Deflecting suicide and mental health calls from law enforcement with an embedded crisis line specialist at a 911 center: A synthetic control analysis

Deflecting suicide and mental health calls from law enforcement with an embedded crisis line specialist at a 911 center: A synthetic control analysis | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Deflecting suicide and mental health calls from law enforcement with an embedded crisis line specialist at a 911 center: A synthetic control analysis Leonard Swanson, Aaron Gottlieb, Margaret Thomas This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9226560/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Law enforcement are the default responders to mental health- and suicide-related calls, and law enforcement can escalate mental health crises to hospitalizations, uses of force, and deaths. Mental health crisis lines, such as 988, are an alternative underutilized mental health crisis resource compared to estimated annual mental health-related 911 calls. This study assesses whether embedding a counselor at a 911 center could reduce mental health- and suicide-related police responses by transferring 911 calls to a mental health crisis line. A quasi-experimental synthetic control analysis assessed police response rates after embedding a crisis line counselor at a 911 emergency communications center in Mesa, AZ in 2021. Authors used publicly available 911 data from 27 cities to create donor pools for synthetic control analyses of police responses to suicide and mental health calls. Embedding a 911 crisis line counselor was associated with 22.27–47.96 fewer police responses to suicide calls per quarter per 100,000 people, relative to a synthetic control unit. The lower suicide call rate translated to an estimated 228.1–490.6 police hours reduced per quarter in Mesa. There was no evidence to suggest that Mesa’s police response rate to mental health calls decreased. The embedded presence of a crisis line counselor at a 911 center may facilitate interdisciplinary trust that enables call routing to crisis lines. Emergency communication centers should consider embedding crisis line staff to divert calls from police, limiting the expansion of the police footprint. crisis line 988 911 embedded diversion suicide crisis response Figures Figure 1 Figure 2 Introduction Law enforcement officers are the predominant responders to suicide and mental health crises (M. E. Balfour et al., 2022 ; Borum et al., 1997 ; Gasior, 2020; Koziarski et al., 2022 ). Officers can struggle with suicide and mental health crisis responses (Compton et al., 2014 ), as their presence can incite fears that escalate to arrests and use of force, or hospitalization (Cojean et al., 2020 ; Fuller et al., 2015 ; Morabito et al., 2017 ). Law enforcement trainings to increase officer awareness of mental health issues, self-efficacy, and de-escalation skills have proliferated (Watson et al., 2017 ). Trainings, such as the 40-hour training in the Crisis Intervention Team model, can increase officer referrals to mental health crisis resources (Comartin et al., 2019 ), but system discontinuities prevent transfers from traditional 911 pathways to the mental health crisis response continuum (M. E. Balfour et al., 2022 ). Law enforcement predominately transport people in mental health crises to emergency departments (Bakko et al., 2025 ), which can be time consuming, frustrating and ineffective (Harada et al., 2021 ; Hickey et al., 2020 ). Emergency departments can be traumatic: mental health crisis patients sometimes wait weeks for placement at psychiatric hospitals in a process often referred to as ‘ED-boarding’ (K. Nordstrom et al., 2019 ). After immediate care, emergency departments struggle with connecting people to post-crisis mental health treatment (Currier et al., 2010 ; Vakkalanka et al., 2021 ). The traditional public mental health crisis response system needs reform. The Black Lives Matter protests in 2020 spurred momentum for alternative response models that decrease the role and presence of law enforcement in mental health crises. Some mental health crisis models can handle calls that would otherwise be handled by police, decreasing the total incidents that could escalate to law enforcement use of force or arrest (M. E. Balfour et al., 2022 ; Compton et al., 2024 ). In-person community response teams are one example of a model that can address 911 calls involving social and mental health needs, suggesting that 911 mental health crisis calls can indeed be safely resolved by civilians (Davis et al., 2025 ; Dee & Pyne, 2022 ; Waters, 2021 ). Crises resolved upstream (e.g. mental health crisis lines, mobile crisis teams) could prevent the utilization of downstream acute crisis services (e.g. emergency departments, psychiatric hospitalization, arrests) that can be traumatic, expensive, and often in high demand (Munetz & Griffin, 2006 ; Swanson et al., 2022 ; Ward-Ciesielski & Rizvi, 2021 ). Mental health systems are developing systemic frameworks to standardize and expand crisis response services as alternatives to 911 default law enforcement pathways. The Substance Abuse and Mental Health Services Administration describes crisis system components as a three-part framework: “someone to contact” (e.g., call centers), “someone to respond” (e.g., mobile crisis teams), and “a safe place for help” (e.g., stabilization facilities) (SAMHSA, 2025). In July, 2022, the nationwide three-digit 988 crisis hotline launched as a toll-free number for around-the-clock suicidal crisis counseling, e.g., “someone to contact” ( 988 Lifeline , 2022; M. Balfour & Minkoff, 2022 ). The 988 hotline allowed people to reach the National Suicide Prevention Lifeline (NSPL) with an easy-to-remember three-digit number. When someone contacts a mental health crisis line, call takers answer, listen, provide support, empathize, and connect the caller with local resources. Crisis line callers describe 988 as helpful, even life-saving (Gould et al., 2025 ). When NSPLs make follow-up calls to callers, clients report that the intervention prevents them from killing themselves and keeps them safe (Gould et al., 2018 ). Rather than a single call center, there are 200 distinct 988 call centers spread across the country; the decentralization is meant to encourage call center familiarity with and referral to local resources ( 988 Lifeline , 2022). Crisis line staff determine whether a call presents an imminent risk, which is when a person expresses both immediate intent and capability of suicide (Gould et al., 2016 ). Crisis line calls that present imminent risk can be routed to emergency services (e.g. law enforcement, emergency medical services). Less than 2% of 988 calls require emergency assistance and are routed to 911 emergency services (SAMHSA, 2023). About one in four crisis line calls involving imminent risk are routed to emergency services without collaboration with the caller (Gould et al., 2016 ). Goldman and colleagues ( 2025 ) assessed the characteristics of calls routed from 988 to 911: people who are Black, male, or at imminent risk of harm were more likely to be rerouted from 988 to 911 for active rescue. However, only a third of 988 callers with serious distress indicated they would call 988 again (Purtle et al., 2023 ). Some mental health crisis lines have the capability of dispatching a mobile crisis team, offering an alternative in-person response to law enforcement. There is a thin but growing body of research on the mobile crisis model. Behavioral health practitioners have been responding to mental health crises in community settings since the 1960s (Center for Police Research and Policy, 2021 ). Attending to crises outside of the hospital or healthcare setting is the key distinction that separates mobile crisis from other crisis services provided by behavioral health service agencies (Alexander & Zealberg, 1999 ; Geller et al., 1995 ; Gillig et al., 1990 ; Guo et al., 2001 ). Mobile crisis teams generally attempt to de-escalate crises, screen clients to determine eligibility for psychiatric hospitalization, provide safety plans, and link people to post-crisis mental health care (Center for Police Research and Policy, 2021 ; Dyches et al., 2002 ; Watson et al., 2019 ). Compared to emergency department-based crisis services, mobile crisis teams are effective in diverting people from hospitalizations (Guo et al., 2001 ; Hugo et al., 2002 ; Vakkalanka et al., 2021 ) and connecting people to mental health services after a crisis incident (Currier et al., 2010 ; Dyches et al., 2002 ; Kim & Kim, 2017 ). Compared to law enforcement responses, mobile crisis teams may also prevent future arrests (Swanson et al., 2025 ). Despite crisis lines’ ability to de-escalate mental health crises and connect people to mental health services, there are reasons to suspect that people call 911 for mental health crises instead. The most utilized public resource is 911 (Transform 911, 2022 ); an estimated 240 million 911 calls occur every day (National Emergency Number Association, 2021 ). Several 911 calls do not fit the original scope of law enforcement response (Transform 911, 2022 ). A recent historical analysis discusses how the 911 call for service system has shaped and expanded the role of law enforcement to attend to caller needs beyond those relating to crime (Gillooly & Thacher, 2024 ). Mental health-related 911 calls are a prime example of the expanded police footprint, as its utilization as a public resource has grown to include specialized emergencies beyond the capabilities of the responders (Gillooly & Thacher, 2024 ). There are no uniform metrics to determine the proportion of 911 calls that are mental health-related, let alone those that flag suitability to route to mental health crisis lines. Some analyses estimate 20% of 911 calls are mental health-related (Center for American Progress, 2020). However, even a conservative estimate of 10% of the annual U.S. 911 calls (24 million) would far outnumber the annual calls to 988 (6 million) (Saunders, 2025). It is possible that several mental health-related 911 calls are escalated with active weapons or threats where law enforcement presence is necessary and would not appropriate for a mental health crisis line. People who call 911 may provoke police action by presenting situations that involve weapons, barricades, or otherwise aggressive behavior that warrants in-person police responses (Patton & Fremouw, 2016 ). However, it is unlikely that all of the estimated 24 million mental health-related 911 calls are escalated to levels where police response is necessary, and several may be suitable to deflect to mental health service responses. One reason behind the utilization differences between 911 and mental health crisis lines is the lack of awareness of mental health crisis lines. A recent survey suggests that only about one in five Americans were aware of 988, whereas over half of respondents said they would call 911 (Ueda et al., 2026 ). After receiving information about 988, respondents expressed a willingness to use it (Ueda et al., 2026 ). Other reasons for lack of 988 utilization include law enforcement involvement, forced hospitalization, and service fees (Ueda et al., 2026 ), though these reasons would not be applicable to 911 callers, since people who call 911 often expect police responses. It is likely that people call 911 for mental health crises because they do not know who else to call, and such callers would benefit from an intervention that transfers 911 calls to mental health crisis resources. To our knowledge, interventions that route 911 calls to mental health crisis lines have not been empirically tested. Over-the-phone crisis resolutions are upstream interventions that could deflect calls from legal entanglement (Beck et al., 2020; Munetz & Griffin, 2006 ). Emergency Communications Centers (ECCs), often also referred to as public safety answering points, could be key to diverting calls from law enforcement (Gillooly & Thacher, 2024 ), but most ECCs do not have standard protocols for routing 911 calls to crisis resources (Pew, 2021). There are an estimated 240 million 911 calls answered by 10,000 ECCs in the U.S. annually (Federal Communications Commission, 2024; NENA, 2021 ), each with their own governance and routing protocols (Choi et al., 2024 ). 911 professionals often default to law enforcement responses, expressing concern about legal liability and lack of trust with engaging alternative crisis resources (Hedden-Clayton et al., 2025 ; Pew, 2021). Some communities deescalate 911 crisis calls over the phone, though their approaches vary (Beck et al., 2020). Mental health crisis service models based at ECCs include: training 911 professionals about mental health calls ( CIT International - CIT Support Training for 911 , n.d.), embedding crisis professionals in 911 centers to answer calls directly (Aldridge, 2025), advising transfer of 911 calls to other crisis resources (Beck et al., 2020), or a combination of these strategies. This study examines whether embedding a crisis line counselor in Mesa, Arizona’s 911 ECC diverted law enforcement from mental health and suicidal crises by transferring calls to a local mental health crisis line. A synthetic control analysis compares Mesa’s trends of police response rates to mental health and suicide calls after the embedded crisis line counselor intervention. Research questions were: when compared to synthetic control groups, did embedding a crisis line counselor at a 911 ECC decrease police responses to A) suicide calls or B) mental health calls? We hypothesized that embedding a crisis line counselor in an ECC would significantly reduce police responses to both call types. Materials and Methods Study Setting The study setting is Mesa, Arizona, a metropolitan city in the United States with 510,680 people, or 3,635 people per square mile. Mesa’s median household income is $73,766, and 10.5% of households live under the federal poverty line (U.S. Census Bureau, 2023). Mesa is 61% white, has 27% people of two or more races, 91% of adults have graduated high school, and 31.5% have a Bachelor’s degree or higher according to the U.S. Census (U.S. Census Bureau, 2023). Areas with higher proportions of non-white people, private health insurance, and lower levels of education have shown lower 911 suicidal call rates (O’Connell et al., 2022). The Mesa Police Department (MPD) operates the ECC, receiving around 1.4 million calls per year, and assumes responsibility for dispatching police, fire, and medical calls. Mesa dispatches police to ~245,000 911 calls per year (48 calls per 100 people per year) (City of Mesa, 2024a). Solari is a crisis line that takes 988 calls for the state of Arizona. Solari launched in 2007 as a 24/7 crisis line and has since expanded to serious mental illness determinations, peer warmlines, and 211 resource navigation responses. Solari GPS-tracks and dispatches mobile crisis teams in central Arizona, receiving over 35,000 Arizona crisis calls per month (Solari, 2025). Call diversion programs in Mesa Solari and the Mesa Police Department worked together to identify mental health and suicide 911 calls that could be routed to a Solari’s crisis line. First, the two agencies targeted suicide calls. The Mesa Police Department drafted a policy that permitted routing of suicidal threats without plans or means, effective on October 16, 2019. In the month before the policy was enacted, the 911 professionals at the Mesa Police department received one hour of training that described the updated policy language, what the Solari crisis line did, how to identify suicidal and mental health behavior appropriate for call routing, how to categorize the calls, and how to transfer the calls to Solari. The training was jointly facilitated by Solari and the Mesa Police Department. 911 professionals would identify and code the calls before recommending the option to the caller, who would consent before their call would be transferred (Mesa Police Department, 2019). On August 17, 2020, the Mesa Police Department enacted a similar policy and training that enabled routing of mental health-related calls. The logistics of routing mental health calls resembled those of the suicide call policy change: for a mental health call to be eligible for transfer, there could not be apparent danger to self or others, and the caller had to consent to the transfer (Mesa Police Department, 2020). On June 5, 2021, Solari embedded a licensed professional counselor at the ECC to help with the identification of additional transferable calls. The crisis line counselor learned the ECC procedures, norms, and coding practices. The crisis line specialist did not assume call taking responsibilities but rather advised in routing and dispatch decisions made by the 911 professionals. Specifically, he listened for the types and severity of calls that the Solari crisis line would be able to handle. The crisis line counselor was involved in coaching and direct training of the 911 professionals during shifts. Making decisions on specific calls in a real-world environment gave clearer, concrete examples of appropriate calls for routing to the mental health crisis line. Data and Measures This study seeks to quantify the impact of the embedded counselor on the frequency of law enforcement responses to suicide and mental health calls using publicly available call for service data. The Mesa, AZ data came from the city’s open data portal (City of Mesa, 2024b). The data contained all calls for service dispatched to MPD officers through its Computer Aided Dispatch system. The data included a unique event number, call date, time, duration, priority level, case type, and call source. Officer-initiated call sources (responses to something an officer saw in the field) were filtered out, as they could not have been impacted by the embedded crisis line counselor. Mesa’s data were compared to a synthetic control unit, which drew from several donor sites that were reasonably comparable to the test site (Abadie et al., 2015). Researchers searched for donor site data that consisted of police responses to mental health incidents initiated by a 911 call, had publicly available 911 data, and contained at least one call type relating to suicide or mental health. We identified donor sites by searching public websites from every city in the United States whose population was greater than 100,000 people (n=369) to mimic Mesa’s urban environment. We performed Google searches using “[city name] open data” and “[city name] calls for service” from October 2024 – March 2025. We used “Computer Aided Dispatch”, “call for service”, or “police incident” data; “police stop” or “crime” data were excluded, since they did not contain 911-initiated mental health calls. Donor sites also could not have implemented a 911 embedded counselor program during the study period. To check for treatment contamination, we asked ChatGPT-5 to search for press releases that described embedded clinician programs at 911 centers at each of the cities; we verified its findings with manual Google searches. Table 1 lists donor sites (n=26), their states, populations, and rates of 911 calls for suicide, mental health, and all calls per quarter per 100,000 people. [insert table 1 around here] We counted Mesa’s “Suicide” and “Behavioral Health” initial call types as the outcome variables[1]. The trends of Mesa’s “Suicide” calls were compared against a synthetic control group of similar call types (e.g. suicidal, suicide attempt, etc.). Mesa’s “Behavioral Health” calls were compared with a synthetic control group of similar call types (e.g. mental patient, mentally disturbed person, etc.)