Association of Arizona’s Implementation of a Behavioral Health Crisis Response System With Suicide Hospitalizations
Abstract
Objective:
In July 2022, the 988 Suicide and Crisis Lifeline went live. The Lifeline is part of larger federal and state efforts to build comprehensive behavioral health crisis response systems that include mobile crisis units and crisis diversion and stabilization centers. Comprehensive response systems are anticipated to reduce hospitalizations for suicide and other behavioral health crises; however, research testing this assumption has been limited. The authors used Arizona—a state known for its comprehensive crisis system—to determine the association between state implementation of a comprehensive behavioral health crisis response system and suicide-related hospitalizations.
Methods:
A comparative interrupted time-series (CITS) design was used to compare changes in suicide-related hospitalizations after the 2015 implementation of Arizona’s crisis response system (N=215,063). Data were from the 2010–2019 Healthcare Cost and Utilization Project (HCUP) State Inpatient Databases (SID). Nevada (N=84,091 hospitalizations) was used as a comparison state because it is a western state that had not yet implemented a comprehensive crisis system and had available HCUP SID data. The CITS model included controls for time-varying differences in state demographic composition.
Results:
From 2010 to 2014 to 2019, annual suicide-related hospitalizations in Arizona increased from 122.0 to 324.2 to 584.5, respectively, per 100,000 people, and in Nevada, hospitalizations increased from 94.7 to 263.2 to 595.5, respectively, per 100,000 people. Arizona’s crisis response system was associated with a significant relative decrease in the quarterly trend of 2.57 suicide-related hospitalizations per 100,000 people (p=0.033).
Conclusions:
More research is needed to understand how the implementation of a comprehensive crisis response system may affect suicide-related hospitalizations.
HIGHLIGHTS
•
The implementation of Arizona’s comprehensive behavioral health crisis system in 2015 was associated with a decrease of 2.57 suicide-related hospitalizations per 100,000 people per quarter relative to suicide-related hospitalizations in Nevada.
•
Despite the approximately 2%–3% relative decrease in suicide-related hospitalizations over 4 years, Arizona’s suicide-related hospitalizations increased from 324.2 per 100,000 people per year in 2014 to 584.5 in 2019.
•
More research is needed now that the 988 Suicide and Crisis Lifeline has been implemented to determine how comprehensive crisis systems affect hospitalizations for behavioral health crises.
In July 2022, the 988 Suicide and Crisis Lifeline was rolled out as a universal telephone number for individuals having a behavioral health crisis to call or text to receive brief crisis counseling. The National Suicide Hotline Designation Act (https://www.congress.gov/bill/116th-congress/senate-bill/2661), signed into law in 2020, authorized 988 as a new three-digit number to replace the 10-digit National Suicide Prevention Lifeline telephone number. The rollout of 988 was part of a larger national and state effort to improve the ability of the nation’s behavioral health system to respond to individuals in crisis. In 2020, the Substance Abuse and Mental Health Services Administration (SAMHSA) released the National Guidelines for Behavioral Health Crisis Care: Best Practice Toolkit, in which three essential elements of a behavioral health crisis response system are described: regional 24/7 clinically staffed crisis call centers that use best practices in suicide prevention; crisis mobile teams able to reach anyone, anywhere in a timely manner; and crisis receiving and stabilization facilities that provide short-term (<24-hour) observation and crisis stabilization services in a nonhospital environment (1). These three components are often summarized as “someone to call,” “someone to come,” and a “safe place to go.” The SAMHSA guidelines reflect the National Action Alliance for Suicide Prevention’s framework, also referred to as the “crisis now model” or “Arizona model” (2, 3).
A key goal of 988, and of comprehensive crisis response systems in general, is to reduce suicide attempts and deaths. A second goal is to shift the locus of care for suicide events, and other behavioral health crises, away from general hospitals to settings that offer faster, and more patient-centered, access to specialty mental health treatment (4). Policy analysts developed a calculator for crisis resource needs to support states and municipalities in their rollout of crisis response systems; the calculator assumes that, once implemented, comprehensive crisis response systems will reduce hospital inpatient stays from 68% of individuals in crisis to 20% (5, 6).
