a much larger proportion of in the 10.8% of calls for service involving PwPMI, while simultaneously being ready to intervene should the context of a given call require it.

Shifting away from broader police reform efforts toward more immediate implications for police practice, findings from the present study, as well as the fact that the contemporary police mandate has far less to do with crime problems than social problems (Wuschke et al., 2018), suggests that police data practices may need to be modified to more closely align with the scope of the present mandate. Indeed, as shown here, existing police data practices have failed to capture a considerable proportion of calls for service that involve PwPMI. As such, even though frontline police officers have a tumultuous relationship with the existing abundance of paperwork they are required to complete (Huey et al., 2021), standardized modifications to police data practices are needed which—at the very least—enable officers to denote whether a particular call involved PwPMI or not. This would, in turn, enable police services to conduct similar analyses to those conducted here, but with their localized context being of focus.

From these rudimentary analyses, police services could subsequently formulate more nuanced approaches to calls for service involving PwPMI that are tailored for their respective contexts. For example, call classifications which experience a higher frequency or higher likelihood of PwPMI involvement could be allotted a higher share of operational resources, such as staffing hours or training time, to better account and prepare for the multi-layered nature of these particular calls for service. Such calls could also be reserved for officers who are more specialized in mental health than their frontline colleagues, thus resulting in a more efficient and effective use of multi-skilled officers than would be had by dispatching them to other calls where their skills cannot be used to the fullest extent.

Conclusion

The present study’s findings reveal that the police role of ‘street corner psychiatrist’ is more prominent than initially thought, especially in and across particular calls for service classifications. These results are especially important in informing discussions and debates around removing the police as primary responders to PwPMI. That is, by simply removing the police from the ~1% of all calls for service under the ‘Mental Health’ call classification, we only remove the police from a small proportion of all calls that do indeed involve PwPMI. However, untangling the police from other call classifications that also involve PwPMI may not be straightforward due to a lack of alternatives beyond the police that have a broad operational mandate, adequate resourcing, and around-the-clock availability. For these calls for service then, reform efforts should not focus on who can replace the police but how the police can improve their responses to calls that may involve PwPMI. For the short-term, however, police data recording practices are in-need of modification to more accurately record PwPMI involvement with and across calls for service. Improving recording practices on this front could enable the development of more nuanced and localized approaches to calls involving PwPMI.

References

Bailey, K., Paquet, S. R., Ray, B. R., Grommon, E., Lowder, E. M., & Sightes, E. (2018). Barriers and Facilitators to Implementing an Urban Co-Responding Police-Mental Health Team. Health & Justice, 6(1), 21. doi:10.1186/s40352-018-0079-0

Bittner, E. (1967). Police Discretion in Emergency Apprehension of Mentally Ill Persons. Social Problems, 14(3), 278–292. doi:10.1525/sp.1967.14.3.03a00040

Blais, E., Landry, M., Elazhary, N., Carrier, S., & Savard, A.-M. (2020). Assessing the Capability of a Co-Responding Police-Mental Health Program to Connect Emotionally Disturbed People with Community Resources and Decrease Police Use-of-Force. Journal of Experimental Criminology, 1–25. doi:10.1007/s11292-020-09434-x

Bohrman, C., Wilson, A. B., Watson, A., & Draine, J. (2018). How Police Officers Assess for Mental Illnesses. Victims & Offenders, 13(8), 1077–1092. doi:10.1080/15564886.2018.1504844

Charette, Y., Crocker, A. G., & Billette, I. (2011). The Judicious Judicial Dispositions Juggle: Characteristics of Police Interventions Involving People with a Mental Illness. The Canadian Journal of Psychiatry, 56(11), 677–685. doi:10.1177/070674371105601106

Crocker, A. G., Hartford, K., & Heslop, L. (2009). Gender Differences in Police Encounters Among Persons With and Without Serious Mental Illness. Psychiatric Services, 60(1), 86–93. doi:10.1176/ps.2009.60.1.86

Fahim, C., Semovski, V., & Younger, J. (2016). The Hamilton Mobile Crisis Rapid Response Team: A First-Responder Mental Health Service. Psychiatric Services, 67(8), 929–929. doi:10.1176/appi.ps.670802