[2]. Among the 26 donor sites, 20 included suicide-related categories, and 17 included mental health-related categories. Calls that were resolved over the phone were excluded in all but one site that did not have relevant data filters (Charlotte). The number of law enforcement calls for service for each type were aggregated by quarter per site to capture seasonal variation. It is recommended that there is low volatility in the outcome variable when running a synthetic control analysis (Wiltshire, 2025), and quarterly rates were less volatile than monthly rates. The suicide call rate in a given city and quarter was captured using this equation: SuicideCallRate qct = ((SuicideCalls qct ) / (Pop yct ))*(100,000), with SuicideCalls qct representing the number of calls dispatched to law enforcement in a given quarter and Pop yct indicating the size of a city’s population in each year from US census data (United States Census Bureau, 2024). The suicide call rate equaled the total number of suicide calls in each city that were dispatched to law enforcement per 100,000 people each quarter. The same formula was applied to mental health calls: MentalHealthCallRate qct = ((MentalHealthCalls qct ) / (Pop yct ))*(100,000). Analytic Strategy To examine the effects of the embedded crisis line counselor, we treated Mesa as a quantitative case study and employed quasi-experimental synthetic controls (Abadie et al., 2015; Bartos et al., 2020; Bartos & Kubrin, 2018; Gottlieb, Epperson, et al., 2024; Gottlieb, Harper, et al., 2024). The synthetic control method extends the difference-in-difference framework, which assumes trends across treatment and control groups continue in parallel if not for the intervention. It is often difficult to find a reasonable comparison site with similar pre-period trends (Bartos & Kubrin, 2018). The synthetic control method draws weights from a pool of potential donor sites to mimic the treatment group’s pre-period trends (time-invariant, non-negative, and add to one). We used all prior values of the outcome variable as covariates to optimize pretreatment fit between Mesa and its synthetic control (Bartos et al., 2020; Gottlieb, Epperson, et al., 2024), as prior outcome values would be more predictive of future values than any other covariate (Ferman et al., 2020). The pretreatment periods should begin as early as possible while retaining a sufficient number of donor sites. There is no established minimum of donor sites for a synthetic control analysis, though having too few donor sites risks a type 2 error. For the suicide call rate analysis, 20 donor sites had pretreatment data starting in 2018q1. For the mental health call rate analysis, 17 donor sites had pretreatment data starting in 2018q3, when Mesa began to track mental health-related calls. Both pretreatment periods ended after 2021q2, the last quarter that was majority unaffected by the embedded counselor. It is often difficult to isolate the effect of one intervention from those of other policy interventions at the study site. The initial crisis line transfer policies in 2019q4 and 2020q3 occurred before the embedded counselor intervention in 2021q2, and their effects would likely have made the most impact before the post-treatment period began in 2021q3. Any outcome values affected by the initial crisis line transfer policies before embedding the crisis line counselor would have been included in the synthetic control weighting. It is possible there were lagged effects of the initial crisis line transfer policies that contaminated the outcome values after the crisis line counselor was embedded. However, a lagged year-and-a-half impact of the initial crisis line transfer policies would be less plausibly responsible for a drop in police responses than the new impact of embedding a crisis line counselor in the 911 center. The posttreatment started in 2021q3 and continued through 2023q2 to capture at least two years of post-intervention data; extending beyond two years of post-intervention data would invite concerns that other confounding influences may affect the police response rates. The post-intervention trends of the treatment group were compared against those of the synthetic controls. We used Stata 19.5’s synth command to employ the synthetic control method (Abadie et al., 2015). We first described the pre-intervention goodness of fit using the root mean squared prediction error (RMSPE) term and compared the post- and pre-intervention RMSPEs. If the post-intervention RMSPE was substantially larger, the comparison suggested that the effects of Mesa’s embedded counselor was substantively important (Bartos & Kubrin, 2018). To assess magnitude, we compared the call rate differences between Mesa and ‘synthetic Mesa’ in the pre-intervention period with the differences the post-intervention period. Determining statistical significance when using the synthetic control method is not like traditional regression models (Abadie et al., 2015). To assess whether the observed effects happened by chance, placebo in-space estimates (i.e. creating new synthetic control groups for each of the donor sites) test whether the effect in Mesa was larger than almost all of the other jurisdictions (Abadie et al., 2015; Bartos & Kubrin, 2018; Galiani & Quistorff, 2017). If it was, it is unlikely that the observed effects were due to chance (Abadie et al., 2015; Galiani & Quistorff, 2017). We used Stata 18’s synth-runner command, which provided a standardized p-value for each post-intervention quarter (Galiani & Quistorff, 2017). To estimate the law enforcement hours saved, we calculated the average duration of on-scene time for suicide and mental health calls in Mesa and multiplied it by two officers. We then multiplied the officer hours by the synthetic control coefficients and adjusted the rate per 100,000 to Mesa’s population average=510,680. [1] In late 2020, Mesa showed an unusual change in trends among its mental health and suicide 911 call data. Mental health-related calls peaked for a five-month period, and suicide-related calls dipped in the same five-month period by similar proportions. The change occurred directly after the August, 2020 MPD policy change that expanded the categorization of mental health calls, which may have caused confusion among 911 call takers. To adjust for the suspected data issue, we assumed that the excess mental health call counts were suicide call counts and imputed accordingly. Call trends of the raw data and the authors’ imputation strategy are discussed in the supplemental appendix, as are robustness checks that remove the months in question and analyze the raw counts. [2] The term ‘behavioral health’ encompasses issues of both mental health and substance use. Substance use-related 911 calls are often medical emergencies that are often not appropriate for routing to crisis lines. Mesa’s data used the term “behavioral health”. The term ‘mental health’ was more common among the comparison ECC sites and was chosen here for conceptual relevance. Results Suicide call rates First, we report the synthetic control analysis on police responses to suicide calls. The weights that comprised the synthetic control unit were: Everett=.213, New Orleans=.037, Norman=.036, Seattle=.334, Tacoma=.38, where ‘synthetic Mesa’ was drawn from 21.3% of Everett’s suicide call rates, 3.7% of New Orleans’ suicide call rates, etc. Figure 1 shows the suicide call rate trends in Mesa (solid black line) and synthetic Mesa (dashed black line) pre- and post-intervention. Pre-intervention, the absolute value difference between Mesa and synthetic Mesa remained within 18 calls per 100,000 people. We treat post-treatment differences between Mesa and synthetic Mesa as the effect of embedding a counselor at the ECC. If the differences between Mesa and synthetic Mesa were larger than those of nearly all the donor cities, it is unlikely that the differences were due to chance. [insert figure 1 around here] Figure 1 shows Mesa’s suicide call rate was visibly lower than synthetic Mesa after the embedded counselor intervention. Mesa’s call rate increased relative to synthetic Mesa in 2021q3 by 13.43 calls, then decreased in 2021q4 to -22.27 calls, decreased in 2022q1 to -35.29 calls, decreased in 2022q2 to -36.58 calls, decreased in 2022q3 to -43.07 calls, decreased in 2022q4 to -47.96 calls, increased in 2023q1 to -44.75 calls, and increased in 2023q2 to -35.07 calls. Table 2 quantifies the synthetic control analysis. Mesa’s difference was greater than all but one of the placebo in-space estimates in 2021q4 (p=.05), all but three placebo estimates in 2022q1 (p=.15) all but two of the placebo estimates in 2022q2 (p=.10), all but one placebo estimate in 2022q3 (p=.05), all placebo estimates in 2022q4 (p<.01), all but two placebo estimates in 2023q1 (p=.10) and all but four placebo estimates in 2023q2 (p=.2). [insert table 2 around here] Mental health call rates Next, we report the synthetic control analysis on police responses to mental health calls. Table 3 shows the composition of the synthetic control unit from cities with mental health call rates most like Mesa in the pre-treatment period. The weights were: Chandler=.178, Chula Vista=.778, New York=.001, Portland=.043. Pretreatment fit was poor. As such, we are less confident in speculating whether post-treatment differences are a result of the embedded crisis line counselor at 911. Figure 2 shows the mental health call rate trends in Mesa (solid black line) and synthetic Mesa (dashed black line) pre- and post-intervention. [insert figure 2 around here] Figure 2 shows Mesa’s mental health call rate fluctuated above and below synthetic Mesa’s before and after the embedded counselor intervention. Given the poor pretreatment fit, and the continued fluctuation after the intervention, we cannot say whether the embedded counselor had any effect on the mental health call rate. Table 3 quantifies the results of the synthetic control analysis. [insert table 3 around here] Before the embedded crisis counselor at 911, the mental health call rate difference between Mesa and synthetic Mesa ranged between -6.61 to 8.87 calls per 100,000 people. After the intervention, Mesa’s mental health call rate fluctuated between -9.09 calls lower to 4.84 calls higher than synthetic Mesa’s. The post-intervention differences between Mesa and synthetic Mesa remained firmly within the normal distribution of placebo estimates; the standardized p-values never dropped below p=.59. Because of the poor pretreatment fit, we are not confident that the synthetic control analysis represents a meaningful estimate of the causal association. Because there were no obvious changes in the Mesa’s mental health call rate’s trajectory after the embedded counselor intervention, we doubt there was substantial impact. Police hours saved We translated the synthetic control results from the suicide rate analysis into approximations of police hours saved. MPD averaged 60.09 minutes of on-scene time for suicide calls from 2021-2024. After embedding a crisis line counselor at its ECC, Mesa’s 22.27–47.96 lower call rate per 100,000 people translated to 228.1–490.6 police hours saved per quarter. We did not estimate police hours saved for mental health call rates because they did not appear to significantly decrease. Discussion This study assessed the impact of embedding a crisis line counselor at a 911 center on rates of law enforcement responses to suicide and mental health calls in Mesa, AZ using a synthetic control approach. To our knowledge, this is the first study empirically testing a 911 to crisis line call routing intervention. After embedding a crisis line counselor at 911, Mesa’s suicide call rate was lower than synthetic Mesa’s by 22.27–47.96 calls per quarter per 100,000 people. Placebo estimates suggested Mesa’s difference was greater than all placebo sites in one post-intervention quarter (p<.01), all but one placebo site in two post-intervention quarters (p=.05), and all but two placebo sites in two post-intervention quarters (p=.10). We did not find evidence that the mental health call rate decreased. These results suggest that embedding a crisis line counselor at an ECC can reduce police responses to suicide calls, avoiding potentially problematic incidents of police escalation. Routing calls ‘upstream’ could reduce need for other ‘downstream’ resources such as emergency departments (Bakko et al., 2025 ), which often present time and financial burdens for consumers, police, and health systems (Harada et al., 2021 ; Hickey et al., 2020 ; B. R. Nordstrom, 2011 ). Emergency department visits are expensive for consumers and can often result in long ‘ED boarding’ delays while waiting for psychiatric inpatient placement (K. Nordstrom et al., 2019 ), tying up time and space for other health patients. Crisis lines primarily address low-risk over-the-phone suicidal ideation, but decreasing police presence may also prevent some incidents from dangerous escalations, particularly among people of color (Saleh et al., 2018 ). People and family members with lived experience of mental health crises prefer models without police, as police presence can escalate the situation (Pope et al., 2023 ). The reduction of police hours in this study shows a potential benefit for both people in mental health distress (through reduced contact hours with police during a crisis), and for law enforcement agencies (through reduced labor associated with mental health crises). Law enforcement agencies operate a majority of ECCs (Choi et al., 2024 ), and would likely be involved in the approval of a program involving the deflection of 911 calls to mental health crisis resources. Seeing the quantifiable reduction of police hours could serve as an incentive for law enforcement administrators to explore embedded 911 counselor or similar mental heath deflection programs. Routing 911 calls to mental health crisis lines could also provide benefits for the person in distress. Mental health crisis lines can de-escalate the caller, activate mobile crisis, and connect people to ongoing mental health resources (Gould et al., 2025 ), but people may not know about their local crisis line and call 911 instead (M. Balfour & Minkoff, 2022 ). Enabling transfers from 911 to crisis lines and other mental health resources could improve care continuity and long-term mental health. Future studies could track outcomes of calls routed from 911 to crisis lines to quantify the downstream impact. Transfer policies should be mindful of racial bias among call-takers; if crisis line call-takers are more likely to transfer Black callers back to 911 (Goldman et al., 2025 ), 911 call-takers may also be less likely to transfer Black callers to mental health crisis lines. These findings suggest that crisis line specialists provide important assistance with 911 to crisis line call routing. A Solari representative learning the ECC’s culture, policies, procedures, practices, and norms likely built personal- and agency-level trust. 911 professionals feel personal liability to callers expressing suicidal ideation (Hedden-Clayton et al., 2025 ; Pew, 2021), and establishing stronger connections between 911 and 988 professionals can assuage fears that the calls would be mishandled or unanswered. The crisis line specialist assisted with translating ECC call routing policies into actual calls, rather than hypothetical training scenarios, which may be important for sustained routing practices. ECCs should welcome collaboration with crisis lines and arrange protocols for meaningful involvement of interdisciplinary crisis line staff. Future qualitative studies could examine mechanisms key to successful routing interventions. We did not find a meaningful change in Mesa’s mental health-related call rate. It is possible that suicidal calls are often first-person calls, making them more appropriate to route to a crisis line and speak directly to crisis line specialists. Third-party bystanders may not have a relationship with the person in distress and may not wish to get involved themselves, preferring law enforcement come out to address the situation. Additionally, a person in mental health distress may experience anosognosia, or unawareness of their mental health condition. If someone experiences a delusion of intruders, for example, they may not have any interest in talking to a crisis line; they may refuse the transfer and only request a police presence. The broader concept of ‘mental health’ calls may also make it difficult to articulate appropriate calls to route. Suicidality has clearer, more distinct concepts (i.e. ideation, threats, means), whereas 911 professionals may be less confident in identifying mental health concepts that vary situationally or require clinical judgment (i.e. speaking ‘quickly’, ‘erratic’ behaviors, ‘disorganized’ thought patterns, etc.). 911 professionals often have limited information, and they may not have the benefit of knowing the accuracy of the caller’s claims. When faced with uncertainty, 911 professionals often default to dispatching law enforcement (Hedden-Clayton et al., 2025 ; Pew, 2021). Studies that analyze the content of the suicide and mental health calls could assess the routing potential with more specificity. Limitations Limited and inconsistent language across ECCs made it difficult to parse meaningful differences in publicly available data. We used data labeled as “Computer Aided Dispatch” (CAD), “calls for service”, and “police incidents”. The variance between the terms suggested potential differences in the types of data being compared: “police dispatch events” in Mesa may represent different activities than “calls for service” in New Orleans. ECC call file names were often accompanied by brief descriptions, though not infrequently the terms like “calls for service” and “police incidents” were used interchangeably, making it difficult to parse whether the differences were meaningful or semantic. There was variation in how the ECCs coded suicide and mental health calls, as mental health calls are often aggregated with larger categories like “check welfare” or “public disturbance”. Suicide calls are a more distinct concept, making the call type more comparable across ECCs. Further analysis could filter calls by priority level, as calls with high priority levels involving weapons or active violence are not likely to be appropriate for routing to 988. However, priority levels were inconsistently defined across the databases, and applying consistent criteria in filtering out high priority calls was not possible. Moreover, given that Mesa’s interventions involved the re-classification of calls previously deemed to require police responses, some of the routable calls in the donor sites were likely to have been classified as high priority incidents. As such, all suicide- and mental health-related calls were included, regardless of priority level, recognizing some of the targeted call counts still required police responses. Cities with particularly low suicide call rates may instead be tracking suicide attempts, rather than callers with suicidal ideation, and would be inappropriate comparisons to Mesa. Unfortunately, we do not know whether the cities track suicide calls differently or whether the cities had lower rates of 911 suicidal ideation calls. When dropping four sites with low suicide call rates (< 10/100,000 people/quarter), the effect size did not change and remained larger than all placebo sites in one post-treatment quarter (p<.01), and larger than all but one placebo in two post-treatment quarters (p=.0625). The supplemental appendix describes robustness checks that dropped individual cities to determine outsized influence and examined non-suicidal call trends to assess possible confounders. Finally, ECC governance structure would likely affect whether ECCs know about 988 and route calls to it. Police-operated ECCs would be less likely to route calls than independently operated ECCs (Choi et al., 2024 ). We did not have a large enough subsample of police-operated ECCs to compare Mesa to ECCs with similar police-operated governance structure; in this study, the police-operated Mesa ECC was compared to a mixed group of independent- and police-operated ECCs, and the suicide call rates were still associated with significant relative declines. Police responses may differ across state-level policies or regional-level cultural differences, but only one southwestern city other than Mesa (Chandler) did not have a similar embedded crisis line counselor intervention and availed downloadable law enforcement response data that contained enough information to perform the analyses. We suggest cautious interpretations of the mental health call analysis given its poor pretreatment fit. Conclusion This study used synthetic controls to assess the impact of embedding a crisis line counselor at a 911 center on law enforcement responses to mental health and suicide calls. Embedding a crisis line counselor at 911 was associated with 22.27–47.96 fewer police responses to suicide calls per quarter per 100,000 people compared to a synthetic control unit, saving Mesa an estimated 228.1–490.6 police hours per quarter. The mental health call response rate did not decrease. Fewer law enforcement responses to suicide calls could decrease acute service utilization and escalated police incidents. ECCs should consider embedding crisis line staff to assist with identification of suicidal calls for appropriate transfer. Deflecting suicide-related 911 calls to mental health crisis lines could invite the mental health crisis system to assume some of the responsibilities associated with the expanded police footprint. Declarations Data availability statement: The authors created a dataset of cities with publicly available 911 data and corresponding URLs, available upon request. Processed data of monthly mental health- and suicide-related 911 call frequencies by city are also available upon request. Acknowledgements: [blinded for peer review] Funding Statement: The authors received no financial support for the research, authorship, and/or publication of this article. Ethics approval statement: This project used publicly available data and was determined to be non-human subjects research by the authors’ university IRB [blinded for peer review]. Conflict of interest statement: The authors have no conflicts of interest to declare. Author contributions: [blinded for peer review] References 988 Lifeline . (2022). 