Although the assumption that crisis response systems will result in fewer inpatient hospital stays is supported by some studies of individual components of crisis systems, we know of no empirical studies evaluating the effect of a comprehensive crisis response system on hospitalization rates in the United States. Recent reviews of mental health crisis hotlines in the United States identified four studies that found that callers reported reduced distress and suicidality after hotline engagement, but the reviews identified no studies on whether hotlines reduced psychiatric hospitalizations (7, 8). Some research has found that use of mobile crisis teams is associated with fewer psychiatric emergency department (ED) visits and hospitalizations (9–13). A review of 12 studies across six countries (Australia, Belgium, Canada, the Netherlands, the United Kingdom, and the United States) found that hospital-based crisis stabilization units were associated with a reduction in ED visits and inpatient stays (14). Research on the effect of non–hospital-based crisis stabilization units is lacking (15).
In this study, we evaluated whether the implementation of Arizona’s behavioral health crisis response system was associated with a decrease in suicide-related hospitalizations, compared with Nevada, a state that had not yet implemented a comprehensive crisis response system. We focused on Arizona because it is the only state that had a comprehensive state crisis response system, as defined by SAMHSA’s National Guidelines for Behavioral Health Crisis Care, as of 2020, and it is considered a model for other states to replicate (16, 17). Arizona’s crisis response system was implemented in 2015 as a result of a 2014 lawsuit settlement with the Arizona Department of Health Services that required the department to develop a crisis response system that included a crisis hotline, mobile crisis teams, and crisis stabilization settings that provide short-term crisis stabilization services (up to 72 hours) to successfully resolve the crisis, returning individuals to the community instead of transitioning them to a higher level of care (i.e., a general hospital setting) (18). Arizona’s crisis response system was made available to all individuals, regardless of insurance type (19). To fund its system, Arizona expanded Medicaid billing codes for crisis services and blended federal grant, state, and county monies. Arizona disseminated the funding to regional behavioral health authorities, who contracted with providers to deliver crisis services (18). Arizona’s rollout of its crisis response system was reported to have resulted in expansion of alternatives to hospital inpatient settings. For example, in 2015 the number of crisis beds in Maricopa County increased from <50 beds to about 150 (20).
Methods
A challenge in evaluating comprehensive crisis intervention response systems is that one cannot use an experimental design to determine the system’s effectiveness. To evaluate the effect of the comprehensive crisis response system implemented in Arizona, we used a version of the difference-in-differences (DiD) statistical approach known as comparative interrupted time-series (CITS) analysis. DiD methods are commonly used to evaluate complex policy interventions, including state- and country-level policies that are anticipated to affect suicide outcomes. For example, Lewitzka et al. (21) evaluated the effectiveness of national suicide prevention programs by using a DiD framework to compare four countries that implemented them with four countries that did not. We followed best practices for conducting DiD analyses by identifying a counterfactual state that did not implement a crisis response system, testing the common trends assumption that no time-varying differences existed between the intervention and control populations through visual inspection and statistical testing, and adjusting for any time-varying characteristics of the intervention and control populations (22).
Comparison State
To identify states comparable to Arizona, we required that states meet the following criteria: located in the western United States, had Healthcare Cost and Utilization Project (HCUP) State Inpatient Databases (SID) data over the 2010–2019 period, and had not yet implemented a comprehensive crisis response system. Six western states had HCUP SID data available for 2010–2019. Of these six, Nevada was the only western state that had not fully implemented any component of the crisis response system as reflected in 2020 Community Mental Health Services Block Grant applications and other public information (23–25).
Data
Data on suicide-related hospital discharges (referred to here as “hospitalizations”) came from the HCUP SID. SID is an all-payer database of all community hospital discharges in each state (26). Community hospitals include academic medical centers and specialty hospitals, such as obstetrics, gynecology, otolaryngology, short-term rehabilitation, orthopedic, and pediatric hospitals. Community hospitals exclude U.S. Department of Veterans Affairs, Department of Defense, and Indian Health Service hospitals, as well as long-term hospitals, psychiatric hospitals, alcohol or drug treatment facilities, and hospital units within institutions such as prisons. For each discharge, SID captures principal and other diagnoses; procedure codes; and demographic information, such as age, race, ethnicity, primary payer, and other information. Each state submits their data to the federal government. We used SID data from Arizona and Nevada for 2010–2019.
Identification of Suicide-Related Hospitalizations
To identify suicide-related hospitalizations, we used ICD-9-CM and ICD-10-CM diagnosis codes. In October 2015, the United States transitioned from using ICD-9-CM codes to ICD-10-CM codes. We used ICD-9-CM codes for data collected before October 2015 and ICD-10-CM codes for data collected thereafter.
