Implementation of Extreme Risk Protection Orders in Colorado from 2020 to 2022: Firearm relinquishment and return and petitioner characteristics

Abstract

Introduction

Firearm injury remains a public health problem, with nearly 50,000 firearm-related deaths in the US in 2021. Extreme risk protection orders (ERPOs) are civil restraining orders that intend to reduce firearm deaths by temporarily removing firearms from individuals who are threatening violence to themselves or others. We described ERPO use by petitioner type and implementation including firearm removal.

Methods

All ERPO petitions filed in Colorado (1/1/2020–12/31/2022) were analyzed using an established abstraction tool and team-based approach. Case data abstracted from petitions and court documents were analyzed descriptively.

Results

Over three years, there were 353 ERPO petitions filed in Colorado. Only 39 % percent of granted petitions had documentation of firearms being relinquished. The average number firearms relinquished was 1.8 with a range of 1 to 31 firearms. One third (37.7 %) of petitions mentioned a mental health issue, 10 % had a renewal request, and half (54.6 %) of petitions were filed by law enforcement (LE). LE petitions filed were more likely to be granted temporary ERPOs (94.3 % vs 35.0 %, p < 0.0001) and full year ERPOs (79.7 % vs 39.3 %, p < 0.0001) compared to non-LE petitions.

Conclusion

Results from these analyses shed light on data gaps surrounding ERPO use and implementation. Differences in LE vs others’ ERPO outcomes suggest a need for additional research and training. ERPOs’ efficacy hinges on removing access to firearms among those at risk, and a lack of documentation limits the ability to evaluate these policies. This suggests a need to standardize reporting to ensure ERPO utilization and impact can be evaluated.

Mini abstract: This descriptive study assessed use, implementation and data gaps surrounding Extreme Risk Protection Orders in Colorado.

Abbreviations: Extreme risk protection orders (ERPOs) are civil restraining orders that intend to reduce firearm deaths by temporarily removing firearms from individuals who are threatening violence to themselves or others.

1. Introduction

Firearm violence is a public health concern. Nearly 50,000 firearm deaths occurred in the US in 2021, 54 % of these were suicides (New Report Highlights U.S, 2021). From 2019 to 2020 the firearm homicide rate increased by 45 % and the suicide rate increased by 10 % in the U.S., representing nearly 10,000 additional firearm deaths (New Report Highlights U.S, 2021). In Colorado, 73 % (Workbook: COVDRS, 2023) of homicides and 54 % (Workbook, 2023) of suicides occurred using a firearm, and there were more than twice as many firearm suicides as homicides. Mass shootings in the U.S. have also increased in recent years; one study reports 20 % of recorded mass shootings occurred in the last five years (Silva, 2022). In Colorado, there were 60 mass shootings causing 1,874 injuries or deaths from 2014 to 2022 (Barnard et al., 2023).

In response to an increase in firearm violence, legislators across the country have enacted Extreme Risk Protection Order (ERPO) Laws. In 2021, the Federal government also passed the Safer Communities Act to provide grant funding for states to implement ERPOs (McBath, 2021). ERPOs are civil restraining orders that intend to reduce firearm deaths by temporarily removing firearms from individuals who are at imminent risk of inflicting violence upon themselves or others (Extreme Risk Protection Order: | Bloomberg American Health Initiative, 2022). As of September 2023, 21 states and the District of Columbia have passed ERPO laws, including Colorado (Extreme Risk Protection Orders | Colorado General Assembly, 2020). Colorado's ERPO law was enacted in 2019 and implemented on January 1st, 2020. In Colorado, from 2020 to 2022 eligible petitioners (those filing an ERPO petition) were law enforcement (LE) or family members (including romantic partners, those who share children, and roommates).