988 Lifeline. https://988lifeline.org/ Abadie, A., Diamond, A., & Hainmueller, J. (2015). Comparative Politics and the Synthetic Control Method. American Journal of Political Science , 59 (2), 495–510. https://doi.org/10.1111/ajps.12116 Aldridge, O. (2025, January 9). Behind the scenes of mental health calls at Austin’s 911 call center . KUT Radio, Austin’s NPR Station. https://www.kut.org/health/2025-01-09/austin-tx-911-mental-health-crisis-calls-24-hours-podcast Alexander, C., & Zealberg, J. J. (1999). Mobile crisis: Moving emergency psychiatry out of the hospital setting. New Directions for Mental Health Services , 1999 (82), 93–99. https://doi.org/10.1002/yd.23319998212 Bakko, M., Swanson, L., Zettner, C., Kok, K., Fukuzawa, H., & Kubiak, S. (2025). A Comparison of Behavioral Health Crisis Response Models in Meeting Behavioral Health Goals and Improving Criminal Legal Diversion. Community Mental Health Journal . https://doi.org/10.1007/s10597-024-01447-4 Balfour, M. E., Hahn Stephenson, A., Delany-Brumsey, A., Winsky, J., & Goldman, M. L. (2022). Cops, Clinicians, or Both? Collaborative Approaches to Responding to Behavioral Health Emergencies. Psychiatric Services , 73 (6), 658–669. https://doi.org/10.1176/appi.ps.202000721 Balfour, M., & Minkoff, K. (2022). A Look at the 988 Soft Launch. Psychiatric Times . https://www.psychiatrictimes.com/view/a-look-at-the-988-soft-launch Bartos, B. J., & Kubrin, C. E. (2018). Can We Downsize Our Prisons and Jails Without Compromising Public Safety? Criminology & Public Policy , 17 (3), 693–715. https://doi.org/https://doi.org/10.1111/1745-9133.12378 Bartos, B. J., McCleary, R., Mazerolle, L., & Luengen, K. (2020). Controlling Gun Violence: Assessing the Impact of Australia’s Gun Buyback Program Using a Synthetic Control Group Experiment. Prevention Science , 21 (1), 131–136. https://doi.org/10.1007/s11121-019-01064-8 Beck, J., Reuland, M., & Pope, L. (2020, November). CASE STUDY: Robust Crisis Care and Diverting 911 Calls to Crisis Lines . Vera Institute of Justice. https://www.vera.org/behavioral-health-crisis-alternatives/robust-crisis-care-and-diverting-911-calls-to-crisis-lines Borum, R., Swanson, J., Swartz, M., & Hiday, V. (1997). Substance Abuse, Violent Behavior, and Police Encounters among Persons with Severe Mental Disorder. Journal of Contemporary Criminal Justice , 13 (3), 236–250. https://doi.org/10.1177/1043986297013003004 Center for American Progress. (2020, October 28). The Community Responder Model . https://www.americanprogress.org/article/community-responder-model/ Center for Police Research and Policy. (2021). Assessing the Impact of Mobile Crisis Teams: A Review of Research . University of Cincinnati. https://www.informedpoliceresponses.com/_files/ugd/313296_8d01cdc7187a489893197f2d07300ee6.pdf Choi, H., Lerner, J., & Neusteter, R. (2024). Understanding 911 Governance: Examining Factors Influencing and Related to the Governance of our Nation’s Emergency Communications Centers (Transform911). University of Chicago Health Lab. https://bpb-us-w2.wpmucdn.com/voices.uchicago.edu/dist/e/2911/files/2024/11/Making-911-Independent-and-Equal_Brief_FINAL.pdf CIT International—CIT Support Training for 911 . (n.d.). Retrieved August 28, 2025, from https://www.citinternational.org/CITST911 City of Mesa. (2024a). Data Portal | Mesa Data Hub . https://data.mesaaz.gov/ City of Mesa. (2024b, December 5). Police Dispatch Events 2021-present . Mesa Data Hub. https://data.mesaaz.gov/Police/Police-Dispatch-Events-2021-present/izhu-764k/about_data Cojean, S., Combalbert, N., & Taillandier-Schmitt, A. (2020). Psychological and sociological factors influencing police officers’ decisions to use force: A systematic literature review. International Journal of Law and Psychiatry , 70 , 101569. https://doi.org/10.1016/j.ijlp.2020.101569 Comartin, E. B., Swanson, L., & Kubiak, S. (2019). Mental Health Crisis Location and Police Transportation Decisions: The Impact of Crisis Intervention Team Training on Crisis Center Utilization. Journal of Contemporary Criminal Justice , 35 (2), 241–260. https://doi.org/10.1177/1043986219836595 Compton, M. T., Bakeman, R., Broussard, B., Hankerson-Dyson, D., Husbands, L., Krishan, S., Stewart-Hutto, T., D’Orio, B. M., Oliva, J. R., Thompson, N. J., & Watson, A. C. (2014). The Police-Based Crisis Intervention Team (CIT) Model: II. Effects on Level of Force and Resolution, Referral, and Arrest. Psychiatric Services , 65 (4), 523–529. https://doi.org/10.1176/appi.ps.201300108 Compton, M. T., Pope, L. G., & Watson, A. C. (2024). Community-Based Mental Health Crisis Response: An Overview of Models and Workforce Implications. Psychiatric Clinics of North America . https://doi.org/10.1016/j.psc.2024.04.004 Currier, G. W., Fisher, S. G., & Caine, E. D. (2010). Mobile Crisis Team Intervention to Enhance Linkage of Discharged Suicidal Emergency Department Patients to Outpatient Psychiatric Services: A Randomized Controlled Trial. Academic Emergency Medicine , 17 (1), 36–43. https://doi.org/10.1111/j.1553-2712.2009.00619.x Davis, J., Norris, S., Schmitt, J., Shem-Tov, Y., & Strickland, C. (2025). Mobile Crisis Response Teams Support Better Policing: Evidence from CAHOOTS (Working Paper No. 33761). National Bureau of Economic Research. https://doi.org/10.3386/w33761 Dee, T. S., & Pyne, J. (2022). A community response approach to mental health and substance abuse crises reduced crime. Science Advances , 8 (23), eabm2106. https://doi.org/10.1126/sciadv.abm2106 Dyches, H., Biegel, D. E., Johnsen, J. A., Guo, S., & Min, M. O. (2002). The Impact of Mobile Crisis Services on the Use of Community-Based Mental Health Services. Research on Social Work Practice , 12 (6), 731–751. https://doi.org/10.1177/104973102237470 Federal Communications Commission. (2024, October 31). 911 Master PSAP Registry . https://www.fcc.gov/general/9-1-1-master-psap-registry Ferman, B., Pinto, C., & Possebom, V. (2020). Cherry Picking with Synthetic Controls. Journal of Policy Analysis and Management , 39 (2), 510–532. https://doi.org/10.1002/pam.22206 Fuller, D., Lamb, R., Biasotti, M., & Snook, J. (2015). Overlooked in the Undercounted: The role of mental illness in fatal law enforcement encounters . Treatment Advocacy Center. https://www.treatmentadvocacycenter.org/storage/documents/overlooked-in-the-undercounted.pdf Galiani, S., & Quistorff, B. (2017). The Synth_Runner Package: Utilities to Automate Synthetic Control Estimation Using Synth. The Stata Journal , 17 (4), 834–849. https://doi.org/10.1177/1536867X1801700404 Gasior, M. (2020, December 22). Policing the mentally ill . PowerDMS. https://www.powerdms.com/policy-learning-center/policing-the-mentally-ill Geller, J. L., Fisher, W. H., & McDermeit, M. (1995). A national survey of mobile crisis services and their evaluation. Psychiatric Services (Washington, D.C.) , 46 (9), 893–897. https://doi.org/10.1176/ps.46.9.893 Gillig, P., Dumaine, M., & Hillard, J. (1990). Whom Do Mobile Crisis Services Serve? Hospital & Community Psychiatry , 41 (7), 804–805. https://doi.org/10.1176/ps.41.7.804 Gillooly, J. W., & Thacher, D. (2024). How the Public Became the Caller: The Emergence of Reactive Policing, 1880–1970. Law & Social Inquiry , 1–33. https://doi.org/10.1017/lsi.2024.9 Goldman, M., Elser, A., Yeh, H.-H., McDaniel, M., Ma, L., Ahmedani, B. K., & Foster, A. A. (2025). Demographic and Clinical Characteristics of Mental Health Crisis Line Callers Who Were Transferred to 911. Psychiatric Services , 76 (3), 282–285. https://doi.org/10.1176/appi.ps.20240050 Gottlieb, A., Epperson, M., Giugno, L., & Jung, H.-M. (2024). What happens to imprisonment rates when a progressive prosecutor is elected: Quasi-experimental evidence from Cook County. Punishment & Society , 14624745241288185. https://doi.org/10.1177/14624745241288185 Gottlieb, A., Harper, T., & Jung, H.-M. (2024). Does racial impact statement reform reduce Black–White disparities in imprisonment: Mixed methods evidence from Minnesota. Law & Policy , n/a (n/a). https://doi.org/10.1111/lapo.12252 Gould, M. S., Lake, A. M., Galfalvy, H., Kleinman, M., Munfakh, J. L., Wright, J., & McKeon, R. (2018). Follow-up with Callers to the National Suicide Prevention Lifeline: Evaluation of Callers’ Perceptions of Care. Suicide and Life-Threatening Behavior , 48 (1), 75–86. https://doi.org/10.1111/sltb.12339 Gould, M. S., Lake, A. M., Munfakh, J. L., Galfalvy, H., Kleinman, M., Williams, C., Glass, A., & McKeon, R. (2016). Helping Callers to the National Suicide Prevention Lifeline Who Are at Imminent Risk of Suicide: Evaluation of Caller Risk Profiles and Interventions Implemented. Suicide and Life-Threatening Behavior , 46 (2), 172–190. https://doi.org/10.1111/sltb.12182 Gould, M. S., Lake, A. M., Port, M. S., Kleinman, M., Hoyte‐Badu, A. M., Rodriguez, C. L., Chowdhury, S. J., Galfalvy, H., & Goldstein, A. (2025). National Suicide Prevention Lifeline (Now 988 Suicide and Crisis Lifeline): Evaluation of Crisis Call Outcomes for Suicidal Callers. Suicide & Life-Threatening Behavior , 55 (3), e70020. https://doi.org/10.1111/sltb.70020 Guo, S., Biegel, D. E., Johnsen, J. A., & Dyches, H. (2001). Assessing the Impact of Community-Based Mobile Crisis Services on Preventing Hospitalization. Psychiatric Services , 52 (2), 223–228. https://doi.org/10.1176/appi.ps.52.2.223 Harada, M. Y., Lara-Millán, A., & Chalwell, L. E. (2021). Policed Patients: How the Presence of Law Enforcement in the Emergency Department Impacts Medical Care. Annals of Emergency Medicine , 78 (6), 738–748. https://doi.org/10.1016/j.annemergmed.2021.04.039 Hedden-Clayton, B., Carl, A., Swanson, L., Comartin, E., & Kubiak, S. (2025). Deflecting people experiencing a mental health crisis from the criminal-legal system at dispatch: Findings from calls-for-service reports and statewide survey data. Evaluation and Program Planning , 102727. https://doi.org/10.1016/j.evalprogplan.2025.102727 Hickey, K., Walther, C., King, J., Layne, H., Besler, C., Andrzejewski, V., & Gijada, J. (2020). Connecting Law Enforcement and Emergency Department Providers to Improve Access to Mental Health Services. Journal of Psychosocial Nursing and Mental Health Services , 58 (8), 24–30. https://doi.org/https://doi.org/10.3928/02793695-20200624-02 Hugo, M., Smout, M., & Bannister, J. (2002). A comparison in hospitalization rates between a community-based mobile emergency service and a hospital-based emergency service. Australian and New Zealand Journal of Psychiatry , 36 (4), 504–508. https://doi.org/10.1046/j.1440-1614.2002.01042.x Kim, S., & Kim, H. (2017). Determinants of the use of community-based mental health services after mobile crisis team services: An empirical approach using the Cox proportional hazard model. Journal of Community Psychology , 45 (7), 877–887. https://doi.org/10.1002/jcop.21899 Koziarski, J., Ferguson, L., & Huey, L. (2022). Shedding Light on the Dark Figure of Police Mental Health Calls for Service. Policing: A Journal of Policy and Practice , 16 (4), 696–706. https://doi.org/10.1093/police/paac006 Mesa Police Department. (2019, October 16). Overdose or Suicide Attempt, COM 13.4 . Mesa Police Department. (2020, August 17). Persons Affected by Mental Illness or in Crisis . Morabito, M. S., Socia, K., Wik, A., & Fisher, W. H. (2017). The nature and extent of police use of force in encounters with people with behavioral health disorders. International Journal of Law and Psychiatry , 50 , 31–37. https://doi.org/10.1016/j.ijlp.2016.10.001 Munetz, M. R., & Griffin, P. A. (2006). Use of the Sequential Intercept Model as an Approach to Decriminalization of People With Serious Mental Illness. PSYCHIATRIC SERVICES , 57 (4). National Emergency Number Association. (2021). 9-1-1 Statistics . https://www.nena.org/page/911statistics NENA. (2021). 9-1-1 Statistics . https://www.nena.org/page/911Statistics Nordstrom, B. R. (2011). The effects of neighborhood characteristics on rearrest after participation in drug teatment court: A re-analysis of the Baltimore city drug treatment court experiment. Dissertations Available from ProQuest , 1–123. Nordstrom, K., Berlin, J. S., Nash, S. S., Shah, S. B., Schmelzer, N. A., & Worley, L. L. M. (2019). Boarding of Mentally Ill Patients in Emergency Departments: American Psychiatric Association Resource Document. Western Journal of Emergency Medicine , 20 (5), 690–695. https://doi.org/10.5811/westjem.2019.6.42422 O’Connell, K. L., Jacobson, S. V., Ton, A. T., & Law, K. C. (2022). Association between race and socioeconomic factors and suicide-related 911 call rate. Social Science & Medicine , 306 , 115106. https://doi.org/10.1016/j.socscimed.2022.115106 Patton, C. L., & Fremouw, W. J. (2016). Examining “suicide by cop”: A critical review of the literature. Aggression and Violent Behavior , 27 , 107–120. https://doi.org/10.1016/j.avb.2016.03.003 Pew. (2021, October 26). New Research Suggests 911 Call Centers Lack Resources to Handle Behavioral Health Crises . https://pew.org/30LhARu Pope, L. G., Patel, A., Fu, E., Zingman, M., Warnock, A., Ellis, S., Ashekun, O., Watson, A., Wood, J., & Compton, M. T. (2023). Crisis Response Model Preferences of Mental Health Care Clients With Prior Misdemeanor Arrests and of Their Family and Friends. Psychiatric Services , 74 (11), 1163–1170. https://doi.org/10.1176/appi.ps.20220363 Purtle, J., McSorley, A.-M. M., Adera, A. L., & Lindsey, M. A. (2023). Use, Potential Use, and Awareness of the 988 Suicide and Crisis Lifeline by Level of Psychological Distress. JAMA Network Open , 6 (10), e2341383. https://doi.org/10.1001/jamanetworkopen.2023.41383 Saleh, A. Z., Appelbaum, P. S., Liu, X., Scott Stroup, T., & Wall, M. (2018). Deaths of people with mental illness during interactions with law enforcement. International Journal of Law and Psychiatry , 58 , 110–116. https://doi.org/10.1016/j.ijlp.2018.03.003 SAMHSA. (2023, June 15). 988 Frequently Asked Questions . https://www.samhsa.gov/mental-health/988/faqs SAMHSA. (2025). 2025 National Guidelines for a Behavioral Health Coordinated System of Crisis Care (Nos. PEP24-01–037). https://library.samhsa.gov/sites/default/files/national-guidelines-crisis-care-pep24-01-037.pdf Saunders, H. (2025, July 14). Demand for 988 Continues to Grow at Third Anniversary. KFF . https://www.kff.org/mental-health/demand-for-988-continues-to-grow-at-third-anniversary/ Solari. (2025, April). Crisis Response Network Performance Dashboard . Tableau Public. https://public.tableau.com/app/profile/crisis.network/viz/AZ600StatewideDashboard/AZ600StatewideDashboard Swanson, L., Nelson, V., Comartin, E. B., Kubiak, S., Putans, L., Hambrick, N., Ray, B., Tillander, L., Washington, A., Butkiewicz, R., & Costello, M. (2022). Assessing County-Level Behavioral Health and Justice Systems with the Sequential Intercept Model Practices, Leadership, and Expertise Scorecard. Community Mental Health Journal . https://doi.org/10.1007/s10597-022-01042-5 Swanson, L., Zettner, C., Watson, A., Hinojosa, M., Roddy, J., & Kubiak, S. (2025). Eleven‐Month Arrest Outcomes Among Three Crisis Response Models in Michigan. Psychiatric Research and Clinical Practice , n/a-n/a. https://doi.org/10.1176/appi.prcp.20240100 Transform 911. (2022). Transform 911: A Blueprint for Change . University of Chicago Health Lab. https://bpb-us-w2.wpmucdn.com/voices.uchicago.edu/dist/e/2911/files/2022/08/Transform911-Blueprint-for-Change.pdf Ueda, M., Heflin, C. M., Liu, Y., Yu, Q., & Ramanathan, S. (2026). Understanding Factors Associated with 911 and 988 Use in Mental Health Crises. Community Mental Health Journal , 62 (3), 527–536. https://doi.org/10.1007/s10597-025-01545-x United States Census Bureau. (2024). Census Bureau Data . Explore Census Data. https://data.census.gov/ U.S. Census Bureau. (2023). Mesa city, Arizona . https://www.census.gov/quickfacts/fact/table/mesacityarizona/PST045223 Vakkalanka, J. P., Neuhaus, R. A., Harland, K. K., Clemsen, L., Himadi, E., & Lee, S. (2021). Mobile Crisis Outreach and Emergency Department Utilization: A Propensity Score-matched Analysis. Western Journal of Emergency Medicine , 22 (5), 1086–1094. https://doi.org/10.5811/westjem.2021.6.52276 Ward-Ciesielski, E. F., & Rizvi, S. L. (2021). The potential iatrogenic effects of psychiatric hospitalization for suicidal behavior: A critical review and recommendations for research. Clinical Psychology: Science and Practice , 28 (1), 60–71. https://doi.org/10.1111/cpsp.12332 Waters, R. (2021). Enlisting Mental Health Workers, Not Cops, In Mobile Crisis Response. Health Affairs , 40 (6), 864–869. https://doi.org/10.1377/hlthaff.2021.00678 Watson, A. C., Compton, M. T., & Draine, J. N. (2017). The crisis intervention team (CIT) model: An evidence-based policing practice? Behavioral Sciences & the Law , 35 (5–6), 431–441. https://doi.org/10.1002/bsl.2304 Watson, A. C., Compton, M. T., & Pope, L. G. (2019). Crisis Response Services for People with Mental Illnesses or Intellectual and Developmental Disabilities: A review of the literature on police-based and other first response models . Vera Institute of Justice. (https://www.vera.org/). https://www.vera.org/publications/crisis-response-services-for-people-with-mental-illnesses-or-intellectual-and-developmental-disabilities Wiltshire, J. C. (2025). allsynth: (Stacked) Synthetic Control Bias-Correction Utilities for Stata. Working Paper . https://static1.squarespace.com/static/5e0fdcef27e0945c43fab131/t/67aa6bac859390516fcf8f21/1739221941939/allsynth_Wiltshire.pdf Tables Table 1. Characteristics of cities in the synthetic control analyses City State Population (mean 2018-2022) ECC governance Quarterly average per 100k population All 911 calls Mental health Suicide Boise City ID 233,662 Police 7,784 1260.7 - Boston MA 673,948 Police 3,276 - 2.4 Cary NC 173,971 Independent 671 - 6.9 Chandler AZ 271,218 Police 7,150 52.6 71.0 Charlotte NC 882,538 Police 2,493 - 14.7 Chula Vista CA 274,343 Police 6,683 0.6 2.8 Cincinnati OH 306,777 Independent 29,510 625.4 - College Station TX 120,333 Police 20,695 - 1.1 Everett WA 111,229 Independent 22,112 667.7 151.5 Fort Lauderdale FL 182,509 Police 22,946 252.4 81.0 Hartford CT 120,869 Independent 43,932 400.9 87.2 Irving TX 249,326 Police 17,969 - 28.8 Lincoln NE 290,308 Independent 2,794 - 34.2 Los Angeles CA 3,901,675 Police 9,940 - 75.3 Mesa (test case) AZ 510,680 Police 7,112 17.2 132.2 Miami FL 453,098 Police 16,807 245 11.7 Nashville TN 677,047 Independent 23,044 29.2 75.9 New Orleans LA 382,330 Independent 22,107 274.6 94.0 New York NY 8,453,041 Police 19,712 988.3 - Norman OK 127,106 Police 11,572 - 179.6 Portland OR 647,354 Independent 9,462 200 - San Diego CA 1,399,116 Police 10,148 300 - San Francisco CA 850,494 Independent 19,328 387.7 108.9 San Jose CA 1,002,486 Police 6,398 80 - Seattle WA 743,473 Independent 4,363 110.1 90.0 St. Petersburg FL 262,077 Police 18,247 410 119.0 Tacoma* WA 219,008 Independent 15,641 - 186.5 Mean 871,112 14,144 350.1 74.0 Standard Deviation 1,685,134 9,855 345.3 58.5 *Tacoma launched an embedded clinician program in June, 2023; its effects would not impact this study period Table 2 . Synthetic control weights, pre-intervention fit, and post-intervention outcomes of suicide 911 call rates per 100,000 people. Weights: Everett=.213, New Orleans=.037, Norman=.036, Seattle=.334, Tacoma=.38 Quarter Suicide call rate per 100,000 people Standardized p-value Mesa, AZ Synthetic Control Difference 2018q1 158.48 157.02 1.46 - 2018q2 165.37 160.41 4.96 - 2018q3 163.60 160.46 3.14 - 2018q4 157.89 160.74 -2.85 - 2019q1 144.78 147.45 -2.67 - 2019q2 155.40 158.58 -3.18 - 2019q3 170.46 164.90 5.56 - 2019q4 148.45 149.36 -0.91 - 2020q1 148.25 154.14 -5.89 - 2020q2 134.61 133.57 1.04 - 2020q3 159.32 142.14 17.19 - 2020q4 132.64 139.42 -6.78 - 2021q1 124.10 126.54 -2.45 - 2021q2 125.86 135.24 -9.38 - 2021q3 147.66 134.23 13.43 0.20 2021q4 110.15 132.42 -22.27 0.05 2022q1 93.12 128.40 -35.29 0.15 2022q2 97.41 133.99 -36.58 0.10 2022q3 97.61 140.67 -43.07 0.05 2022q4 84.92 132.88 -47.96 0.00 2023q1 89.91 134.66 -44.75 0.10 2023q2 98.70 133.78 -35.07 0.20 Table 3. Synthetic control weights, pre-intervention fit, and post-intervention outcomes of mental health call rates Quarter Mental health call rate per 100,000 people Standardized p-value Mesa, AZ Synthetic Control Difference 2018q3 21.07 22.36 -1.29 - 2018q4 24.61 22.44 2.17 - 2019q1 28.57 19.70 8.87 - 2019q2 24.71 22.04 2.67 - 2019q3 19.50 20.38 -0.88 - 2019q4 20.66 17.58 3.08 - 2020q1 16.21 17.41 -1.20 - 2020q2 14.83 19.91 -5.08 - 2020q3 18.58 19.94 -1.36 - 2020q4 18.38 15.96 2.43 - 2021q1 11.19 17.80 -6.61 - 2021q2 12.57 18.37 -5.80 - 2021q3 24.74 19.90 4.84 0.65 2021q4 16.30 18.43 -2.13 0.71 2022q1 13.86 16.29 -2.43 0.59 2022q2 12.49 16.56 -4.07 0.71 2022q3 12.10 18.74 -6.64 0.71 2022q4 11.52 16.74 -5.22 0.71 2023q1 9.19 18.28 -9.09 0.71 2023q2 13.49 18.13 -4.65 0.82 Additional Declarations No competing interests reported. Supplementary Files MesasynthmethodsappendixCMHJ032526.