We used the Agency for Healthcare Research and Quality’s (AHRQ’s) Clinical Classifications Software to identify suicide-related diagnosis codes. The Clinical Classifications Software groups diagnosis codes into clinically meaningful categories. The diagnosis codes used are listed in section A of the online supplement to this article. The software identified suicide discharges as those that included a suicide-related diagnosis code in any position (primary, secondary, tertiary, etc.). To further ensure that suicide or a behavioral health condition was the primary reason for the hospitalization, we limited the sample to hospitalizations with a principal diagnosis code related to mental, behavioral, and neurodevelopmental disorders and injury, poisoning, or any suicide-related ICD codes (e.g., suicidal ideation). This approach resulted in the removal of hospitalizations for which the principal diagnosis was osteoporosis or glaucoma, for example, leading to the elimination of 6% of the discharges with any suicide-related diagnosis.
The outcome of interest was the rate of suicide-related hospitalizations per 100,000 people in the state per quarter (i.e., 3-month period). We used American Community Survey data to determine the state population size. We summed the number of suicide-related hospitalizations by state and quarter and then divided it by the state population to calculate the state rate of suicide-related hospitalizations per quarter.
Statistical Approach
We estimated the following CITS model:
where Yit is the suicide-related hospitalization rate per 100,000 people for state i (Arizona or Nevada) in quarter t; Tt is the time in quarters; Zi is 1 for Arizona and 0 for Nevada; Tt×Zi is an interaction term between time and state; Xt is a dummy variable indicating the crisis response system implementation period (0 before 2016 and 1 beginning in 2016); At is the time in quarters since the crisis response system was implemented, starting with zero in the first quarter of 2016 and incrementing by one each quarter after that; Zi×Xt is an interaction between the state and the start of the crisis response system, and Zi×At is an interaction between the state and the quarters since the start of the crisis response system. PopChar is a vector of state demographic characteristics. As is appropriate for models with fixed state effects and time dummy variables, we included only state characteristic variables that had significant time and state variation: the percentage of the population that was female, the percentage of non-Hispanic Black individuals in the state, and the percentage of “other” race, where “other” encompassed Asian, Native Hawaiian, two or more races, or race “other.” εit is an error term.
The coefficient β6 is interpreted as the difference in the shift in the level of suicide-related hospitalization rates between Arizona and Nevada in the first quarter after the crisis response system was implemented (first-quarter effect). The coefficient β7 indicates the difference between Arizona and Nevada in the trend of suicide-related hospitalization rates after initiation of the crisis response system implementation in Arizona, compared with the preimplementation trend (i.e., slope effect) (27). β8 indicates a vector of state population characteristics from the American Community Survey for each year in the 2010–2019 period.
To avoid unreliable data for the period when hospitals were undergoing the changeover from ICD-9 to ICD-10, we omitted the 2015 data in the CITS analysis. This omission follows the approach taken by AHRQ in their evaluation of suicide-related hospitalization trends with HCUP data (28). We also tested for autocorrelation using a Cumby-Huizinga general test and observed statistically significant results for one lag; we therefore used an AR(1) model with Newey-West standard errors (29). This study was conducted from September 2021 to August 2022 and did not require institutional review board approval.
Results
Table 1 describes the characteristics of the Arizona and Nevada populations based on 2010–2019 American Community Survey data. Because of the large population sizes, most differences between Arizona and Nevada were statistically significant. However, most differences were small in terms of percentage point difference. Arizona had a greater percentage of residents who identified as non-Hispanic Native American or non-Hispanic White and fewer who identified as non-Hispanic Black or non-Hispanic “other” race.
| Characteristic | Arizona (M±SD) | Nevada (M±SD) | Mean difference | t | pb |
|---|---|---|---|---|---|
| Age in years | | | | | |
| <18 | 25.3±.4 | 24.4±.4 | .9 | 9.3 | <.001 |
| 18–64 | 59.4±1.0 | 62.0±1.1 | −2.6 | −11.3 | <.001 |
| ≥65 | 15.3±1.2 | 13.5±1.2 | 1.8 | 6.9 | <.001 |
| Female | 50.8±.0 | 50.0±.1 | .8 | 41.2 | <.001 |
| Hispanic | 30.3±.6 | 27.6±.7 | 2.7 | 18.5 | <.001 |
| Non-Hispanic White | 56.7±1.1 | 52.1±1.7 | 4.6 | 14.6 | <.001 |
| Non-Hispanic Black | 4.1±.1 | 8.3±.4 | −4.2 | −70.1 | <.001 |
| Non-Hispanic Native American | 4.4±.0 | 1.1±.1 | 3.3 | 244.9 | <.001 |
| Non-Hispanic other race | 12.9±.8 | 21.0±2.2 | −8.1 | −22.3 | <.001 |
| Unemployed | 8.6±2.0 | 10.1±2.6 | −1.5 | −2.9 | .004 |
| Povertyc | 24.4±2.8 | 21.6±2.6 | 2.8 | 4.7 | <.001 |
a
Source: authors’ analysis of American Community Survey data, 2010–2019. M±SD values are percentages unless indicated otherwise.