Under Colorado’s ERPO law, a petitioner asks a court to prohibit an at-risk person (Respondent) from possessing or acquiring firearms. The court reviews the petition in an emergency (“ex parte” or “temporary”) hearing and may grant a Temporary ERPO (TERPO). If this is granted, the Respondent is required to relinquish any firearms and may not obtain new ones for the duration of the order. The TERPO can last up to two weeks until a hearing where the 364-day, full ERPO may be granted. If the full ERPO is denied, the firearms are returned to the Respondent, and they can acquire additional ones. Once the full ERPO expires, firearms are either returned or a petitioner may re-petition a court to renew the ERPO.

Prior studies have assessed the implementation of ERPOs across time and among specific groups.. In California, data showed that ERPO uptake across counties increased over time (Pallin et al., 2020). In Colorado, ERPO use in self-described “2nd Amendment sanctuary” counties was shown to occur at lower rates than in counties which did not have any such declarations (Barnard et al., 2021) and JAMA REF. In Washington, a study was conducted assessing ERPO use and firearm removal (Rowhani-Rahbar et al., 2020) among individuals with dementia (Prater et al., 2021). ERPO implementation is a crucial precursor to ERPOs ultimate impact of reducing firearm-related harms.

Evidence regarding the impact of ERPO laws in reducing firearm violence is growing. Studies from Connecticut (Kivisto and Phalen, 2018) and Indiana (Swanson et al., 2019) have shown that ERPOs are associated with a reduction in firearm suicide. However, in San Diego County, California from 2016 to 2019, the law was not associated with changes in all forms of intentional firearm violence (Pallin et al., 2020). Descriptive process analyses highlight how ERPOs have been used to prevent mass shootings, including in specifically in California (Wintemute et al., 2019) and across a set of six US states (Zeoli et al., 2022), but neither study could assess if these mass shootings would have occurred without the ERPO in place.

While implementation of ERPO petitions has been examined previously, a more complete conceptualization of their utility requires an understanding of firearm relinquishment after a granted petition. One analysis of orders granted in California showed that firearms were removed in just over half (55.9 %) of cases, far less than the 84.2 % of cases in which the Respondent was believed to own a firearm (Pear et al., 2022) and in Washington state only 64 % of ERPO Respondents had documentation of a firearm being removed (Rooney et al., 2021). To our knowledge, no such analyses have occurred in states which do not have administrative data about firearm ownership or a standardized tool to report firearm surrender/ERPO compliance, – where documentation of firearm relinquishment is arguably more important. Additionally, critics of ERPOs frequently cite concerns about due process protections of Respondents’ firearm rights, including the right to have firearms returned after ERPOs have expired or been vacated. No previous studies have examined this process. While previous studies show that LE petitions have been granted more frequently than non-LE petitions, less is known about differences in case circumstances and firearm relinquishment between petitions made by these different groups. Without understanding what and when different petitioner types are using ERPOs as a tool, practitioners may not be able to advocate for policy change and educations may not be equipped to train eligible groups.

In the present study, we sought to (1) describe overall use of ERPOs in Colorado including how many petitions are filed, granted, renewed and expired; (2) describe implementation of ERPOs meaning if a Respondent has access to firearms, if those firearms removed or relinquished and alternatively, are firearms given back when a petition expires; and (3) describe petitioner type and examine differences among LE vs non-LE petitioners. Understanding ERPO experiences and challenges can serve to better inform policy creation and enaction in other states, including identifying how best to address concerns and facilitate evaluation and education surrounding ERPO laws.

2. Methods

2.1. Data collection

We obtained court records for all ERPO petitions filed in Colorado between January 1, 2020, and December 31, 2022. Petitions were collected through public information requests to the Colorado Judicial Department. All court documents are publicly available upon request. Once case numbers and court information were received, individual petitions were downloaded from the State of Colorado Judicial Department Public Access Terminal court database or by requesting information from individual county court clerks.