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9226560","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":614027892,"identity":"40509333-69df-48a7-ae85-e4b7ae79ffa9","order_by":0,"name":"Leonard Swanson","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA6ElEQVRIiWNgGAWjYBACPgjFzMDPwMDGkAAVlcCnhQ2mRbKBZC0GB2BsglrYm489+LnH2t74RvKzBw9q7Oz6GZgP3ubBp4XnWLphz7P0xG030swNEo4lJ89sYEu2xqtFIsdMgufA4QSzGzlsEokNzMkGB3jMpPFqkX9jJvnnwGF74xlgLfVALfzf8GuRAJl54DDjBgmwlsN2QFvY8GvhSUuTljmQnjjjzDMziYRjxxMkm9mMLefg0cLPfviY5JsD1vb87cnPJH/UVNvzszc/vPEGjxYMAAwBUpSDgD2pGkbBKBgFo2D4AwD4qESFtcROLQAAAABJRU5ErkJggg==","orcid":"","institution":"University of Chicago","correspondingAuthor":true,"prefix":"","firstName":"Leonard","middleName":"","lastName":"Swanson","suffix":""},{"id":614027893,"identity":"b08c0f15-4d71-4d10-b0bf-431693da62b4","order_by":1,"name":"Aaron Gottlieb","email":"","orcid":"","institution":"University of Chicago","correspondingAuthor":false,"prefix":"","firstName":"Aaron","middleName":"","lastName":"Gottlieb","suffix":""},{"id":614027894,"identity":"c0b376a7-e626-4f81-85b2-904fd6c0c2bb","order_by":2,"name":"Margaret Thomas","email":"","orcid":"","institution":"University of Chicago","correspondingAuthor":false,"prefix":"","firstName":"Margaret","middleName":"","lastName":"Thomas","suffix":""}],"badges":[],"createdAt":"2026-03-25 19:08:58","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9226560/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9226560/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":105831359,"identity":"2f791b2e-0d54-4b20-ae96-485c45bff472","added_by":"auto","created_at":"2026-03-31 14:47:31","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":98913,"visible":true,"origin":"","legend":"\u003cp\u003eSynthetic control analysis of Mesa’s suicide call rate per 100,000 people after embedding a crisis line counselor in a 911 emergency communications center.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-9226560/v1/c80e1a86faf75dd1b102f443.png"},{"id":105831360,"identity":"6fdbfb93-7a69-4104-841c-664727eb0e53","added_by":"auto","created_at":"2026-03-31 14:47:32","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":95815,"visible":true,"origin":"","legend":"\u003cp\u003eSynthetic control analysis of Mesa’s mental health call rate per 100,000 people after embedding a crisis line counselor in a 911 emergency communications center.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-9226560/v1/1ed691362f4c645aa41ac0bf.png"},{"id":106093113,"identity":"d9f13ed6-3f6e-4abd-a05c-8f8e76b6d751","added_by":"auto","created_at":"2026-04-03 11:34:57","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1021830,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9226560/v1/0d53e397-056a-46ca-bd2c-0e060a60ddb4.pdf"},{"id":105831358,"identity":"f86d3876-2846-4823-bd95-91d8c4c8a5cd","added_by":"auto","created_at":"2026-03-31 14:47:31","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":113548,"visible":true,"origin":"","legend":"","description":"","filename":"MesasynthmethodsappendixCMHJ032526.docx","url":"https://assets-eu.researchsquare.com/files/rs-9226560/v1/30189c35b4c45d9277871553.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Deflecting suicide and mental health calls from law enforcement with an embedded crisis line specialist at a 911 center: A synthetic control analysis","fulltext":[{"header":"Introduction","content":"\u003cp\u003eLaw enforcement officers are the predominant responders to suicide and mental health crises (M. E. Balfour et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Borum et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e1997\u003c/span\u003e; Gasior, 2020; Koziarski et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Officers can struggle with suicide and mental health crisis responses (Compton et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2014\u003c/span\u003e), as their presence can incite fears that escalate to arrests and use of force, or hospitalization (Cojean et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Fuller et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Morabito et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Law enforcement trainings to increase officer awareness of mental health issues, self-efficacy, and de-escalation skills have proliferated (Watson et al., \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Trainings, such as the 40-hour training in the Crisis Intervention Team model, can increase officer referrals to mental health crisis resources (Comartin et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), but system discontinuities prevent transfers from traditional 911 pathways to the mental health crisis response continuum (M. E. Balfour et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Law enforcement predominately transport people in mental health crises to emergency departments (Bakko et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), which can be time consuming, frustrating and ineffective (Harada et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Hickey et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Emergency departments can be traumatic: mental health crisis patients sometimes wait weeks for placement at psychiatric hospitals in a process often referred to as \u0026lsquo;ED-boarding\u0026rsquo; (K. Nordstrom et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). After immediate care, emergency departments struggle with connecting people to post-crisis mental health treatment (Currier et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Vakkalanka et al., \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The traditional public mental health crisis response system needs reform.\u003c/p\u003e \u003cp\u003eThe Black Lives Matter protests in 2020 spurred momentum for alternative response models that decrease the role and presence of law enforcement in mental health crises. Some mental health crisis models can handle calls that would otherwise be handled by police, decreasing the total incidents that could escalate to law enforcement use of force or arrest (M. E. Balfour et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Compton et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). In-person community response teams are one example of a model that can address 911 calls involving social and mental health needs, suggesting that 911 mental health crisis calls can indeed be safely resolved by civilians (Davis et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Dee \u0026amp; Pyne, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Waters, \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Crises resolved upstream (e.g. mental health crisis lines, mobile crisis teams) could prevent the utilization of downstream acute crisis services (e.g. emergency departments, psychiatric hospitalization, arrests) that can be traumatic, expensive, and often in high demand (Munetz \u0026amp; Griffin, \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Swanson et al., \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Ward-Ciesielski \u0026amp; Rizvi, \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eMental health systems are developing systemic frameworks to standardize and expand crisis response services as alternatives to 911 default law enforcement pathways. The Substance Abuse and Mental Health Services Administration describes crisis system components as a three-part framework: \u0026ldquo;someone to contact\u0026rdquo; (e.g., call centers), \u0026ldquo;someone to respond\u0026rdquo; (e.g., mobile crisis teams), and \u0026ldquo;a safe place for help\u0026rdquo; (e.g., stabilization facilities) (SAMHSA, 2025). In July, 2022, the nationwide three-digit 988 crisis hotline launched as a toll-free number for around-the-clock suicidal crisis counseling, e.g., \u0026ldquo;someone to contact\u0026rdquo; (\u003cem\u003e988 Lifeline\u003c/em\u003e, 2022; M. Balfour \u0026amp; Minkoff, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The 988 hotline allowed people to reach the National Suicide Prevention Lifeline (NSPL) with an easy-to-remember three-digit number. When someone contacts a mental health crisis line, call takers answer, listen, provide support, empathize, and connect the caller with local resources. Crisis line callers describe 988 as helpful, even life-saving (Gould et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). When NSPLs make follow-up calls to callers, clients report that the intervention prevents them from killing themselves and keeps them safe (Gould et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Rather than a single call center, there are 200 distinct 988 call centers spread across the country; the decentralization is meant to encourage call center familiarity with and referral to local resources (\u003cem\u003e988 Lifeline\u003c/em\u003e, 2022). Crisis line staff determine whether a call presents an imminent risk, which is when a person expresses both immediate intent and capability of suicide (Gould et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Crisis line calls that present imminent risk can be routed to emergency services (e.g. law enforcement, emergency medical services). Less than 2% of 988 calls require emergency assistance and are routed to 911 emergency services (SAMHSA, 2023). About one in four crisis line calls involving imminent risk are routed to emergency services without collaboration with the caller (Gould et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Goldman and colleagues (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) assessed the characteristics of calls routed from 988 to 911: people who are Black, male, or at imminent risk of harm were more likely to be rerouted from 988 to 911 for active rescue. However, only a third of 988 callers with serious distress indicated they would call 988 again (Purtle et al., \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eSome mental health crisis lines have the capability of dispatching a mobile crisis team, offering an alternative in-person response to law enforcement. There is a thin but growing body of research on the mobile crisis model. Behavioral health practitioners have been responding to mental health crises in community settings since the 1960s (Center for Police Research and Policy, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Attending to crises outside of the hospital or healthcare setting is the key distinction that separates mobile crisis from other crisis services provided by behavioral health service agencies (Alexander \u0026amp; Zealberg, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e1999\u003c/span\u003e; Geller et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e1995\u003c/span\u003e; Gillig et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e1990\u003c/span\u003e; Guo et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2001\u003c/span\u003e). Mobile crisis teams generally attempt to de-escalate crises, screen clients to determine eligibility for psychiatric hospitalization, provide safety plans, and link people to post-crisis mental health care (Center for Police Research and Policy, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Dyches et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2002\u003c/span\u003e; Watson et al., \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Compared to emergency department-based crisis services, mobile crisis teams are effective in diverting people from hospitalizations (Guo et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2001\u003c/span\u003e; Hugo et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2002\u003c/span\u003e; Vakkalanka et al., \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) and connecting people to mental health services after a crisis incident (Currier et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Dyches et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2002\u003c/span\u003e; Kim \u0026amp; Kim, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Compared to law enforcement responses, mobile crisis teams may also prevent future arrests (Swanson et al., \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eDespite crisis lines\u0026rsquo; ability to de-escalate mental health crises and connect people to mental health services, there are reasons to suspect that people call 911 for mental health crises instead. The most utilized public resource is 911 (Transform 911, \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2022\u003c/span\u003e); an estimated 240\u0026nbsp;million 911 calls occur every day (National Emergency Number Association, \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Several 911 calls do not fit the original scope of law enforcement response (Transform 911, \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). A recent historical analysis discusses how the 911 call for service system has shaped and expanded the role of law enforcement to attend to caller needs beyond those relating to crime (Gillooly \u0026amp; Thacher, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Mental health-related 911 calls are a prime example of the expanded police footprint, as its utilization as a public resource has grown to include specialized emergencies beyond the capabilities of the responders (Gillooly \u0026amp; Thacher, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). There are no uniform metrics to determine the proportion of 911 calls that are mental health-related, let alone those that flag suitability to route to mental health crisis lines. Some analyses estimate 20% of 911 calls are mental health-related (Center for American Progress, 2020). However, even a conservative estimate of 10% of the annual U.S. 911 calls (24\u0026nbsp;million) would far outnumber the annual calls to 988 (6\u0026nbsp;million) (Saunders, 2025). It is possible that several mental health-related 911 calls are escalated with active weapons or threats where law enforcement presence is necessary and would not appropriate for a mental health crisis line. People who call 911 may provoke police action by presenting situations that involve weapons, barricades, or otherwise aggressive behavior that warrants in-person police responses (Patton \u0026amp; Fremouw, \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). However, it is unlikely that all of the estimated 24\u0026nbsp;million mental health-related 911 calls are escalated to levels where police response is necessary, and several may be suitable to deflect to mental health service responses.\u003c/p\u003e \u003cp\u003eOne reason behind the utilization differences between 911 and mental health crisis lines is the lack of awareness of mental health crisis lines. A recent survey suggests that only about one in five Americans were aware of 988, whereas over half of respondents said they would call 911 (Ueda et al., \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2026\u003c/span\u003e). After receiving information about 988, respondents expressed a willingness to use it (Ueda et al., \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2026\u003c/span\u003e). Other reasons for lack of 988 utilization include law enforcement involvement, forced hospitalization, and service fees (Ueda et al., \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2026\u003c/span\u003e), though these reasons would not be applicable to 911 callers, since people who call 911 often expect police responses. It is likely that people call 911 for mental health crises because they do not know who else to call, and such callers would benefit from an intervention that transfers 911 calls to mental health crisis resources.\u003c/p\u003e \u003cp\u003eTo our knowledge, interventions that route 911 calls to mental health crisis lines have not been empirically tested. Over-the-phone crisis resolutions are upstream interventions that could deflect calls from legal entanglement (Beck et al., 2020; Munetz \u0026amp; Griffin, \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). Emergency Communications Centers (ECCs), often also referred to as public safety answering points, could be key to diverting calls from law enforcement (Gillooly \u0026amp; Thacher, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), but most ECCs do not have standard protocols for routing 911 calls to crisis resources (Pew, 2021). There are an estimated 240\u0026nbsp;million 911 calls answered by 10,000 ECCs in the U.S. annually (Federal Communications Commission, 2024; NENA, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), each with their own governance and routing protocols (Choi et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). 911 professionals often default to law enforcement responses, expressing concern about legal liability and lack of trust with engaging alternative crisis resources (Hedden-Clayton et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Pew, 2021). Some communities deescalate 911 crisis calls over the phone, though their approaches vary (Beck et al., 2020). Mental health crisis service models based at ECCs include: training 911 professionals about mental health calls (\u003cem\u003eCIT International - CIT Support Training for 911\u003c/em\u003e, n.d.), embedding crisis professionals in 911 centers to answer calls directly (Aldridge, 2025), advising transfer of 911 calls to other crisis resources (Beck et al., 2020), or a combination of these strategies.\u003c/p\u003e \u003cp\u003eThis study examines whether embedding a crisis line counselor in Mesa, Arizona\u0026rsquo;s 911 ECC diverted law enforcement from mental health and suicidal crises by transferring calls to a local mental health crisis line. A synthetic control analysis compares Mesa\u0026rsquo;s trends of police response rates to mental health and suicide calls after the embedded crisis line counselor intervention. Research questions were: when compared to synthetic control groups, did embedding a crisis line counselor at a 911 ECC decrease police responses to A) suicide calls or B) mental health calls? We hypothesized that embedding a crisis line counselor in an ECC would significantly reduce police responses to both call types.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cp\u003e\u003cstrong\u003eStudy Setting\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study setting is Mesa, Arizona, a metropolitan city in the United States with 510,680 people, or 3,635 people per square mile. Mesa\u0026rsquo;s median household income is $73,766, and 10.5% of households live under the federal poverty line (U.S. Census Bureau, 2023). Mesa is 61% white, has 27% people of two or more races, 91% of adults have graduated high school, and 31.5% have a Bachelor\u0026rsquo;s degree or higher according to the U.S. Census (U.S. Census Bureau, 2023). Areas with higher proportions of non-white people, private health insurance, and lower levels of education have shown lower 911 suicidal call rates (O\u0026rsquo;Connell et al., 2022). The Mesa Police Department (MPD) operates the ECC, receiving around 1.4 million calls per year, and assumes responsibility for dispatching police, fire, and medical calls. Mesa dispatches police to ~245,000 911 calls per year (48 calls per 100 people per year) (City of Mesa, 2024a). Solari is a crisis line that takes 988 calls for the state of Arizona. Solari launched in 2007 as a 24/7 crisis line and has since expanded to serious mental illness determinations, peer warmlines, and 211 resource navigation responses. Solari GPS-tracks and dispatches mobile crisis teams in central Arizona, receiving over 35,000 Arizona crisis calls per month (Solari, 2025).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCall diversion programs in Mesa\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSolari and the Mesa Police Department worked together to identify mental health and suicide 911 calls that could be routed to a Solari\u0026rsquo;s crisis line. First, the two agencies targeted suicide calls. The Mesa Police Department drafted a policy that permitted routing of suicidal threats without plans or means, effective on October 16, 2019. In the month before the policy was enacted, the 911 professionals at the Mesa Police department received one hour of training that described the updated policy language, what the Solari crisis line did, how to identify suicidal and mental health behavior appropriate for call routing, how to categorize the calls, and how to transfer the calls to Solari. The training was jointly facilitated by Solari and the Mesa Police Department. 911 professionals would identify and code the calls before recommending the option to the caller, who would consent before their call would be transferred (Mesa Police Department, 2019). On August 17, 2020, the Mesa Police Department enacted a similar policy and training that enabled routing of mental health-related calls. The logistics of routing mental health calls resembled those of the suicide call policy change: for a mental health call to be eligible for transfer, there could not be apparent danger to self or others, and the caller had to consent to the transfer (Mesa Police Department, 2020). \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOn June 5, 2021, Solari embedded a licensed professional counselor at the ECC to help with the identification of additional transferable calls. The crisis line counselor learned the ECC procedures, norms, and coding practices. The crisis line specialist did not assume call taking responsibilities but rather advised in routing and dispatch decisions made by the 911 professionals. Specifically, he listened for the types and severity of calls that the Solari crisis line would be able to handle. The crisis line counselor was involved in coaching and direct training of the 911 professionals during shifts. Making decisions on specific calls in a real-world environment gave clearer, concrete examples of appropriate calls for routing to the mental health crisis line.