b
Statistical significance of differences between means was assessed with two-sided t tests (df=78 for all comparisons).
c
Defined as income <138% of the federal poverty line.
Table 2 describes the characteristics of patients with suicide-related hospital discharge. The average age was 35.1 years in Arizona and 38.1 years in Nevada; 51.1% and 47.8% of discharged patients were female in Arizona and Nevada, respectively. Compared with Nevada, Arizona had a lower percentage of discharges who identified as non-Hispanic Black (6.7% vs. 13.8%) and a greater percentage of those who identified as Native American (4.8% vs. 1.7%) or Hispanic (17.3% vs. 8.2%). Arizona had fewer Medicare discharges (14.4% vs. 23.0%) and more Medicaid (41.3% vs. 37.2%) and privately insured (28.0% vs. 21.3%) discharges than Nevada.
| Characteristic | Arizona (M±SD; N=240,625) | Nevada (M±SD; N=96,577) |
|---|---|---|
| Age in years | 35.1±16.5 | 38.1±17.6 |
| Female | 51.1±.5 | 47.8±.5 |
| Race | | |
| Non-Hispanic White | 69.7±.5 | 64.3±.5 |
| Non-Hispanic Black | 6.7±.3 | 13.8±.3 |
| Non-Hispanic Asian or Pacific Islander | 1.4±.1 | 1.9±.1 |
| Non-Hispanic Native American | 4.8±.2 | 1.7±.1 |
| Non-Hispanic other | .1±.0 | 10.1±.3 |
| Hispanic | 17.3±.4 | 8.2±.3 |
| Primary expected payer | | |
| Medicare | 14.4±.4 | 23.0±.4 |
| Medicaid | 41.3±.5 | 37.2±.5 |
| Private insurance | 28.0±.5 | 21.3±.4 |
| Uninsured | 16.2±.4 | 18.5±.4 |
| Patient location | | |
| Micropolitan statistical area | 2.8±.2 | NA |
| Metropolitan statistical area | 96.5±.2 | NA |
a
Source: authors’ analyses of the Healthcare Cost and Utilization Project state inpatient data set, 2010–2019. Values are percentages unless indicated otherwise. NA, not applicable.
Figure 1 displays the rate of suicide-related hospitalizations per 100,000 people per quarter from 2010 to 2019 in Arizona and Nevada. Arizona’s annual suicide-related hospital discharge rate per 100,000 people increased from 122.0 in 2010 to 324.2 in 2014 to 584.5 in 2019 (Table 3). Nevada’s annual suicide-related hospital discharge rate increased from 94.7 in 2010 to 263.2 in 2014 to 595.5 in 2019. Arizona and Nevada hospitalization trends were similar in the preimplementation period (before 2015) (Figure 1). Arizona’s suicide-related hospitalization rate was higher than Nevada’s in most quarters before 2015, then dipped below Nevada in some quarters after 2015.
FIGURE 1. Suicide-related hospital discharges per quarter in Arizona and Nevada, 2010–2019a
aData for 2015 were omitted because hospitals were undergoing a change from ICD-9 to ICD-10 codes.
| | N of hospitalizations | Rate per 100,000 people | ||
|---|---|---|---|---|
| Year | Arizona | Nevada | Arizona | Nevada |
| 2010 | 7,824 | 2,560 | 122.0 | 94.7 |
| 2011 | 13,181 | 3,027 | 203.3 | 111.2 |
| 2012 | 15,787 | 6,126 | 240.9 | 222.0 |
| 2013 | 18,205 | 6,990 | 274.7 | 250.5 |
| 2014 | 21,823 | 7,472 | 324.2 | 263.2 |
| 2015 | 25,562 | 12,486 | 374.4 | 431.9 |
| 2016 | 24,991 | 8,078 | 360.6 | 274.8 |
| 2017 | 31,907 | 14,046 | 454.8 | 468.5 |
| 2018 | 38,800 | 17,449 | 541.0 | 575.0 |
| 2019 | 42,545 | 18,343 | 584.5 | 595.5 |
a
Source: authors’ analyses of the Healthcare Cost and Utilization Project state inpatient data set, 2010–2019.