2.2. Data abstraction

Data elements were abstracted by trained staff using a standardized guide including 77 elements (Appendix A). Petitions with the same Respondent, petitioner and case information were considered duplicates and were removed. The abstraction guide included case circumstances across petitions – including, but not limited to data on petitioners, Respondents, case characteristics, precipitating events, petition renewals, firearm relinquishment, firearm return upon ERPO expiration, and court case outcomes. Variables collected varied in terms of required, distinct variables (i.e. DOB, sex, race) or check boxes (i.e. petitioner type, reasons for filing the ERPO), and description in court documents (i.e. open ended descriptions on the supporting evidence as to who an ERPO should be granted, descriptions in court proceedings about who a petitions is denied). Generally, demographic criteria were filled out by the petitioner and abstracted from distinct elements in court documents and ERPO petitions. Petitioner type and reasons for filing were collected via the EPRO petition as a check box. Case outcomes were also collected via distinct elements in court documents. Case circumstances and reasons for denial were all collected via court documents. There was no standardized form or way to collect information on firearm relinquishment and return, this information was often in court records or proceedings ad-hoc. The abstraction team (LB, MM, NB, CR, KT, CK) met on a biweekly basis to discuss questions related to abstraction process and data quality. Trained staff had inter-rater reliability: 85 % of a random 1 % sample and a random 10 % subsample of all cases were reviewed by a senior researcher (LB) to ensure consistent abstraction.

2.3. Data analysis

First, we used univariate analysis to summarize the petitioner and Respondent demographic characteristics, the number of and reasons for petitions being granted or denied and renewed. Using 2020–2022 one-year US census estimates data, we calculated an average rate of ERPO usage per year. Next, we examined ERPO implementation including documentation of firearm relinquishment among those cases where ERPOs were granted, and firearm return upon expiration of an order. Finally, we compared patterns of petition filing and granting by petitioner (LE vs non-LE). Case information and outcomes were compared using chi-squared or Fisher’s exact tests depending on cell size. This study was deemed exempt from IRB review by the Colorado Multisite Institutional Review Board, study number: 20-2487.

3. Results

3.1. Overall

From 2020 to 2022 there were 353 unique ERPO petitions filed in Colorado, including 96 in 2020, 103 in 2021 and 154 in 2022 (Fig. 1). We found 22 duplicate cases including the same petitioner, Respondent, and case information which were excluded from this analysis This was an average of 117.7 petitions per year or 2.0 per 100,000 people. The majority of Respondents were male (84.1 %) and White (64.0 %); 5.4 % of Respondents were Black and 2.8 % were Hispanic. The average age of Respondents was 41.4 years old (SD = 15.8) (Table 1).

Fig. 1.

Flow chart of ERPO cases and outcomes 2020–2022.

Over the study period, two thirds (238; 67.8 %) of petitions were granted temporary two-week TERPOs; of those, 167 (70.1 %) were subsequently granted full 365-day ERPOs. Of all granted full ERPOs, 16 (9.5 %) had a renewal request, all but one of which was granted (93.8 %). One Respondent was under an ERPO for the entire study period (meaning their ERPO was renewed twice). The most common reason for a petition being denied was that it did not meet the burden of proof (11.0 %) followed by an inappropriate relationship being listed (on the petition (i.e. the petitioner was not allowed under Colorado law to petition against the respondent, 5.0 %). Those relationships were most commonly a neighbor, community member, or co-worker. Over one third (37.7 %) of petitions mentioned a mental health issue (specific mental illness or recurring mental health issue, specified or not) in the petition and 5 % of these were ordered to have a mental health evaluation as part of the petition. One Respondent was under an ERPO for the entire study period (meaning their ERPO was renewed twice). The most common reason for a petition being denied was that it did not meet the burden of proof (11.0 %) followed by an inappropriate relationship being listed (5.0 %). Those relationships were most commonly a neighbor, community member, or co-worker. Over one third (37.7 %) of petitions mentioned a mental health issue (specific mental illness or recurring mental health issue, specified or not) in the petition and 5 % of these were ordered to have a mental health evaluation as part of the petition.