Frederick, T., O’Connor, C., & Koziarski, J. (2018). Police Interactions with People Perceived to Have a Mental Health Problem: A Critical Review of Frames, Terminology, and Definitions. Victims & Offenders, 13(8), 1037–1054. doi:10.1080/15564886.2018.1512024

Hartford, K., Heslop, L., Stitt, L., & Hoch, J. S. (2005). Design of an Algorithm to Identify Persons with Mental Illness in a Police Administrative Database. International Journal of Law and Psychiatry, 28(1), 1–11. doi:10.1016/j.ijlp.2004.12.001

Hoch, J. S., Hartford, K., Heslop, L., & Stitt, L. (2009). Mental Illness and Police Interactions in a Mid-Sized Canadian City: What the Data Do and Do Not Say. Canadian Journal of Community Mental Health, 28(1), 49–66. doi:10.7870/cjcmh-2009-0005

Hodgkinson, T., & Andresen, M. A. (2019). Understanding the Spatial Patterns of Police Activity and Mental Health in a Canadian City. Journal of Contemporary Criminal Justice, 35(2), 221–240. doi:10.1177/1043986219842014

Huey, L., Ferguson, L., & Koziarski, J. (2021). The Irrationalities of Rationality in Police Data Processes. Policing and Society, 1–16. doi:10.1080/10439463.2021.2007245

Kirst, M., Pridham, K. F., Narrandes, R., Matheson, F., Young, L., Niedra, K., & Stergiopoulos, V. (2015). Examining Implementation of Mobile, Police-Mental Health Crisis Intervention Teams in a Large Urban Center. Journal of Mental Health, 24(6), 369–374. doi:10.3109/09638237.2015.1036970

Kisely, S., Campbell, L. A., Peddle, S., Hare, S., Pyche, M., Spicer, D., & Moore, B. (2010). A Controlled Before-and-after Evaluation of a Mobile Crisis Partnership between Mental Health and Police Services in Nova Scotia. The Canadian Journal of Psychiatry, 55(10), 662–668. doi:10.1177/070674371005501005

Koegel, P., Burnam, M. A., & Farr, R. K. (1988). The Prevalence of Specific Psychiatric Disorders Among Homeless Individuals in the Inner City of Los Angeles. Archives of General Psychiatry, 45(12), 1085–1092. doi:10.1001/archpsyc.1988.01800360033005

Kouyoumdjian, F. G., Wang, R., Mejia-Lancheros, C., Owusu-Bempah, A., Nisenbaum, R., O’Campo, P., Stergiopoulos, V., & Hwang, S. W. (2019). Interactions between Police and Persons Who Experience Homelessness and Mental Illness in Toronto, Canada: Findings from a Prospective Study. The Canadian Journal of Psychiatry, 64(10), 718–725. doi:10.1177/0706743719861386

Koziarski, J. (2020). Examining the Spatial Concentration of Mental Health Calls for Police Service in a Small City. Policing: A Journal of Policy and Practice. doi:10.1093/police/paaa093

Koziarski, J., & Huey, L. (2021). #Defund or #Re-Fund? Re-Examining Bayley’s Blueprint for Police Reform. International Journal of Comparative and Applied Criminal Justice, 1–16. doi:10.1080/01924036.2021.1907604

Koziarski, J., O’Connor, C., & Frederick, T. (2020). Policing Mental Health: The Composition and Perceived Challenges of Co-Response Teams and Crisis Intervention Teams in the Canadian Context. Police Practice and Research, 22(1), 1–19. doi:10.1080/15614263.2020.1786689

Lamanna, D., Shapiro, G. K., Kirst, M., Matheson, F. I., Nakhost, A., & Stergiopoulos, V. (2018). Co‐responding Police–Mental Health Programmes: Service User Experiences and Outcomes in a Large Urban Centre. International Journal of Mental Health Nursing, 27(2), 891–900. doi:10.1111/inm.12384

Lamb, H. R., & Weinberger, L. E. (1998). Persons With Severe Mental Illness in Jails and Prisons: A Review. Psychiatric Services, 49(4), 483–492. doi:10.1176/ps.49.4.483

Langton, S., Bannister, J., Ellison, M., Haleem, M. S., & Krzemieniewska-Nandwani, K. (2021). Policing and Mental Ill-Health: Using Big Data to Assess the Scale and Severity of, and the Frontline Resources Committed to, Mental Ill-Health-Related Calls-for-Service. Policing: A Journal of Policy and Practice. doi:10.1093/police/paab035