Table 4 summarizes the coefficient estimates of a model assessing the impact of Arizona’s crisis response system implementation on quarterly suicide-related hospitalizations per 100,000 people, compared with Nevada. Of particular importance are results for β6 (first-quarter) and β7 (trend) effects. The model failed to reject the parallel trends test, indicating that before the crisis response system intervention, Arizona and Nevada had similar trends in suicide-related discharges. Arizona’s implementation of the crisis response system was not associated with an immediate first-quarter change in suicide-related hospitalizations at the start of 2016 relative to Nevada. However, it was associated with a small but statistically significant (p=0.033) decrease in trend (i.e., trend effect) of −2.57 hospitalizations per 100,000 people per quarter compared with Nevada.
| Variable | Coefficient | 95% CI | SE | p |
|---|---|---|---|---|
| T (time in quarters) | 3.56 | 2.83 to 4.29 | 9.8 | <.001 |
| Z (Arizona=1; Nevada=0) | −57.7 | −323.4 to 207.9 | −.43 | .665 |
| T×Z interaction | −.07 | −1.01 to .87 | −.14 | .889 |
| X (crisis system implementation period indicator) | −.53 | −21.7 to 20.6 | −.05 | .960 |
| A (time in quarters after start of crisis implementation period) | 4.36 | 1.24 to 7.47 | 2.79 | .007 |
| Z×X interaction (first-quarter effect) | −7.86 | −27.3 to 11.6 | 9.74 | .423 |
| Z×A interaction (trend effect) | −2.57 | −4.92 to −.22 | 1.18 | .033 |
| % female | −68.3 | −135.7 to −.97 | 33.7 | .047 |
| % non-Hispanic Black | −39.1 | −97.9 to 19.7 | 29.4 | .189 |
| % non-Hispanic other race | 3.91 | −3.50 to 11.3 | 3.71 | .295 |
a
Data for 2015 were omitted from the analysis because hospitals were undergoing a change from ICD-9 to ICD-10 codes.
Discussion
Arizona’s comprehensive crisis response system is often cited as a model for other states to emulate. Some budget impact models estimate that if a state implemented a comprehensive crisis response system such as Arizona’s, the state’s behavioral health hospitalizations could drop to a third of its baseline levels. Using Nevada, a state with a noncomprehensive crisis response system, as a comparator, we found that Arizona’s comprehensive crisis response system was associated with a relatively small quarterly decrease of 2.57 suicide-related hospitalizations per 100,000 people. For context, annual suicide-related hospitalizations in Arizona were 584.5 per 100,000 people in 2019, meaning that its crisis response system had an effect of about 1.8% fewer suicide hospitalizations in 2019 (2.57×4/584.5×100=1.76%).
We can only conjecture as to why suicide-related hospitalizations did not decline more substantially after Arizona implemented its comprehensive crisis response system. One reason may be that although Arizona expanded crisis diversion centers and other alternatives to hospitalizations in community settings, the expansion may not have been large enough to meet demand for crisis services. Counting crisis diversion centers is challenging because no standard definition, licensing criterion, or billing code is available that defines a crisis diversion center. Another explanation may be that crisis hotlines had only a short-term impact on preventing suicide crises because they did not provide follow-up (30). Repeat brief contacts with individuals at risk for suicide have shown a positive effect of reducing suicide-related outcomes and may be a way to enhance the effect of crisis hotlines (31). A final explanation may be that people experiencing a suicidal crisis did not receive evidence-based interventions that may prevent psychiatric hospitalizations, such as family psychoeducation, peer support, supported housing, supported employment, assertive community treatment, dialectical behavior therapy, and post-ED or hospital follow-up (32).