3.2. Firearm relinquishment and return

Of all petitions that were granted, 146 (87.4 %) noted that the Respondent had access to firearms. Only 66 of these case files (45.2 %) included documentation of firearms being relinquished. The average number of firearms relinquished was 1.8 (range = 1–31). Only one Respondent refused to relinquish their firearms, resulting in LE filing a search warrant. Only (20.6 %) of eligible petitions had documentation of firearms being returned upon expiration of the order.

3.3. Law enforcement vs. non-law enforcement petitioners

Out of all 353 ERPO petitions filed in the study period, 193 (54.6 %) were filed by LE. Among non-LE petitioners, 64(40.0 %) were current or former intimate partners, 38(23.8 %) were related by blood, marriage, or adoption, 16(10.0 %) resided with the Respondent, 14(8.9 %) were a legal guardian, 11(6.9 %) shared a child with the Respondent and 9(5.6) did not fit into any of the legal categories allowed by the ERPO law (all of which were denied). LE petitions filed were more likely to be granted TERPOs compared to non– LE petitions (94.3 % vs 35.0 %, p < 0.0001) and full year ERPOs (79.7 % vs 39.3 %, p < 0.0001) LE petitions were more likely (p = 0.0003) to be filed for threats against self (19.9 % vs 7.9 %) or both self and others (28.0 % vs 18.0), while non-LE petitions were more likely to be for threats against others alone (52.2 % vs 74.2 %). LE petitions filed were more likely to have immediate access to firearms compared to non-law enforcement petitions (65.8 % vs 11.9 %, p < 0.0001) and more likely to have documentation of firearm relinquishment compared to non-law enforcement petitions (48.8 % vs 21.1 %, p = 0.02; Table 2).

Table 2.

Law Enforcement vs non-Law Enforcement Petitioners of ERPO filed in Colorado from 2020 to 2023.

	LE (1 9 3) n(%)	Non-LE (1 6 0) n(%)	p-value
Petitioner			N/A
Law Enforcement	193 (100 %)	N/A
Current / former intimate partners	N/A	64(40.0)
Related by blood, marriage, or adoption	N/A	38(23.8)
Resided with the Respondent	N/A	16(10.0)
Legal guardian	N/A	14(8.9)
Shared a child	N/A	11(6.9)
None of these	N/A	9(5.6)
Unknown	N/A	8(5.0)
Case Outcomes
TERPO			0<.0001
Granted	182(94.3)	56(35.0)
Denied/dismissed	11(5.7)	104(65.0)
ERPO			0<.0001
Granted	145(79.7)	22(39.3)
Denied/dismissed	37(20.3)	34(60.7)
Filed for threats to			0.0003
Self	37(19.9)	10(7.9)
Others	97(52.2)	94(74.0)
Both	52(28.0)	23(18.1)
Implementation (only among those where TERPO or ERPO was granted)
Firearm access			<0.0001
Yes	127(65.8)	19(11.9)
No/unknown	66(34.2)	141(88.1)
Firearm relinquishment (only where firearm access is “Yes”)			0.0233
Yes	62(48.8)	4(21.1)
No/unknown	65(51.2)	15(78.9)

4. Discussion

Results from these analyses shed light on data gaps surrounding ERPO use and implementation. Specifically, data around petitioner types, documenting relinquishment of firearms when orders are granted, and returning firearms when orders expire or are lifted is lacking. We additionally found 16 ERPOs that were renewed and one that had been renewed twice. This may suggest that Respondents under an ERPO are able to reduce their perceived risk of violence such that their firearm rights are legally restored following the 364-day ERPO period. However, we found few Respondents with a documented mental health issue who were required to undergo professional evaluation as part of the order, although any voluntary (non-court-ordered) evaluations would not have been included in publicly-available ERPO documentation. Future research should build upon studies examining criminal justice and suicide outcomes, including subsequent arrests and charges after an ERPO(Swanson et al., 2019) is granted to evaluate orders’ long-term impact. Particularly, examining impacts on Respondents, including future criminal justice involvement and health-related outcomes (i.e. cause of death or hospitalization) to fully understand the impact of ERPOs on respondents at an individual level.