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData and Measures\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;\u003c/strong\u003eThis study seeks to quantify the impact of the embedded counselor on the frequency of law enforcement responses to suicide and mental health calls using publicly available call for service data. The Mesa, AZ data came from the city\u0026rsquo;s open data portal (City of Mesa, 2024b). The data contained all calls for service dispatched to MPD officers through its Computer Aided Dispatch system. The data included a unique event number, call date, time, duration, priority level, case type, and call source. Officer-initiated call sources (responses to something an officer saw in the field) were filtered out, as they could not have been impacted by the embedded crisis line counselor.\u003c/p\u003e\n\u003cp\u003eMesa\u0026rsquo;s data were compared to a synthetic control unit, which drew from several donor sites that were reasonably comparable to the test site (Abadie et al., 2015). Researchers searched for donor site data that consisted of police responses to mental health incidents initiated by a 911 call, had publicly available 911 data, and contained at least one call type relating to suicide or mental health. We identified donor sites by searching public websites from every city in the United States whose population was greater than 100,000 people (n=369) to mimic Mesa\u0026rsquo;s urban environment. We performed Google searches using \u0026ldquo;[city name] open data\u0026rdquo; and \u0026ldquo;[city name] calls for service\u0026rdquo; from October 2024 \u0026ndash; March 2025. We used \u0026ldquo;Computer Aided Dispatch\u0026rdquo;, \u0026ldquo;call for service\u0026rdquo;, or \u0026ldquo;police incident\u0026rdquo; data; \u0026ldquo;police stop\u0026rdquo; or \u0026ldquo;crime\u0026rdquo; data were excluded, since they did not contain 911-initiated mental health calls. Donor sites also could not have implemented a 911 embedded counselor program during the study period. To check for treatment contamination, we asked ChatGPT-5 to search for press releases that described embedded clinician programs at 911 centers at each of the cities; we verified its findings with manual Google searches. Table 1 lists donor sites (n=26), their states, populations, and rates of 911 calls for suicide, mental health, and all calls per quarter per 100,000 people.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e[insert table 1 around here]\u003c/p\u003e\n\u003cp\u003eWe counted Mesa\u0026rsquo;s \u0026ldquo;Suicide\u0026rdquo; and \u0026ldquo;Behavioral Health\u0026rdquo; initial call types as the outcome variables[1]. The trends of Mesa\u0026rsquo;s \u0026ldquo;Suicide\u0026rdquo; calls were compared against a synthetic control group of similar call types (e.g. suicidal, suicide attempt, etc.). Mesa\u0026rsquo;s \u0026ldquo;Behavioral Health\u0026rdquo; calls were compared with a synthetic control group of similar call types (e.g. mental patient, mentally disturbed person, etc.)[2]. Among the 26 donor sites, 20 included suicide-related categories, and 17 included mental health-related categories. Calls that were resolved over the phone were excluded in all but one site that did not have relevant data filters (Charlotte). The number of law enforcement calls for service for each type were aggregated by quarter per site to capture seasonal variation. It is recommended that there is low volatility in the outcome variable when running a synthetic control analysis (Wiltshire, 2025), and quarterly rates were less volatile than monthly rates. The suicide call rate in a given city and quarter was captured using this equation: \u003cem\u003eSuicideCallRate\u003csub\u003eqct\u0026thinsp;\u003c/sub\u003e=\u0026thinsp;((SuicideCalls\u003csub\u003eqct\u003c/sub\u003e)\u003csub\u003e/\u003c/sub\u003e(Pop\u003csub\u003eyct\u003c/sub\u003e\u003c/em\u003e))*(100,000), with \u003cem\u003eSuicideCalls\u003csub\u003eqct\u003c/sub\u003e\u003c/em\u003e representing the number of calls dispatched to law enforcement in a given quarter and \u003cem\u003ePop\u003c/em\u003e\u003csub\u003eyct\u003c/sub\u003e indicating the size of a city\u0026rsquo;s population in each year from US census data (United States Census Bureau, 2024). The suicide call rate equaled the total number of suicide calls in each city that were dispatched to law enforcement per 100,000 people each quarter. The same formula was applied to mental health calls: \u003cem\u003eMentalHealthCallRate\u003csub\u003eqct\u0026thinsp;\u003c/sub\u003e=\u0026thinsp;((MentalHealthCalls\u003csub\u003eqct\u003c/sub\u003e)\u003csub\u003e/\u003c/sub\u003e(Pop\u003csub\u003eyct\u003c/sub\u003e\u003c/em\u003e))*(100,000).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAnalytic Strategy\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;\u003c/strong\u003eTo examine the effects of the embedded crisis line counselor, we treated Mesa as a quantitative case study and employed quasi-experimental synthetic controls (Abadie et al., 2015; Bartos et al., 2020; Bartos \u0026amp; Kubrin, 2018; Gottlieb, Epperson, et al., 2024; Gottlieb, Harper, et al., 2024). The synthetic control method extends the difference-in-difference framework, which assumes trends across treatment and control groups continue in parallel if not for the intervention. It is often difficult to find a reasonable comparison site with similar pre-period trends (Bartos \u0026amp; Kubrin, 2018). The synthetic control method draws weights from a pool of potential donor sites to mimic the treatment group\u0026rsquo;s pre-period trends (time-invariant, non-negative, and add to one). We used all prior values of the outcome variable as covariates to optimize pretreatment fit between Mesa and its synthetic control (Bartos et al., 2020; Gottlieb, Epperson, et al., 2024), as prior outcome values would be more predictive of future values than any other covariate (Ferman et al., 2020).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe pretreatment periods should begin as early as possible while retaining a sufficient number of donor sites. There is no established minimum of donor sites for a synthetic control analysis, though having too few donor sites risks a type 2 error. For the suicide call rate analysis, 20 donor sites had pretreatment data starting in 2018q1. For the mental health call rate analysis, 17 donor sites had pretreatment data starting in 2018q3, when Mesa began to track mental health-related calls. Both pretreatment periods ended after 2021q2, the last quarter that was majority unaffected by the embedded counselor.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIt is often difficult to isolate the effect of one intervention from those of other policy interventions at the study site. The initial crisis line transfer policies in 2019q4 and 2020q3 occurred before the embedded counselor intervention in 2021q2, and their effects would likely have made the most impact before the post-treatment period began in 2021q3. Any outcome values affected by the initial crisis line transfer policies before embedding the crisis line counselor would have been included in the synthetic control weighting. It is possible there were lagged effects of the initial crisis line transfer policies that contaminated the outcome values after the crisis line counselor was embedded. However, a lagged year-and-a-half impact of the initial crisis line transfer policies would be less plausibly responsible for a drop in police responses than the new impact of embedding a crisis line counselor in the 911 center. The posttreatment started in 2021q3 and continued through 2023q2 to capture at least two years of post-intervention data; extending beyond two years of post-intervention data would invite concerns that other confounding influences may affect the police response rates. The post-intervention trends of the treatment group were compared against those of the synthetic controls.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp; We used Stata 19.5\u0026rsquo;s \u003cem\u003esynth\u003c/em\u003e command to employ the synthetic control method (Abadie et al., 2015). We first described the pre-intervention goodness of fit using the root mean squared prediction error (RMSPE) term and compared the post- and pre-intervention RMSPEs. If the post-intervention RMSPE was substantially larger, the comparison suggested that the effects of Mesa\u0026rsquo;s embedded counselor was substantively important (Bartos \u0026amp; Kubrin, 2018). To assess magnitude, we compared the call rate differences between Mesa and \u0026lsquo;synthetic Mesa\u0026rsquo; in the pre-intervention period with the differences the post-intervention period. Determining statistical significance when using the synthetic control method is not like traditional regression models (Abadie et al., 2015). To assess whether the observed effects happened by chance, placebo in-space estimates (i.e. creating new synthetic control groups for each of the donor sites) test whether the effect in Mesa was larger than almost all of the other jurisdictions (Abadie et al., 2015; Bartos \u0026amp; Kubrin, 2018; Galiani \u0026amp; Quistorff, 2017). If it was, it is unlikely that the observed effects were due to chance (Abadie et al., 2015; Galiani \u0026amp; Quistorff, 2017). We used Stata 18\u0026rsquo;s synth-runner command, which provided a standardized p-value for each post-intervention quarter (Galiani \u0026amp; Quistorff, 2017).\u003c/p\u003e\n\u003cp\u003eTo estimate the law enforcement hours saved, we calculated the average duration of on-scene time for suicide and mental health calls in Mesa and multiplied it by two officers. We then multiplied the officer hours by the synthetic control coefficients and adjusted the rate per 100,000 to Mesa\u0026rsquo;s population average=510,680.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e[1] In late 2020, Mesa showed an unusual change in trends among its mental health and suicide 911 call data. Mental health-related calls peaked for a five-month period, and suicide-related calls dipped in the same five-month period by similar proportions. The change occurred directly after the August, 2020 MPD policy change that expanded the categorization of mental health calls, which may have caused confusion among 911 call takers. To adjust for the suspected data issue, we assumed that the excess mental health call counts were suicide call counts and imputed accordingly. Call trends of the raw data and the authors\u0026rsquo; imputation strategy are discussed in the supplemental appendix, as are robustness checks that remove the months in question and analyze the raw counts.\u003c/p\u003e\n\u003cdiv id=\"ftn2\"\u003e\n \u003cp\u003e[2] The term \u0026lsquo;behavioral health\u0026rsquo; encompasses issues of both mental health and substance use. Substance use-related 911 calls are often medical emergencies that are often not appropriate for routing to crisis lines. Mesa\u0026rsquo;s data used the term \u0026ldquo;behavioral health\u0026rdquo;. The term \u0026lsquo;mental health\u0026rsquo; was more common among the comparison ECC sites and was chosen here for conceptual relevance.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eSuicide call rates\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFirst, we report the synthetic control analysis on police responses to suicide calls. The weights that comprised the synthetic control unit were: Everett=.213, New Orleans=.037, Norman=.036, Seattle=.334, Tacoma=.38, where \u0026lsquo;synthetic Mesa\u0026rsquo; was drawn from 21.3% of Everett\u0026rsquo;s suicide call rates, 3.7% of New Orleans\u0026rsquo; suicide call rates, etc.\u0026nbsp;Figure 1 shows the suicide call rate trends in Mesa (solid black line) and synthetic Mesa (dashed black line) pre- and post-intervention. Pre-intervention, the absolute value difference between Mesa and synthetic Mesa remained within 18 calls per 100,000 people. We treat post-treatment differences between Mesa and synthetic Mesa as the effect of embedding a counselor at the ECC. If the differences between Mesa and synthetic Mesa were larger than those of nearly all the donor cities, it is unlikely that the differences were due to chance.\u003c/p\u003e\n\u003cp\u003e[insert figure 1 around here]\u003c/p\u003e\n\u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Figure 1 shows Mesa\u0026rsquo;s suicide call rate was visibly lower than synthetic Mesa after the embedded counselor intervention. Mesa\u0026rsquo;s call rate increased relative to synthetic Mesa in 2021q3 by 13.43 calls, then decreased in 2021q4 to -22.27 calls, decreased in 2022q1 to -35.29 calls, decreased in 2022q2 to -36.58 calls, decreased in 2022q3 to -43.07 calls, decreased in 2022q4 to -47.96 calls, increased in 2023q1 to -44.75 calls, and increased in 2023q2 to -35.07 calls. Table 2 quantifies the synthetic control analysis. Mesa\u0026rsquo;s difference was greater than all but one of the placebo in-space estimates in 2021q4 (p=.05), all but three placebo estimates in 2022q1 (p=.15) all but two of the placebo estimates in 2022q2 (p=.10), all but one placebo estimate in 2022q3 (p=.05), all placebo estimates in 2022q4 (p\u0026lt;.01), all but two placebo estimates in 2023q1 (p=.10) and all but four placebo estimates in 2023q2 (p=.2).\u003c/p\u003e\n\u003cp\u003e[insert table 2 around here]\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMental health call rates\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNext, we report the synthetic control analysis on police responses to mental health calls. Table 3 shows the composition of the synthetic control unit from cities with mental health call rates most like Mesa in the pre-treatment period. The weights were: Chandler=.178, Chula Vista=.778, New York=.001, Portland=.043. Pretreatment fit was poor. As such, we are less confident in speculating whether post-treatment differences are a result of the embedded crisis line counselor at 911. Figure 2 shows the mental health call rate trends in Mesa (solid black line) and synthetic Mesa (dashed black line) pre- and post-intervention.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e[insert figure 2 around here]\u003c/p\u003e\n\u003cp\u003eFigure 2 shows Mesa\u0026rsquo;s mental health call rate fluctuated above and below synthetic Mesa\u0026rsquo;s before and after the embedded counselor intervention. Given the poor pretreatment fit, and the continued fluctuation after the intervention, we cannot say whether the embedded counselor had any effect on the mental health call rate. Table 3 quantifies the results of the synthetic control analysis.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;[insert table 3 around here]\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eBefore the embedded crisis counselor at 911, the mental health call rate difference between Mesa and synthetic Mesa ranged between -6.61 to 8.87 calls per 100,000 people. After the intervention, Mesa\u0026rsquo;s mental health call rate fluctuated between -9.09 calls lower to 4.84 calls higher than synthetic Mesa\u0026rsquo;s. The post-intervention differences between Mesa and synthetic Mesa remained firmly within the normal distribution of placebo estimates; the standardized p-values never dropped below p=.59. Because of the poor pretreatment fit, we are not confident that the synthetic control analysis represents a meaningful estimate of the causal association. Because there were no obvious changes in the Mesa\u0026rsquo;s mental health call rate\u0026rsquo;s trajectory after the embedded counselor intervention, we doubt there was substantial impact.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePolice hours saved\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe translated the synthetic control results from the suicide rate analysis into approximations of police hours saved. MPD averaged 60.09 minutes of on-scene time for suicide calls from 2021-2024. After embedding a crisis line counselor at its ECC, Mesa\u0026rsquo;s 22.27\u0026ndash;47.96 lower call rate per 100,000 people translated to 228.1\u0026ndash;490.6 police hours saved per quarter. We did not estimate police hours saved for mental health call rates because they did not appear to significantly decrease.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study assessed the impact of embedding a crisis line counselor at a 911 center on rates of law enforcement responses to suicide and mental health calls in Mesa, AZ using a synthetic control approach. To our knowledge, this is the first study empirically testing a 911 to crisis line call routing intervention. After embedding a crisis line counselor at 911, Mesa\u0026rsquo;s suicide call rate was lower than synthetic Mesa\u0026rsquo;s by 22.27\u0026ndash;47.96 calls per quarter per 100,000 people. Placebo estimates suggested Mesa\u0026rsquo;s difference was greater than all placebo sites in one post-intervention quarter (p\u0026lt;.01), all but one placebo site in two post-intervention quarters (p=.05), and all but two placebo sites in two post-intervention quarters (p=.10). We did not find evidence that the mental health call rate decreased. These results suggest that embedding a crisis line counselor at an ECC can reduce police responses to suicide calls, avoiding potentially problematic incidents of police escalation.\u003c/p\u003e \u003cp\u003eRouting calls \u0026lsquo;upstream\u0026rsquo; could reduce need for other \u0026lsquo;downstream\u0026rsquo; resources such as emergency departments (Bakko et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), which often present time and financial burdens for consumers, police, and health systems (Harada et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Hickey et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; B. R. Nordstrom, \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). Emergency department visits are expensive for consumers and can often result in long \u0026lsquo;ED boarding\u0026rsquo; delays while waiting for psychiatric inpatient placement (K. Nordstrom et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), tying up time and space for other health patients. Crisis lines primarily address low-risk over-the-phone suicidal ideation, but decreasing police presence may also prevent some incidents from dangerous escalations, particularly among people of color (Saleh et al., \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). People and family members with lived experience of mental health crises prefer models without police, as police presence can escalate the situation (Pope et al., \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). The reduction of police hours in this study shows a potential benefit for both people in mental health distress (through reduced contact hours with police during a crisis), and for law enforcement agencies (through reduced labor associated with mental health crises). Law enforcement agencies operate a majority of ECCs (Choi et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), and would likely be involved in the approval of a program involving the deflection of 911 calls to mental health crisis resources. Seeing the quantifiable reduction of police hours could serve as an incentive for law enforcement administrators to explore embedded 911 counselor or similar mental heath deflection programs.\u003c/p\u003e \u003cp\u003eRouting 911 calls to mental health crisis lines could also provide benefits for the person in distress. Mental health crisis lines can de-escalate the caller, activate mobile crisis, and connect people to ongoing mental health resources (Gould et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), but people may not know about their local crisis line and call 911 instead (M. Balfour \u0026amp; Minkoff, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Enabling transfers from 911 to crisis lines and other mental health resources could improve care continuity and long-term mental health. Future studies could track outcomes of calls routed from 911 to crisis lines to quantify the downstream impact. Transfer policies should be mindful of racial bias among call-takers; if crisis line call-takers are more likely to transfer Black callers back to 911 (Goldman et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), 911 call-takers may also be less likely to transfer Black callers to mental health crisis lines.\u003c/p\u003e \u003cp\u003eThese findings suggest that crisis line specialists provide important assistance with 911 to crisis line call routing. A Solari representative learning the ECC\u0026rsquo;s culture, policies, procedures, practices, and norms likely built personal- and agency-level trust. 911 professionals feel personal liability to callers expressing suicidal ideation (Hedden-Clayton et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Pew, 2021), and establishing stronger connections between 911 and 988 professionals can assuage fears that the calls would be mishandled or unanswered. The crisis line specialist assisted with translating ECC call routing policies into actual calls, rather than hypothetical training scenarios, which may be important for sustained routing practices. ECCs should welcome collaboration with crisis lines and arrange protocols for meaningful involvement of interdisciplinary crisis line staff. Future qualitative studies could examine mechanisms key to successful routing interventions.