The results of this study should be understood along with its limitations. First, when evaluating the implementation of a complex system-level intervention such as a statewide crisis response system, it is rarely possible to find the perfect counterfactual. We used Nevada as a comparison state for Arizona because it was a western state, had suicide-related hospitalization and other data available, had not yet implemented a comprehensive crisis response system, and had preperiod trends similar to those in Arizona. We used a CITS approach that controls for time-invariant, unobserved state-level characteristics and observable time-varying characteristics. However, it is possible that the two states differed in some unmeasured, time-varying factors associated with the rate of suicide-related discharges. For example, changes in gun ownership may have differed between the two states, although 2016 data have revealed similar rates of gun ownership in Arizona and Nevada, and unmeasured behavioral health policies may have been implemented in Arizona or Nevada, such as programs aimed at reducing suicide among college students and improving suicide prevention training among behavioral health professionals (33, 34).
A second limitation was that the United States switched from ICD-9 to ICD-10 codes during the period studied. The switch was associated with an increase in suicide cases documented in hospital medical records (28). Per AHRQ’s approach to analyzing the HCUP SID data, in our CITS analysis we omitted the HCUP SID data in 2015 because hospitals transitioned from ICD-9 to ICD-10 during that time. This also meant that the first year of Arizona’s implementation was omitted. We assumed that the switch to ICD-10 affected Arizona and Nevada equally and thus would have no effect on the findings; however, this may not have been the case.
Third, our analysis focused on suicide-related crises because the 988 Suicide and Crisis Lifeline and the crisis response system are very focused on suicide, and there is no evidence that comprehensive crisis response systems reduce suicide-related hospital admissions. Arizona’s crisis response system may have had a larger effect on deterring hospitalizations for non–suicide-related behavioral health crises than suicide hospitalizations. It is important to understand how comprehensive crisis systems affect a range of outcomes, including suicide mortality rates. However, crisis response systems were developed with a focus on and anticipation of reducing suicide.
Fourth, Arizona’s crisis response system may have reduced suicide-related ED visits. We originally planned to compare Arizona’s and Nevada’s suicide-related ED visits, but the preperiod trends diverged, preventing a valid CITS model estimation. Fifth, the HCUP database excludes freestanding psychiatric hospitals. Finally, our analysis examined the effect of a crisis system before the 988 rollout and additional federal and state funding to support crisis services. Therefore, the results may not be generalizable to 2022 or future years.
Conclusions
Although this study did not find large effects of Arizona’s comprehensive crisis response system on suicide-related hospitalizations, investments in improving states’ crisis response systems are worthwhile even if they do not reduce hospitalizations and offset costs. Too often, individuals do not receive services when they are in crisis or receive services that are traumatic and stigmatizing and make their conditions worse (35, 36). Suicide is the 10th leading cause of death among all age groups and the second leading cause of death in age groups 10–14 and 25–34 years (37). In the United States, suicide rates increased 35% from 1999 to 2018, fell in 2019 and 2020, only to increase again in 2021 (38). In the face of this growing public health emergency, more aggressive responses that demonstrate efficacy are needed.
Supplementary Material
File (appi.ps.20220628.ds001.pdf)
- View/Download
- 190.47 KB
References
1.
National Guidelines for Behavioral Health Crisis Care: Best Practice Toolkit. Rockville, MD, Substance Abuse and Mental Health Services Administration, 2020. https://www.samhsa.gov/sites/default/files/national-guidelines-for-behavioral-health-crisis-care-02242020.pdf
2.
Crisis Now: Transforming Services Is Within Our Reach. Washington, DC, National Action Alliance for Suicide Prevention, Crisis Services Task Force, 2016. https://theactionalliance.org/resource/crisis-now-transforming-services-within-our-reach. Accessed July 17, 2023
3.
Hogan MF, Covington DW: Arizona Crisis System the Equivalent of Everest Base Camp . . . the Final Climb Still Ahead. Alexandria, VA, National Association of State Mental Health Program Directors, Crisis Now, 2020. https://talk.crisisnow.com/arizona-crisis-system-the-equivalent-of-everest-base-camp-the-final-climb-still-ahead. Accessed July 17, 2023
4.
Strategies and Interventions to Reduce Suicide: A Workshop. Washington, DC, National Academies Press, 2022. https://www.nationalacademies.org/our-work/strategies-and-interventions-to-reduce-suicide-a-workshop. Accessed July 17, 2023
5.
Broadway ED, Covington DW: A Comprehensive Crisis System: Ending Unnecessary Emergency Room Admissions and Jail Bookings Associated With Mental Illness. Alexandria, VA, National Association of State Mental Health Program Directors, 2018. https://nasmhpd.org/sites/default/files/TACPaper5_ComprehensiveCrisisSystem_508C.pdf
6.