Livingston, J. D. (2016). Contact Between Police and People With Mental Disorders: A Review of Rates. Psychiatric Services, 67(8), 850–857. doi:10.1176/appi.ps.201500312

Livingston, J. D., Desmarais, S. L., Verdun-Jones, S., Parent, R., Michalak, E., & Brink, J. (2014). Perceptions and Experiences of People with Mental Illness Regarding Their Interactions with Police. International Journal of Law and Psychiatry, 37(4), 334–340. doi:10.1016/j.ijlp.2014.02.003

Lum, C., Koper, C. S., & Wu, X. (2021). Can We Really Defund the Police? A Nine-Agency Study of Police Response to Calls for Service. Police Quarterly, 109861112110350. doi:10.1177/10986111211035002

Markowitz, F. E. (2011). Mental Illness, Crime, and Violence: Risk, Context, and Social Control. Aggression and Violent Behavior, 16(1), 36–44. doi:10.1016/j.avb.2010.10.003

Ratcliffe, J. H. (2021). Policing and Public Health Calls for Service in Philadelphia. Crime Science, 10(1), 5. doi:10.1186/s40163-021-00141-0

Standing Committee on Public Safety and National Security (2014). Economics of Policing: Report of the Standing Committee on Public Safety and National Security.

Semple, T., Tomlin, M., Bennell, C., & Jenkins, B. (2021). An Evaluation of a Community-Based Mobile Crisis Intervention Team in a Small Canadian Police Service. Community Mental Health Journal, 57(3), 567–578. doi:10.1007/s10597-020-00683-8

Shore, K., & Lavoie, J. A. A. (2018). Exploring Mental Health-Related Calls for Police Service: A Canadian Study of Police Officers as ‘Frontline Mental Health Workers.’ Policing: A Journal of Policy and Practice, 13(2), 157–171. doi:10.1093/police/pay017

Simpson, R. (2020). Calling the Police: Dispatchers as Important Interpreters and Manufacturers of Calls for Service Data. Policing: A Journal of Policy and Practice. doi:10.1093/police/paaa040

Teplin, L. A., & Pruett, N. S. (1992). Police as Streetcorner Psychiatrist: Managing the Mentally Ill. International Journal of Law and Psychiatry, 15(2), 139–156. doi:10.1016/0160-2527(92)90010-x

Vaughan, A. D., & Andresen, M. A. (2018). The Cost of Mental Health‑Related Calls on Police Service Evidence from British Columbia. In R. J. Mitchell & L. Huey, Evidence-Based Policing: An Introduction (pp. 173–185). Policy Press.

Vaughan, A. D., Hewitt, A. N., Andresen, M. A., & Brantingham, P. L. (2015). Exploring the Role of the Environmental Context in the Spatial Distribution of Calls-for-Service Associated with Emotionally Disturbed Persons. Policing, 10(2), 121–133. doi:10.1093/police/pav040

Vaughan, A. D., Ly, M., Andresen, M. A., Wuschke, K., Hodgkinson, T., & Campbell, A. (2018). Concentrations and Specialization of Mental Health–Related Calls for Police Service. Victims & Offenders, 13(8), 1153–1170. doi:10.1080/15564886.2018.1512539

Watson, A. C., Pope, L. G., & Compton, M. T. (2021). Police Reform From the Perspective of Mental Health Services and Professionals: Our Role in Social Change. Psychiatric Services, appi.ps.2020005. doi:10.1176/appi.ps.202000572

Watson, A. C., & Wood, J. D. (2017). Everyday Police Work during Mental Health Encounters: A Study of Call Resolutions in Chicago and Their Implications for Diversion. Behavioral Sciences & the Law, 35(5–6), 442–455. doi:10.1002/bsl.2324

White, C., & Goldberg, V. (2018). Hot Spots of Mental Health Crises: A Look at the Concentration of Mental Health Calls and Future Directions for Policing. Policing: An International Journal, 41(3), 401–414. doi:10.1108/pijpsm-12-2017-0155

Wuschke, K. E., Andresen, M. A., Brantingham, P. J., Rattenbury, C., & Richards, A. (2017). What Do Police Do and Where Do They Do It? International Journal of Police Science & Management, 20(1), 19–27. doi:10.1177/1461355717748973

Comments
0
comment
No comments here
Why not start the discussion?