\u003c/p\u003e \u003cp\u003eWe did not find a meaningful change in Mesa\u0026rsquo;s mental health-related call rate. It is possible that suicidal calls are often first-person calls, making them more appropriate to route to a crisis line and speak directly to crisis line specialists. Third-party bystanders may not have a relationship with the person in distress and may not wish to get involved themselves, preferring law enforcement come out to address the situation. Additionally, a person in mental health distress may experience anosognosia, or unawareness of their mental health condition. If someone experiences a delusion of intruders, for example, they may not have any interest in talking to a crisis line; they may refuse the transfer and only request a police presence. The broader concept of \u0026lsquo;mental health\u0026rsquo; calls may also make it difficult to articulate appropriate calls to route. Suicidality has clearer, more distinct concepts (i.e. ideation, threats, means), whereas 911 professionals may be less confident in identifying mental health concepts that vary situationally or require clinical judgment (i.e. speaking \u0026lsquo;quickly\u0026rsquo;, \u0026lsquo;erratic\u0026rsquo; behaviors, \u0026lsquo;disorganized\u0026rsquo; thought patterns, etc.). 911 professionals often have limited information, and they may not have the benefit of knowing the accuracy of the caller\u0026rsquo;s claims. When faced with uncertainty, 911 professionals often default to dispatching law enforcement (Hedden-Clayton et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Pew, 2021). Studies that analyze the content of the suicide and mental health calls could assess the routing potential with more specificity.\u003c/p\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eLimitations\u003c/h2\u003e \u003cp\u003eLimited and inconsistent language across ECCs made it difficult to parse meaningful differences in publicly available data. We used data labeled as \u0026ldquo;Computer Aided Dispatch\u0026rdquo; (CAD), \u0026ldquo;calls for service\u0026rdquo;, and \u0026ldquo;police incidents\u0026rdquo;. The variance between the terms suggested potential differences in the types of data being compared: \u0026ldquo;police dispatch events\u0026rdquo; in Mesa may represent different activities than \u0026ldquo;calls for service\u0026rdquo; in New Orleans. ECC call file names were often accompanied by brief descriptions, though not infrequently the terms like \u0026ldquo;calls for service\u0026rdquo; and \u0026ldquo;police incidents\u0026rdquo; were used interchangeably, making it difficult to parse whether the differences were meaningful or semantic. There was variation in how the ECCs coded suicide and mental health calls, as mental health calls are often aggregated with larger categories like \u0026ldquo;check welfare\u0026rdquo; or \u0026ldquo;public disturbance\u0026rdquo;. Suicide calls are a more distinct concept, making the call type more comparable across ECCs.\u003c/p\u003e \u003cp\u003eFurther analysis could filter calls by priority level, as calls with high priority levels involving weapons or active violence are not likely to be appropriate for routing to 988. However, priority levels were inconsistently defined across the databases, and applying consistent criteria in filtering out high priority calls was not possible. Moreover, given that Mesa\u0026rsquo;s interventions involved the re-classification of calls previously deemed to require police responses, some of the routable calls in the donor sites were likely to have been classified as high priority incidents. As such, all suicide- and mental health-related calls were included, regardless of priority level, recognizing some of the targeted call counts still required police responses. Cities with particularly low suicide call rates may instead be tracking suicide attempts, rather than callers with suicidal ideation, and would be inappropriate comparisons to Mesa. Unfortunately, we do not know whether the cities track suicide calls differently or whether the cities had lower rates of 911 suicidal ideation calls. When dropping four sites with low suicide call rates (\u0026lt;\u0026thinsp;10/100,000 people/quarter), the effect size did not change and remained larger than all placebo sites in one post-treatment quarter (p\u0026lt;.01), and larger than all but one placebo in two post-treatment quarters (p=.0625). The supplemental appendix describes robustness checks that dropped individual cities to determine outsized influence and examined non-suicidal call trends to assess possible confounders.\u003c/p\u003e \u003cp\u003eFinally, ECC governance structure would likely affect whether ECCs know about 988 and route calls to it. Police-operated ECCs would be less likely to route calls than independently operated ECCs (Choi et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). We did not have a large enough subsample of police-operated ECCs to compare Mesa to ECCs with similar police-operated governance structure; in this study, the police-operated Mesa ECC was compared to a mixed group of independent- and police-operated ECCs, and the suicide call rates were still associated with significant relative declines. Police responses may differ across state-level policies or regional-level cultural differences, but only one southwestern city other than Mesa (Chandler) did not have a similar embedded crisis line counselor intervention and availed downloadable law enforcement response data that contained enough information to perform the analyses. We suggest cautious interpretations of the mental health call analysis given its poor pretreatment fit.\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study used synthetic controls to assess the impact of embedding a crisis line counselor at a 911 center on law enforcement responses to mental health and suicide calls. Embedding a crisis line counselor at 911 was associated with 22.27\u0026ndash;47.96 fewer police responses to suicide calls per quarter per 100,000 people compared to a synthetic control unit, saving Mesa an estimated 228.1\u0026ndash;490.6 police hours per quarter. The mental health call response rate did not decrease. Fewer law enforcement responses to suicide calls could decrease acute service utilization and escalated police incidents. ECCs should consider embedding crisis line staff to assist with identification of suicidal calls for appropriate transfer. Deflecting suicide-related 911 calls to mental health crisis lines could invite the mental health crisis system to assume some of the responsibilities associated with the expanded police footprint.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData availability statement:\u0026nbsp;\u003c/strong\u003eThe authors created a dataset of cities with publicly available 911 data and corresponding URLs, available upon request. Processed data of monthly mental health- and suicide-related 911 call frequencies by city are also available upon request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements:\u003c/strong\u003e [blinded for peer review]\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding Statement:\u003c/strong\u003e The authors received no financial support for the research, authorship, and/or publication of this article.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval statement:\u0026nbsp;\u003c/strong\u003eThis project used publicly available data and was determined to be non-human subjects research by the authors’ university IRB [blinded for peer review].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of interest statement:\u003c/strong\u003e The authors have no conflicts of interest to declare.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions:\u0026nbsp;\u003c/strong\u003e[blinded for peer review]\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003e\u003cem\u003e988 Lifeline\u003c/em\u003e. (2022). 988 Lifeline. https://988lifeline.org/\u003c/li\u003e\n \u003cli\u003eAbadie, A., Diamond, A., \u0026amp; Hainmueller, J. (2015). Comparative Politics and the Synthetic Control Method. \u003cem\u003eAmerican Journal of Political Science\u003c/em\u003e, \u003cem\u003e59\u003c/em\u003e(2), 495\u0026ndash;510. https://doi.org/10.1111/ajps.12116\u003c/li\u003e\n \u003cli\u003eAldridge, O. (2025, January 9). \u003cem\u003eBehind the scenes of mental health calls at Austin\u0026rsquo;s 911 call center\u003c/em\u003e. KUT Radio, Austin\u0026rsquo;s NPR Station. https://www.kut.org/health/2025-01-09/austin-tx-911-mental-health-crisis-calls-24-hours-podcast\u003c/li\u003e\n \u003cli\u003eAlexander, C., \u0026amp; Zealberg, J. J. (1999). Mobile crisis: Moving emergency psychiatry out of the hospital setting. \u003cem\u003eNew Directions for Mental Health Services\u003c/em\u003e, \u003cem\u003e1999\u003c/em\u003e(82), 93\u0026ndash;99. https://doi.org/10.1002/yd.23319998212\u003c/li\u003e\n \u003cli\u003eBakko, M., Swanson, L., Zettner, C., Kok, K., Fukuzawa, H., \u0026amp; Kubiak, S. (2025). A Comparison of Behavioral Health Crisis Response Models in Meeting Behavioral Health Goals and Improving Criminal Legal Diversion. \u003cem\u003eCommunity Mental Health Journal\u003c/em\u003e. https://doi.org/10.1007/s10597-024-01447-4\u003c/li\u003e\n \u003cli\u003eBalfour, M. E., Hahn Stephenson, A., Delany-Brumsey, A., Winsky, J., \u0026amp; Goldman, M. L. (2022). Cops, Clinicians, or Both? Collaborative Approaches to Responding to Behavioral Health Emergencies. \u003cem\u003ePsychiatric Services\u003c/em\u003e, \u003cem\u003e73\u003c/em\u003e(6), 658\u0026ndash;669. https://doi.org/10.1176/appi.ps.202000721\u003c/li\u003e\n \u003cli\u003eBalfour, M., \u0026amp; Minkoff, K. (2022). A Look at the 988 Soft Launch. \u003cem\u003ePsychiatric Times\u003c/em\u003e. https://www.psychiatrictimes.com/view/a-look-at-the-988-soft-launch\u003c/li\u003e\n \u003cli\u003eBartos, B. J., \u0026amp; Kubrin, C. E. (2018). Can We Downsize Our Prisons and Jails Without Compromising Public Safety? \u003cem\u003eCriminology \u0026amp; Public Policy\u003c/em\u003e, \u003cem\u003e17\u003c/em\u003e(3), 693\u0026ndash;715. https://doi.org/https://doi.org/10.1111/1745-9133.12378\u003c/li\u003e\n \u003cli\u003eBartos, B. J., McCleary, R., Mazerolle, L., \u0026amp; Luengen, K. (2020). Controlling Gun Violence: Assessing the Impact of Australia\u0026rsquo;s Gun Buyback Program Using a Synthetic Control Group Experiment. \u003cem\u003ePrevention Science\u003c/em\u003e, \u003cem\u003e21\u003c/em\u003e(1), 131\u0026ndash;136. https://doi.org/10.1007/s11121-019-01064-8\u003c/li\u003e\n \u003cli\u003eBeck, J., Reuland, M., \u0026amp; Pope, L. (2020, November). \u003cem\u003eCASE STUDY: Robust Crisis Care and Diverting 911 Calls to Crisis Lines\u003c/em\u003e. Vera Institute of Justice. https://www.vera.org/behavioral-health-crisis-alternatives/robust-crisis-care-and-diverting-911-calls-to-crisis-lines\u003c/li\u003e\n \u003cli\u003eBorum, R., Swanson, J., Swartz, M., \u0026amp; Hiday, V. (1997). Substance Abuse, Violent Behavior, and Police Encounters among Persons with Severe Mental Disorder. \u003cem\u003eJournal of Contemporary Criminal Justice\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(3), 236\u0026ndash;250. https://doi.org/10.1177/1043986297013003004\u003c/li\u003e\n \u003cli\u003eCenter for American Progress. (2020, October 28). \u003cem\u003eThe Community Responder Model\u003c/em\u003e. https://www.americanprogress.org/article/community-responder-model/\u003c/li\u003e\n \u003cli\u003eCenter for Police Research and Policy. (2021). \u003cem\u003eAssessing the Impact of Mobile Crisis Teams: A Review of Research\u003c/em\u003e. University of Cincinnati. https://www.informedpoliceresponses.com/_files/ugd/313296_8d01cdc7187a489893197f2d07300ee6.pdf\u003c/li\u003e\n \u003cli\u003eChoi, H., Lerner, J., \u0026amp; Neusteter, R. (2024). \u003cem\u003eUnderstanding 911 Governance: Examining Factors Influencing and Related to the Governance of our Nation\u0026rsquo;s Emergency Communications Centers\u003c/em\u003e (Transform911). University of Chicago Health Lab. https://bpb-us-w2.wpmucdn.com/voices.uchicago.edu/dist/e/2911/files/2024/11/Making-911-Independent-and-Equal_Brief_FINAL.pdf\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eCIT International\u0026mdash;CIT Support Training for 911\u003c/em\u003e. (n.d.). Retrieved August 28, 2025, from https://www.citinternational.org/CITST911\u003c/li\u003e\n \u003cli\u003eCity of Mesa. (2024a). \u003cem\u003eData Portal | Mesa Data Hub\u003c/em\u003e. https://data.mesaaz.gov/\u003c/li\u003e\n \u003cli\u003eCity of Mesa. (2024b, December 5). \u003cem\u003ePolice Dispatch Events 2021-present\u003c/em\u003e. Mesa Data Hub. https://data.mesaaz.gov/Police/Police-Dispatch-Events-2021-present/izhu-764k/about_data\u003c/li\u003e\n \u003cli\u003eCojean, S., Combalbert, N., \u0026amp; Taillandier-Schmitt, A. (2020). Psychological and sociological factors influencing police officers\u0026rsquo; decisions to use force: A systematic literature review. \u003cem\u003eInternational Journal of Law and Psychiatry\u003c/em\u003e, \u003cem\u003e70\u003c/em\u003e, 101569. https://doi.org/10.1016/j.ijlp.2020.101569\u003c/li\u003e\n \u003cli\u003eComartin, E. B., Swanson, L., \u0026amp; Kubiak, S. (2019). Mental Health Crisis Location and Police Transportation Decisions: The Impact of Crisis Intervention Team Training on Crisis Center Utilization. \u003cem\u003eJournal of Contemporary Criminal Justice\u003c/em\u003e, \u003cem\u003e35\u003c/em\u003e(2), 241\u0026ndash;260. https://doi.org/10.1177/1043986219836595\u003c/li\u003e\n \u003cli\u003eCompton, M. T., Bakeman, R., Broussard, B., Hankerson-Dyson, D., Husbands, L., Krishan, S., Stewart-Hutto, T., D\u0026rsquo;Orio, B. M., Oliva, J. R., Thompson, N. J., \u0026amp; Watson, A. C. (2014). The Police-Based Crisis Intervention Team (CIT) Model: II. Effects on Level of Force and Resolution, Referral, and Arrest. \u003cem\u003ePsychiatric Services\u003c/em\u003e, \u003cem\u003e65\u003c/em\u003e(4), 523\u0026ndash;529. https://doi.org/10.1176/appi.ps.201300108\u003c/li\u003e\n \u003cli\u003eCompton, M. T., Pope, L. G., \u0026amp; Watson, A. C. (2024). Community-Based Mental Health Crisis Response: An Overview of Models and Workforce Implications. \u003cem\u003ePsychiatric Clinics of North America\u003c/em\u003e. https://doi.org/10.1016/j.psc.2024.04.004\u003c/li\u003e\n \u003cli\u003eCurrier, G. W., Fisher, S. G., \u0026amp; Caine, E. D. (2010). Mobile Crisis Team Intervention to Enhance Linkage of Discharged Suicidal Emergency Department Patients to Outpatient Psychiatric Services: A Randomized Controlled Trial. \u003cem\u003eAcademic Emergency Medicine\u003c/em\u003e, \u003cem\u003e17\u003c/em\u003e(1), 36\u0026ndash;43. https://doi.org/10.1111/j.1553-2712.2009.00619.x\u003c/li\u003e\n \u003cli\u003eDavis, J., Norris, S., Schmitt, J., Shem-Tov, Y., \u0026amp; Strickland, C. (2025). \u003cem\u003eMobile Crisis Response Teams Support Better Policing: Evidence from CAHOOTS\u003c/em\u003e (Working Paper No. 33761). National Bureau of Economic Research. https://doi.org/10.3386/w33761\u003c/li\u003e\n \u003cli\u003eDee, T. S., \u0026amp; Pyne, J. (2022). A community response approach to mental health and substance abuse crises reduced crime. \u003cem\u003eScience Advances\u003c/em\u003e, \u003cem\u003e8\u003c/em\u003e(23), eabm2106. https://doi.org/10.1126/sciadv.abm2106\u003c/li\u003e\n \u003cli\u003eDyches, H., Biegel, D. E., Johnsen, J. A., Guo, S., \u0026amp; Min, M. O. (2002). The Impact of Mobile Crisis Services on the Use of Community-Based Mental Health Services. \u003cem\u003eResearch on Social Work Practice\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(6), 731\u0026ndash;751. https://doi.org/10.1177/104973102237470\u003c/li\u003e\n \u003cli\u003eFederal Communications Commission. (2024, October 31). \u003cem\u003e911 Master PSAP Registry\u003c/em\u003e. https://www.fcc.gov/general/9-1-1-master-psap-registry\u003c/li\u003e\n \u003cli\u003eFerman, B., Pinto, C., \u0026amp; Possebom, V. (2020). Cherry Picking with Synthetic Controls. \u003cem\u003eJournal of Policy Analysis and Management\u003c/em\u003e, \u003cem\u003e39\u003c/em\u003e(2), 510\u0026ndash;532. https://doi.org/10.1002/pam.22206\u003c/li\u003e\n \u003cli\u003eFuller, D., Lamb, R., Biasotti, M., \u0026amp; Snook, J. (2015). \u003cem\u003eOverlooked in the Undercounted: The role of mental illness in fatal law enforcement encounters\u003c/em\u003e. Treatment Advocacy Center. https://www.treatmentadvocacycenter.org/storage/documents/overlooked-in-the-undercounted.pdf\u003c/li\u003e\n \u003cli\u003eGaliani, S., \u0026amp; Quistorff, B. (2017). The Synth_Runner Package: Utilities to Automate Synthetic Control Estimation Using Synth. \u003cem\u003eThe Stata Journal\u003c/em\u003e, \u003cem\u003e17\u003c/em\u003e(4), 834\u0026ndash;849. https://doi.org/10.1177/1536867X1801700404\u003c/li\u003e\n \u003cli\u003eGasior, M. (2020, December 22). \u003cem\u003ePolicing the mentally ill\u003c/em\u003e. PowerDMS. https://www.powerdms.com/policy-learning-center/policing-the-mentally-ill\u003c/li\u003e\n \u003cli\u003eGeller, J. L., Fisher, W. H., \u0026amp; McDermeit, M. (1995). A national survey of mobile crisis services and their evaluation. \u003cem\u003ePsychiatric Services (Washington, D.C.)\u003c/em\u003e, \u003cem\u003e46\u003c/em\u003e(9), 893\u0026ndash;897. https://doi.org/10.1176/ps.46.9.893\u003c/li\u003e\n \u003cli\u003eGillig, P., Dumaine, M., \u0026amp; Hillard, J. (1990). Whom Do Mobile Crisis Services Serve? \u003cem\u003eHospital \u0026amp; Community Psychiatry\u003c/em\u003e, \u003cem\u003e41\u003c/em\u003e(7), 804\u0026ndash;805. https://doi.org/10.1176/ps.41.7.804\u003c/li\u003e\n \u003cli\u003eGillooly, J. W., \u0026amp; Thacher, D. (2024). How the Public Became the Caller: The Emergence of Reactive Policing, 1880\u0026ndash;1970. \u003cem\u003eLaw \u0026amp; Social Inquiry\u003c/em\u003e, 1\u0026ndash;33. https://doi.org/10.1017/lsi.2024.9\u003c/li\u003e\n \u003cli\u003eGoldman, M., Elser, A., Yeh, H.-H., McDaniel, M., Ma, L., Ahmedani, B. K., \u0026amp; Foster, A. A. (2025). Demographic and Clinical Characteristics of Mental Health Crisis Line Callers Who Were Transferred to 911. \u003cem\u003ePsychiatric Services\u003c/em\u003e, \u003cem\u003e76\u003c/em\u003e(3), 282\u0026ndash;285. https://doi.org/10.1176/appi.ps.20240050\u003c/li\u003e\n \u003cli\u003eGottlieb, A., Epperson, M., Giugno, L., \u0026amp; Jung, H.-M. (2024). What happens to imprisonment rates when a progressive prosecutor is elected: Quasi-experimental evidence from Cook County. \u003cem\u003ePunishment \u0026amp; Society\u003c/em\u003e, 14624745241288185. https://doi.org/10.1177/14624745241288185\u003c/li\u003e\n \u003cli\u003eGottlieb, A., Harper, T., \u0026amp; Jung, H.-M. (2024). Does racial impact statement reform reduce Black\u0026ndash;White disparities in imprisonment: Mixed methods evidence from Minnesota. \u003cem\u003eLaw \u0026amp; Policy\u003c/em\u003e, \u003cem\u003en/a\u003c/em\u003e(n/a). https://doi.org/10.1111/lapo.12252\u003c/li\u003e\n \u003cli\u003eGould, M. S., Lake, A. M., Galfalvy, H., Kleinman, M., Munfakh, J. L., Wright, J., \u0026amp; McKeon, R. (2018). Follow-up with Callers to the National Suicide Prevention Lifeline: Evaluation of Callers\u0026rsquo; Perceptions of Care. \u003cem\u003eSuicide and Life-Threatening Behavior\u003c/em\u003e, \u003cem\u003e48\u003c/em\u003e(1), 75\u0026ndash;86. https://doi.org/10.1111/sltb.12339\u003c/li\u003e\n \u003cli\u003eGould, M. S., Lake, A. M., Munfakh, J. L., Galfalvy, H., Kleinman, M., Williams, C., Glass, A., \u0026amp; McKeon, R. (2016). Helping Callers to the National Suicide Prevention Lifeline Who Are at Imminent Risk of Suicide: Evaluation of Caller Risk Profiles and Interventions Implemented. \u003cem\u003eSuicide and Life-Threatening Behavior\u003c/em\u003e, \u003cem\u003e46\u003c/em\u003e(2), 172\u0026ndash;190. https://doi.org/10.1111/sltb.12182\u003c/li\u003e\n \u003cli\u003eGould, M. S., Lake, A. M., Port, M. S., Kleinman, M., Hoyte‐Badu, A. M., Rodriguez, C. L., Chowdhury, S. J., Galfalvy, H., \u0026amp; Goldstein, A. (2025). National Suicide Prevention Lifeline (Now 988 Suicide and Crisis Lifeline): Evaluation of Crisis Call Outcomes for Suicidal Callers. \u003cem\u003eSuicide \u0026amp; Life-Threatening Behavior\u003c/em\u003e, \u003cem\u003e55\u003c/em\u003e(3), e70020. https://doi.org/10.1111/sltb.70020\u003c/li\u003e\n \u003cli\u003eGuo, S., Biegel, D. E., Johnsen, J. A., \u0026amp; Dyches, H. (2001). Assessing the Impact of Community-Based Mobile Crisis Services on Preventing Hospitalization. \u003cem\u003ePsychiatric Services\u003c/em\u003e, \u003cem\u003e52\u003c/em\u003e(2), 223\u0026ndash;228. https://doi.org/10.1176/appi.ps.52.2.223\u003c/li\u003e\n \u003cli\u003eHarada, M. Y., Lara-Mill\u0026aacute;n, A., \u0026amp; Chalwell, L. E. (2021). Policed Patients: How the Presence of Law Enforcement in the Emergency Department Impacts Medical Care. \u003cem\u003eAnnals of Emergency Medicine\u003c/em\u003e, \u003cem\u003e78\u003c/em\u003e(6), 738\u0026ndash;748. https://doi.org/10.1016/j.annemergmed.2021.04.039\u003c/li\u003e\n \u003cli\u003eHedden-Clayton, B., Carl, A., Swanson, L., Comartin, E., \u0026amp; Kubiak, S. (2025). Deflecting people experiencing a mental health crisis from the criminal-legal system at dispatch: Findings from calls-for-service reports and statewide survey data. \u003cem\u003eEvaluation and Program Planning\u003c/em\u003e, 102727. https://doi.org/10.1016/j.evalprogplan.2025.102727\u003c/li\u003e\n \u003cli\u003eHickey, K., Walther, C., King, J., Layne, H., Besler, C., Andrzejewski, V., \u0026amp; Gijada, J. (2020). Connecting Law Enforcement and Emergency Department Providers to Improve Access to Mental Health Services. \u003cem\u003eJournal of Psychosocial Nursing and Mental Health Services\u003c/em\u003e, \u003cem\u003e58\u003c/em\u003e(8), 24\u0026ndash;30. https://doi.org/https://doi.org/10.3928/02793695-20200624-02\u003c/li\u003e\n \u003cli\u003eHugo, M., Smout, M., \u0026amp; Bannister, J. (2002). A comparison in hospitalization rates between a community-based mobile emergency service and a hospital-based emergency service. \u003cem\u003eAustralian and New Zealand Journal of Psychiatry\u003c/em\u003e, \u003cem\u003e36\u003c/em\u003e(4), 504\u0026ndash;508. https://doi.org/10.1046/j.1440-1614.2002.01042.x\u003c/li\u003e\n \u003cli\u003eKim, S., \u0026amp; Kim, H. (2017). Determinants of the use of community-based mental health services after mobile crisis team services: An empirical approach using the Cox proportional hazard model. \u003cem\u003eJournal of Community Psychology\u003c/em\u003e, \u003cem\u003e45\u003c/em\u003e(7), 877\u0026ndash;887. https://doi.org/10.1002/jcop.21899\u003c/li\u003e\n \u003cli\u003eKoziarski, J., Ferguson, L., \u0026amp; Huey, L. (2022). Shedding Light on the Dark Figure of Police Mental Health Calls for Service. \u003cem\u003ePolicing: A Journal of Policy and Practice\u003c/em\u003e, \u003cem\u003e16\u003c/em\u003e(4), 696\u0026ndash;706. https://doi.org/10.1093/police/paac006\u003c/li\u003e\n \u003cli\u003eMesa Police Department. (2019, October 16). \u003cem\u003eOverdose or Suicide Attempt, COM 13.4\u003c/em\u003e.