Crisis Resource Need Calculator—Estimate Your Potential Annual Behavioral Health Acute Inpatient and Crisis Care System Cost. Alexandria, VA, National Association of State Mental Health Program Directors, Crisis Now, 2022. https://calculator.crisisnow.com/#. Accessed Oct 16, 2022
7.
Matthews S, Cantor JH, Brooks Holliday S, et al: Mental health emergency hotlines in the United States: a scoping review (2012–2021). Psychiatr Serv 2023; 74:513–522
8.
Hoffberg AS, Stearns-Yoder KA, Brenner LA: The effectiveness of crisis line services: a systematic review. Front Public Health 2020; 7:399
9.
Fendrich M, Ives M, Kurz B, et al: Impact of mobile crisis services on emergency department use among youths with behavioral health service needs. Psychiatr Serv 2019; 70:881–887
10.
Oblath R, Herrera CN, Were LPO, et al: Long-term trends in psychiatric emergency services delivered by the Boston emergency services team. Community Ment Health J 2023; 59:370–380
11.
Bengelsdorf H, Church JO, Kaye RA, et al: The cost effectiveness of crisis intervention. Admission diversion savings can offset the high cost of service. J Nerv Ment Dis 1993; 181:757–762
12.
Guo S, Biegel DE, Johnsen JA, et al: Assessing the impact of community-based mobile crisis services on preventing hospitalization. Psychiatr Serv 2001; 52:223–228
13.
Hugo M, Smout M, Bannister J: A comparison in hospitalization rates between a community-based mobile emergency service and a hospital-based emergency service. Aust N Z J Psychiatry 2002; 36:504–508
14.
Anderson K, Goldsmith LP, Lomani J, et al: Short-stay crisis units for mental health patients on crisis care pathways: systematic review and meta-analysis. BJPsych Open 2022; 8:e144
15.
Johnson S, Dalton-Locke C, Baker J, et al: Acute psychiatric care: approaches to increasing the range of services and improving access and quality of care. World Psychiatry 2022; 21:220–236
16.
Gordon SY: Building State Capacity to Address Behavioral Health Needs Through Crisis Services and Early Intervention. Issue Brief. New York, Millbank Memorial Fund, 2020. https://www.milbank.org/wp-content/uploads/2020/11/Building-States-Capacity-for-BH-Crisis_4.pdf
17.
Hogan MF, Goldman ML: New opportunities to improve mental health crisis systems. Psychiatr Serv 2021; 72:169–173
18.
Jimenez E: Three Lessons Washington Can Learn From How Arizona Helps People in Mental Health Crisis. Seattle Times 2022. https://www.seattletimes.com/seattle-news/mental-health/three-lessons-washington-can-learn-from-how-arizona-helps-people-in-mental-health-crisis. Accessed July 17, 2023
19.
Vistal C: “Arizona Model” for Behavioral Health Crisis Care Gains National Attention From Other States. azcentral 2020. https://www.azcentral.com/story/news/local/arizona-health/2020/02/21/other-states-copy-arizona-model-behavioral-health-crisis-care/4794168002. Accessed July 17, 2023
20.
Schultz J: Maricopa County Mental Health Crisis Services Grow. Cronkite News 2015. https://cronkitenews.azpbs.org/2015/11/03/maricopa-county-mental-health-crisis-care-services-grow. Accessed July 17, 2023
21.
Lewitzka U, Sauer C, Bauer M, et al: Are national suicide prevention programs effective? A comparison of 4 verum and 4 control countries over 30 years. BMC Psychiatry 2019; 19:158
22.
Wing C, Simon K, Bello-Gomez RA: Designing difference in difference studies: best practices for public health policy research. Annu Rev Public Health 2018; 39:453–469
23.
Toward a Comprehensive Crisis Response System in Nevada. Carson City, Nevada Department of Health and Human Services, Office of Suicide Prevention, 2021. https://dpbh.nv.gov/uploadedFiles/dpbhnvgov/content/Programs/ClinicalSAPTA/Nevada%20Crisis%20Systems%20White%20Paper%20Final%203-31.pdf
24.
Leslie S: While We Still Have a Long Way to Go, Nevada Is Making Strides in Suicide Prevention. Reno Gazette Journal 2022. https://www.rgj.com/story/opinion/2022/07/21/while-we-still-have-long-way-go-nevada-making-strides-suicide-prevention/10121065002. Accessed July 17, 2023
25.