\u003c/li\u003e\n \u003cli\u003eMesa Police Department. (2020, August 17). \u003cem\u003ePersons Affected by Mental Illness or in Crisis\u003c/em\u003e.\u003c/li\u003e\n \u003cli\u003eMorabito, M. S., Socia, K., Wik, A., \u0026amp; Fisher, W. H. (2017). The nature and extent of police use of force in encounters with people with behavioral health disorders. \u003cem\u003eInternational Journal of Law and Psychiatry\u003c/em\u003e, \u003cem\u003e50\u003c/em\u003e, 31\u0026ndash;37. https://doi.org/10.1016/j.ijlp.2016.10.001\u003c/li\u003e\n \u003cli\u003eMunetz, M. R., \u0026amp; Griffin, P. A. (2006). Use of the Sequential Intercept Model as an Approach to Decriminalization of People With Serious Mental Illness. \u003cem\u003ePSYCHIATRIC SERVICES\u003c/em\u003e, \u003cem\u003e57\u003c/em\u003e(4).\u003c/li\u003e\n \u003cli\u003eNational Emergency Number Association. (2021). \u003cem\u003e9-1-1 Statistics\u003c/em\u003e. https://www.nena.org/page/911statistics\u003c/li\u003e\n \u003cli\u003eNENA. (2021). \u003cem\u003e9-1-1 Statistics\u003c/em\u003e. https://www.nena.org/page/911Statistics\u003c/li\u003e\n \u003cli\u003eNordstrom, B. R. (2011). The effects of neighborhood characteristics on rearrest after participation in drug teatment court: A re-analysis of the Baltimore city drug treatment court experiment. \u003cem\u003eDissertations Available from ProQuest\u003c/em\u003e, 1\u0026ndash;123.\u003c/li\u003e\n \u003cli\u003eNordstrom, K., Berlin, J. S., Nash, S. S., Shah, S. B., Schmelzer, N. A., \u0026amp; Worley, L. L. M. (2019). Boarding of Mentally Ill Patients in Emergency Departments: American Psychiatric Association Resource Document. \u003cem\u003eWestern Journal of Emergency Medicine\u003c/em\u003e, \u003cem\u003e20\u003c/em\u003e(5), 690\u0026ndash;695. https://doi.org/10.5811/westjem.2019.6.42422\u003c/li\u003e\n \u003cli\u003eO\u0026rsquo;Connell, K. L., Jacobson, S. V., Ton, A. T., \u0026amp; Law, K. C. (2022). Association between race and socioeconomic factors and suicide-related 911 call rate. \u003cem\u003eSocial Science \u0026amp; Medicine\u003c/em\u003e, \u003cem\u003e306\u003c/em\u003e, 115106. https://doi.org/10.1016/j.socscimed.2022.115106\u003c/li\u003e\n \u003cli\u003ePatton, C. L., \u0026amp; Fremouw, W. J. (2016). Examining \u0026ldquo;suicide by cop\u0026rdquo;: A critical review of the literature. \u003cem\u003eAggression and Violent Behavior\u003c/em\u003e, \u003cem\u003e27\u003c/em\u003e, 107\u0026ndash;120. https://doi.org/10.1016/j.avb.2016.03.003\u003c/li\u003e\n \u003cli\u003ePew. (2021, October 26). \u003cem\u003eNew Research Suggests 911 Call Centers Lack Resources to Handle Behavioral Health Crises\u003c/em\u003e. https://pew.org/30LhARu\u003c/li\u003e\n \u003cli\u003ePope, L. G., Patel, A., Fu, E., Zingman, M., Warnock, A., Ellis, S., Ashekun, O., Watson, A., Wood, J., \u0026amp; Compton, M. T. (2023). Crisis Response Model Preferences of Mental Health Care Clients With Prior Misdemeanor Arrests and of Their Family and Friends. \u003cem\u003ePsychiatric Services\u003c/em\u003e, \u003cem\u003e74\u003c/em\u003e(11), 1163\u0026ndash;1170. https://doi.org/10.1176/appi.ps.20220363\u003c/li\u003e\n \u003cli\u003ePurtle, J., McSorley, A.-M. M., Adera, A. L., \u0026amp; Lindsey, M. A. (2023). Use, Potential Use, and Awareness of the 988 Suicide and Crisis Lifeline by Level of Psychological Distress. \u003cem\u003eJAMA Network Open\u003c/em\u003e, \u003cem\u003e6\u003c/em\u003e(10), e2341383. https://doi.org/10.1001/jamanetworkopen.2023.41383\u003c/li\u003e\n \u003cli\u003eSaleh, A. Z., Appelbaum, P. S., Liu, X., Scott Stroup, T., \u0026amp; Wall, M. (2018). Deaths of people with mental illness during interactions with law enforcement. \u003cem\u003eInternational Journal of Law and Psychiatry\u003c/em\u003e, \u003cem\u003e58\u003c/em\u003e, 110\u0026ndash;116. https://doi.org/10.1016/j.ijlp.2018.03.003\u003c/li\u003e\n \u003cli\u003eSAMHSA. (2023, June 15). \u003cem\u003e988 Frequently Asked Questions\u003c/em\u003e. https://www.samhsa.gov/mental-health/988/faqs\u003c/li\u003e\n \u003cli\u003eSAMHSA. (2025). \u003cem\u003e2025 National Guidelines for a Behavioral Health Coordinated System of Crisis Care\u003c/em\u003e (Nos. PEP24-01\u0026ndash;037). https://library.samhsa.gov/sites/default/files/national-guidelines-crisis-care-pep24-01-037.pdf\u003c/li\u003e\n \u003cli\u003eSaunders, H. (2025, July 14). Demand for 988 Continues to Grow at Third Anniversary. \u003cem\u003eKFF\u003c/em\u003e. https://www.kff.org/mental-health/demand-for-988-continues-to-grow-at-third-anniversary/\u003c/li\u003e\n \u003cli\u003eSolari. (2025, April). \u003cem\u003eCrisis Response Network Performance Dashboard\u003c/em\u003e. Tableau Public. https://public.tableau.com/app/profile/crisis.network/viz/AZ600StatewideDashboard/AZ600StatewideDashboard\u003c/li\u003e\n \u003cli\u003eSwanson, L., Nelson, V., Comartin, E. B., Kubiak, S., Putans, L., Hambrick, N., Ray, B., Tillander, L., Washington, A., Butkiewicz, R., \u0026amp; Costello, M. (2022). Assessing County-Level Behavioral Health and Justice Systems with the Sequential Intercept Model Practices, Leadership, and Expertise Scorecard. \u003cem\u003eCommunity Mental Health Journal\u003c/em\u003e. https://doi.org/10.1007/s10597-022-01042-5\u003c/li\u003e\n \u003cli\u003eSwanson, L., Zettner, C., Watson, A., Hinojosa, M., Roddy, J., \u0026amp; Kubiak, S. (2025). Eleven‐Month Arrest Outcomes Among Three Crisis Response Models in Michigan. \u003cem\u003ePsychiatric Research and Clinical Practice\u003c/em\u003e, n/a-n/a. https://doi.org/10.1176/appi.prcp.20240100\u003c/li\u003e\n \u003cli\u003eTransform 911. (2022). \u003cem\u003eTransform 911: A Blueprint for Change\u003c/em\u003e. University of Chicago Health Lab. https://bpb-us-w2.wpmucdn.com/voices.uchicago.edu/dist/e/2911/files/2022/08/Transform911-Blueprint-for-Change.pdf\u003c/li\u003e\n \u003cli\u003eUeda, M., Heflin, C. M., Liu, Y., Yu, Q., \u0026amp; Ramanathan, S. (2026). Understanding Factors Associated with 911 and 988 Use in Mental Health Crises. \u003cem\u003eCommunity Mental Health Journal\u003c/em\u003e, \u003cem\u003e62\u003c/em\u003e(3), 527\u0026ndash;536. https://doi.org/10.1007/s10597-025-01545-x\u003c/li\u003e\n \u003cli\u003eUnited States Census Bureau. (2024). \u003cem\u003eCensus Bureau Data\u003c/em\u003e. Explore Census Data. https://data.census.gov/ U.S. Census Bureau. (2023). \u003cem\u003eMesa city, Arizona\u003c/em\u003e. https://www.census.gov/quickfacts/fact/table/mesacityarizona/PST045223\u003c/li\u003e\n \u003cli\u003eVakkalanka, J. P., Neuhaus, R. A., Harland, K. K., Clemsen, L., Himadi, E., \u0026amp; Lee, S. (2021). Mobile Crisis Outreach and Emergency Department Utilization: A Propensity Score-matched Analysis. \u003cem\u003eWestern Journal of Emergency Medicine\u003c/em\u003e, \u003cem\u003e22\u003c/em\u003e(5), 1086\u0026ndash;1094. https://doi.org/10.5811/westjem.2021.6.52276\u003c/li\u003e\n \u003cli\u003eWard-Ciesielski, E. F., \u0026amp; Rizvi, S. L. (2021). The potential iatrogenic effects of psychiatric hospitalization for suicidal behavior: A critical review and recommendations for research. \u003cem\u003eClinical Psychology: Science and Practice\u003c/em\u003e, \u003cem\u003e28\u003c/em\u003e(1), 60\u0026ndash;71. https://doi.org/10.1111/cpsp.12332\u003c/li\u003e\n \u003cli\u003eWaters, R. (2021). Enlisting Mental Health Workers, Not Cops, In Mobile Crisis Response. \u003cem\u003eHealth Affairs\u003c/em\u003e, \u003cem\u003e40\u003c/em\u003e(6), 864\u0026ndash;869. https://doi.org/10.1377/hlthaff.2021.00678\u003c/li\u003e\n \u003cli\u003eWatson, A. C., Compton, M. T., \u0026amp; Draine, J. N. (2017). The crisis intervention team (CIT) model: An evidence-based policing practice? \u003cem\u003eBehavioral Sciences \u0026amp; the Law\u003c/em\u003e, \u003cem\u003e35\u003c/em\u003e(5\u0026ndash;6), 431\u0026ndash;441. https://doi.org/10.1002/bsl.2304\u003c/li\u003e\n \u003cli\u003eWatson, A. C., Compton, M. T., \u0026amp; Pope, L. G. (2019). \u003cem\u003eCrisis Response Services for People with Mental Illnesses or Intellectual and Developmental Disabilities: A review of the literature on police-based and other first response models\u003c/em\u003e. Vera Institute of Justice. (https://www.vera.org/). https://www.vera.org/publications/crisis-response-services-for-people-with-mental-illnesses-or-intellectual-and-developmental-disabilities\u003c/li\u003e\n \u003cli\u003eWiltshire, J. C. (2025). allsynth: (Stacked) Synthetic Control Bias-Correction Utilities for Stata. \u003cem\u003eWorking Paper\u003c/em\u003e. https://static1.squarespace.com/static/5e0fdcef27e0945c43fab131/t/67aa6bac859390516fcf8f21/1739221941939/allsynth_Wiltshire.pdf\u003cstrong\u003e\u003cbr clear=\"all\"\u003e\u003c/strong\u003e\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003e\u003cstrong\u003eTable 1.\u003c/strong\u003e Characteristics of cities in the synthetic control analyses\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"623\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" rowspan=\"2\" valign=\"bottom\" style=\"width: 130px;\"\u003e\n \u003cp\u003eCity\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" rowspan=\"2\" valign=\"bottom\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eState\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" rowspan=\"2\" valign=\"bottom\" style=\"width: 101px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;Population \u0026nbsp;(mean 2018-2022)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" rowspan=\"2\" valign=\"bottom\" style=\"width: 87px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eECC governance\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" colspan=\"3\" style=\"width: 229px;\"\u003e\n \u003cp\u003eQuarterly average per 100k population\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003eAll 911 calls\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003eMental health\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003eSuicide\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eBoise City\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eID\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e233,662\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003ePolice\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e7,784\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e1260.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eBoston\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eMA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e673,948\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003ePolice\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e3,276\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e2.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eCary\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eNC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e173,971\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003eIndependent\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e671\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e6.9\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eChandler\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eAZ\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e271,218\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003ePolice\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e7,150\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e52.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e71.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eCharlotte\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eNC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e882,538\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003ePolice\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e2,493\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e14.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eChula Vista\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eCA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e274,343\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003ePolice\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e6,683\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e0.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e2.8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eCincinnati\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eOH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e306,777\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003eIndependent\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e29,510\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e625.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e-\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eCollege Station\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eTX\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e120,333\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003ePolice\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e20,695\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e1.1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eEverett\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eWA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e111,229\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003eIndependent\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e22,112\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e667.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e151.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eFort Lauderdale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eFL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e182,509\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003ePolice\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e22,946\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e252.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e81.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eHartford\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eCT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e120,869\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003eIndependent\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e43,932\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e400.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e87.2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eIrving\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eTX\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e249,326\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003ePolice\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e17,969\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e28.8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eLincoln\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eNE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e290,308\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003eIndependent\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e2,794\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e34.2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eLos Angeles\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eCA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e3,901,675\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003ePolice\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e9,940\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e75.3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003e\u003cem\u003eMesa (test case)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u003cem\u003eAZ\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e\u003cem\u003e510,680\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003e\u003cem\u003ePolice\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u003cem\u003e7,112\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u003cem\u003e17.2\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u003cem\u003e132.2\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eMiami\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eFL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e453,098\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003ePolice\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e16,807\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e245\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e11.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eNashville\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eTN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e677,047\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003eIndependent\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e23,044\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e29.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e75.9\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eNew Orleans\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eLA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e382,330\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003eIndependent\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e22,107\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e274.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e94.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eNew York\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eNY\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e8,453,041\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003ePolice\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e19,712\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e988.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eNorman\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eOK\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e127,106\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003ePolice\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e11,572\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e179.6\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003ePortland\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eOR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e647,354\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003eIndependent\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e9,462\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e200\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eSan Diego\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eCA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e1,399,116\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003ePolice\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e10,148\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e300\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eSan Francisco\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eCA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e850,494\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003eIndependent\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e19,328\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e387.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e108.9\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eSan Jose\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eCA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e1,002,486\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003ePolice\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e6,398\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e80\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eSeattle\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eWA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e743,473\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003eIndependent\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e4,363\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e110.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e90.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eSt. Petersburg\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eFL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e262,077\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003ePolice\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e18,247\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e410\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e119.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" style=\"width: 130px;\"\u003e\n \u003cp\u003eTacoma*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003eWA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e219,008\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003eIndependent\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e15,641\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e186.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 130px;\"\u003e\n \u003cp\u003e\u003cem\u003eMean\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e\u003cem\u003e871,112\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 87px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u003cem\u003e14,144\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u003cem\u003e350.1\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u003cem\u003e74.0\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 130px;\"\u003e\n \u003cp\u003e\u003cem\u003eStandard Deviation\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 101px;\"\u003e\n \u003cp\u003e\u003cem\u003e1,685,134\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 87px;\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u003cem\u003e9,855\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u003cem\u003e345.3\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u003cem\u003e58.5\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e*Tacoma launched an embedded clinician program in June, 2023; its effects would not impact this study period\u003cstrong\u003e\u003cbr clear=\"all\"\u003e\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eTable 2\u003c/strong\u003e. Synthetic control weights, pre-intervention fit, and post-intervention outcomes of suicide 911 call rates per 100,000 people. Weights: Everett=.213, New Orleans=.037, Norman=.036, Seattle=.334, Tacoma=.38\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"465\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" rowspan=\"2\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003eQuarter\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" colspan=\"3\" valign=\"bottom\" style=\"width: 314px;\"\u003e\n \u003cp\u003eSuicide call rate per 100,000 people\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" rowspan=\"2\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003eStandardized p-value\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003eMesa, AZ\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003eSynthetic Control\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003eDifference\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003e2018q1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003e158.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e157.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e1.46\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003e2018q2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003e165.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e160.41\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e4.96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003e2018q3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003e163.60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e160.46\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e3.14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003e2018q4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003e157.89\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e160.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e-2.85\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003e2019q1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003e144.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e147.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e-2.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003e2019q2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003e155.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e158.58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e-3.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003e2019q3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003e170.46\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e164.90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e5.56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003e2019q4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003e148.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e149.36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e-0.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003e2020q1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003e148.25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e154.14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e-5.89\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003e2020q2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003e134.61\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e133.57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e1.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003e2020q3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003e159.32\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e142.14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e17.19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003e2020q4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003e132.64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e139.42\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e-6.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003e2021q1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003e124.10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e126.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e-2.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003e2021q2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003e125.86\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e135.24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e-9.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003e2021q3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003e147.66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e134.23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e13.43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e0.20\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003e2021q4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003e110.15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e132.42\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e-22.27\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003e2022q1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003e93.12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e128.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e-35.29\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e0.15\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003e2022q2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003e97.41\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e133.99\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e-36.58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e0.10\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003e2022q3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003e97.61\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e140.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e-43.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003e2022q4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003e84.92\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e132.88\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e-47.96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e0.00\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003e2023q1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003e89.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e134.66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e-44.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e0.10\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 67px;\"\u003e\n \u003cp\u003e2023q2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 95px;\"\u003e\n \u003cp\u003e98.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 114px;\"\u003e\n \u003cp\u003e133.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e-35.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e0.20\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e\u003c/strong\u003e\u003cstrong\u003eTable 3.\u003c/strong\u003e Synthetic control weights, pre-intervention fit, and post-intervention outcomes of mental health call rates\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"422\" class=\"fr-table-selection-hover\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" rowspan=\"2\" valign=\"bottom\" style=\"width: 59px;\"\u003e\n \u003cp\u003eQuarter\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" colspan=\"3\" valign=\"bottom\" style=\"width: 271px;\"\u003e\n \u003cp\u003eMental health call rate per 100,000 people\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" rowspan=\"2\" valign=\"bottom\" style=\"width: 92px;\"\u003e\n \u003cp\u003eStandardized p-value\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 55px;\"\u003e\n \u003cp\u003eMesa, AZ\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003eSynthetic Control\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 110px;\"\u003e\n \u003cp\u003eDifference\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 59px;\"\u003e\n \u003cp\u003e2018q3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 55px;\"\u003e\n \u003cp\u003e21.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e22.36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 110px;\"\u003e\n \u003cp\u003e-1.29\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 92px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 59px;\"\u003e\n \u003cp\u003e2018q4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 55px;\"\u003e\n \u003cp\u003e24.61\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e22.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 110px;\"\u003e\n \u003cp\u003e2.17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 92px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 59px;\"\u003e\n \u003cp\u003e2019q1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 55px;\"\u003e\n \u003cp\u003e28.57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e19.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 110px;\"\u003e\n \u003cp\u003e8.87\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 92px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 59px;\"\u003e\n \u003cp\u003e2019q2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 55px;\"\u003e\n \u003cp\u003e24.71\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e22.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 110px;\"\u003e\n \u003cp\u003e2.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 92px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 59px;\"\u003e\n \u003cp\u003e2019q3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 55px;\"\u003e\n \u003cp\u003e19.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e20.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 110px;\"\u003e\n \u003cp\u003e-0.88\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 92px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 59px;\"\u003e\n \u003cp\u003e2019q4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 55px;\"\u003e\n \u003cp\u003e20.66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e17.58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 110px;\"\u003e\n \u003cp\u003e3.08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 92px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 59px;\"\u003e\n \u003cp\u003e2020q1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 55px;\"\u003e\n \u003cp\u003e16.21\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e17.41\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 110px;\"\u003e\n \u003cp\u003e-1.20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 92px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 59px;\"\u003e\n \u003cp\u003e2020q2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 55px;\"\u003e\n \u003cp\u003e14.83\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e19.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 110px;\"\u003e\n \u003cp\u003e-5.08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 92px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 59px;\"\u003e\n \u003cp\u003e2020q3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 55px;\"\u003e\n \u003cp\u003e18.58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e19.94\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 110px;\"\u003e\n \u003cp\u003e-1.36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 92px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 59px;\"\u003e\n \u003cp\u003e2020q4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 55px;\"\u003e\n \u003cp\u003e18.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e15.96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 110px;\"\u003e\n \u003cp\u003e2.43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 92px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 59px;\"\u003e\n \u003cp\u003e2021q1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 55px;\"\u003e\n \u003cp\u003e11.19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e17.80\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 110px;\"\u003e\n \u003cp\u003e-6.61\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 92px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 59px;\"\u003e\n \u003cp\u003e2021q2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 55px;\"\u003e\n \u003cp\u003e12.57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e18.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 110px;\"\u003e\n \u003cp\u003e-5.80\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 92px;\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 59px;\"\u003e\n \u003cp\u003e2021q3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 55px;\"\u003e\n \u003cp\u003e24.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e19.90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 110px;\"\u003e\n \u003cp\u003e4.84\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 92px;\"\u003e\n \u003cp\u003e0.65\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 59px;\"\u003e\n \u003cp\u003e2021q4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 55px;\"\u003e\n \u003cp\u003e16.30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e18.43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 110px;\"\u003e\n \u003cp\u003e-2.13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 92px;\"\u003e\n \u003cp\u003e0.71\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 59px;\"\u003e\n \u003cp\u003e2022q1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 55px;\"\u003e\n \u003cp\u003e13.86\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e16.29\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 110px;\"\u003e\n \u003cp\u003e-2.43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 92px;\"\u003e\n \u003cp\u003e0.59\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 59px;\"\u003e\n \u003cp\u003e2022q2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 55px;\"\u003e\n \u003cp\u003e12.49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e16.56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 110px;\"\u003e\n \u003cp\u003e-4.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 92px;\"\u003e\n \u003cp\u003e0.71\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 59px;\"\u003e\n \u003cp\u003e2022q3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 55px;\"\u003e\n \u003cp\u003e12.10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e18.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 110px;\"\u003e\n \u003cp\u003e-6.64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 92px;\"\u003e\n \u003cp\u003e0.71\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 59px;\"\u003e\n \u003cp\u003e2022q4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 55px;\"\u003e\n \u003cp\u003e11.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e16.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 110px;\"\u003e\n \u003cp\u003e-5.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 92px;\"\u003e\n \u003cp\u003e0.71\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 59px;\"\u003e\n \u003cp\u003e2023q1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 55px;\"\u003e\n \u003cp\u003e9.19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e18.28\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 110px;\"\u003e\n \u003cp\u003e-9.09\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 92px;\"\u003e\n \u003cp\u003e0.71\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 59px;\"\u003e\n \u003cp\u003e2023q2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 55px;\"\u003e\n \u003cp\u003e13.49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 105px;\"\u003e\n \u003cp\u003e18.13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 110px;\"\u003e\n \u003cp\u003e-4.65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd nowrap=\"\" valign=\"bottom\" style=\"width: 92px;\"\u003e\n \u003cp\u003e0.82\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"crisis line, 988, 911, embedded, diversion, suicide, crisis response","lastPublishedDoi":"10.21203/rs.3.rs-9226560/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9226560/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eLaw enforcement are the default responders to mental health- and suicide-related calls, and law enforcement can escalate mental health crises to hospitalizations, uses of force, and deaths. Mental health crisis lines, such as 988, are an alternative underutilized mental health crisis resource compared to estimated annual mental health-related 911 calls. This study assesses whether embedding a counselor at a 911 center could reduce mental health- and suicide-related police responses by transferring 911 calls to a mental health crisis line. A quasi-experimental synthetic control analysis assessed police response rates after embedding a crisis line counselor at a 911 emergency communications center in Mesa, AZ in 2021. Authors used publicly available 911 data from 27 cities to create donor pools for synthetic control analyses of police responses to suicide and mental health calls. Embedding a 911 crisis line counselor was associated with 22.27\u0026ndash;47.96 fewer police responses to suicide calls per quarter per 100,000 people, relative to a synthetic control unit. The lower suicide call rate translated to an estimated 228.1\u0026ndash;490.6 police hours reduced per quarter in Mesa. There was no evidence to suggest that Mesa\u0026rsquo;s police response rate to mental health calls decreased. The embedded presence of a crisis line counselor at a 911 center may facilitate interdisciplinary trust that enables call routing to crisis lines. Emergency communication centers should consider embedding crisis line staff to divert calls from police, limiting the expansion of the police footprint.\u003c/p\u003e","manuscriptTitle":"Deflecting suicide and mental health calls from law enforcement with an embedded crisis line specialist at a 911 center: A synthetic control analysis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-31 14:47:22","doi":"10.21203/rs.3.rs-9226560/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"7bb44423-f8bb-4108-af74-72aacee5f5bf","owner":[],"postedDate":"March 31st, 2026","published":true,"recentEditorialEvents":[{"type":"reviewerAgreed","content":"284990111029102654155487823959960344441","date":"2026-05-04T23:58:43+00:00","index":29,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-03-31T14:47:22+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-31 14:47:22","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9226560","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9226560","identity":"rs-9226560","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}