Todd C: A Month Into 988, Call Volumes up, but State Behavioral Health Infrastructure Isn’t. Nevada Current 2022. https://www.nevadacurrent.com/2022/08/18/a-month-into-988-call-volumes-up-but-state-behavioral-health-infrastructure-isnt. Accessed July 17, 2023
26.
State Inpatient Databases (SID) Database Documentation. Rockville, MD, Agency for Healthcare Research and Quality, 2021. https://hcup-us.ahrq.gov/db/state/siddbdocumentation.jsp. Accessed July 17, 2023
27.
Linden A: Conducting interrupted time-series analysis for single-and multiple-group comparisons. Stata J 2015; 15:480–500
28.
Owens PL, McDermott KW, Lipari RN, et al: Emergency Department Visits Involving Suicidal Ideation or Suicide Attempt, 2008–2017. HCUP Statistical Brief no 263. Rockville, MD, Agency for Healthcare Research and Quality, 2020. www.hcup-us.ahrq.gov/reports/statbriefs/sb263-Suicide-ED-Visits-2008-2017.pdf
29.
Cumby RE, Huizinga J: Testing the autocorrelation structure of disturbances in ordinary least squares and instrumental variables regressions. Econometrica 1990; 60:185–195
30.
Zabelski S, Kaniuka AR, Robertson AR, et al: Crisis lines: current status and recommendations for research and policy. Psychiatr Serv 2023; 74:505–512
31.
Skopp NA, Smolenski DJ, Bush NE, et al: Caring contacts for suicide prevention: a systematic review and meta-analysis. Psychol Serv 2023; 20:74–83
32.
Drake RE, Bond GR: Psychiatric crisis care and the more is less paradox. Community Ment Health J 2021; 57:1230–1236
33.
Schell TL, Peterson S, Vegatabile BG, et al: State-Level Estimates of Household Firearm Ownership. Santa Monica, CA, RAND Corporation, 2020. https://www.rand.org/pubs/tools/TL354.html. Accessed July 17, 2023
34.
Nevada Office of Suicide Prevention Action Plan, 2017–2019. Carson City, Nevada Department of Health and Human Services, Office of Suicide Prevention, 2017. https://suicideprevention.nv.gov/uploadedFiles/suicideprevention.nv.gov/content/Resources/Reports/2017-19NV_OSP_Action_Plan.pdf
35.
Jina-Pettersen N: Fear, neglect, coercion, and dehumanization: is inpatient psychiatric trauma contributing to a public health crisis? J Patient Exp 2022; 9:23743735221079138
36.
Jones N, Gius B, Shields M, et al: Youths’ and young adults’ experiences of police involvement during initiation of involuntary psychiatric holds and transport. Psychiatr Serv 2022; 73:910–917
37.
About Underlying Cause of Death, 1999–2020. Atlanta, Centers for Disease Control and Prevention, 2022. https://wonder.cdc.gov/ucd-icd10.html. Accessed July 17, 2023
38.
Curtin SC, Garnett MF, Ahmad FB: Provisional Numbers and Rates of Suicide by Month and Demographic Characteristics: United States, 2021. Atlanta, Centers for Disease Control and Prevention, 2022. https://stacks.cdc.gov/view/cdc/120830. Accessed July 17, 2023
Information & Authors
Information
Published In
History
Received: 14 December 2022
Revision received: 23 June 2023
Accepted: 29 June 2023
Published online: 9 August 2023
Published in print: February 01, 2024
Keywords
Authors
Competing Interests
The authors report no financial relationships with commercial interests.
Metrics & Citations
Metrics
Citations
Export Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
For more information or tips please see 'Downloading to a citation manager' in the Help menu.
View Options
View options
PDF/EPUB
View PDF/EPUBLogin options
Already a subscriber? Access your subscription through your login credentials or your institution for full access to this article.
Personal login Institutional Login Open Athens loginNot a subscriber?
PsychiatryOnline subscription options offer access to the DSM-5-TR® library, books, journals, CME, and patient resources. This all-in-one virtual library provides psychiatrists and mental health professionals with key resources for diagnosis, treatment, research, and professional development.
Need more help? PsychiatryOnline Customer Service may be reached by emailing [email protected] or by calling 800-368-5777 (in the U.S.) or 703-907-7322 (outside the U.S.).