Differences in LE vs non-LE petition documentation suggest a need for additional training, education, and structural support surrounding ERPOs. Systematic improvements in recordkeeping requirements could help to improve the low levels of relinquishment documentation observed in cases where petitioners were not law enforcement officers. Examining differences in behavior between petitioner groups for ERPOs is critical in improving knowledge and education regarding when and how to apply for an ERPO. More targeted training could be made available to law enforcement and the public. Previous literature supports this idea among specific groups: social workers (Conrick et al., 2023) in Washington and physicians in Washington (Gause et al., 2022) and Maryland (Hollo et al., 2022) suggest a willingness to file ERPOS with additional education or support. Emphasizing ERPO’s utility in suicide prevention may be especially helpful given that most firearm deaths are suicides (Pear et al., 2022), and the majority of petitions are filed for threats against others in Colorado. Further research is needed to learn more about the circumstances surrounding ERPO cases and how and when LE petitions are pursued at the suggestion of and/or in coordination with family members. Such work may highlight that petitions filed by family members were those that LE deemed to have insufficient evidence of risk, leading to a lower rate of TERPO and ERPO granting among non-LE petitions. Alternatively, as LE has access to criminal history records, a greater understanding of court proceedings, and often enjoy support in filing petitions from district or city attorneys, it is possible that LE petitions are presented in a manner engendering greater perceived credibility and contain more pertinent background information unavailable to non-LE petitioners. Additionally, a new law was passed in Colorado in 2023 that enabled several new professional groups to file ERPO petitions (Extreme Risk Protection Orders | Colorado General Assembly. Accessed August 13, 2021) including licensed medical and mental health professionals, educators (including K-12 and higher education), and district attorneys. Future research should evaluate petitioning among these groups and evaluate characteristics and outcomes of petitions to further target training.

ERPOs cannot be effective if Respondents maintain access to firearms while an order is in place. Ecological policy evaluations for firearm access laws are not sufficient to understand if or how these alws are working. ERPOs are being granted in Colorado − however, without knowing if firearms are being removed from Respondents, we may not be able to effectively evaluate the true impact of granted orders on firearm injuries and deaths. In California only 55.9 % of cases and in Washington state only 64 % of ERPO Respondents had documentation of a firearm being removed from their possession after a full length ERPO was granted. In Colorado, we found only 39 % of granted TERPO and/or ERPO petitions reported firearm removal or relinquishment. Whether the observed lack of available documentation reflects either 1) differences in protocols for documenting firearms which have been relinquished, or 2) evidence that firearms are not being relinquished per statutory requirements, is not clear. Addressing this question is critical to understanding whether ERPOs are affecting the access to firearms, which is necessary to understand whether they can logically reduce firearm injuries or deaths. In the same vein, return of firearms after ERPOs expire is also crucial to due process and the restoration of full rights to Respondents who the courts no longer believe to be at elevated risk for violence. Training and education around this process should be standardized for law enforcement tasked with serving orders which have been granted, collecting, and returning firearms.

The most prevalent limitations to this study are related to data missingness, including a substantial proportion of data about case documentation, including the reasons for the petition being filed and denied. This is because of variability in county processes for requesting and obtaining documents, inconsistent redaction processes by individual counties, and differences in case evidence submitted by petitioners. This study’s team recommends a more standardized process for both reporting and collecting data to ensure EPROs can be properly evaluated.

Results from this evaluation shed light on data gaps surrounding ERPO use and implementation. The rate of LE ERPO petitions being granted vs other petitioners of ERPOs suggest a need for additional research into how these petitions, circumstances or respondents differ and training/education for non-LE petitioners. ERPOs’ efficacy hinges on removing access to firearms among those at risk, and a lack of documentation limits the ability to evaluate these policies and their efficacy. This suggests a need to standardize the process for reporting and collecting petitions to ensure ERPO utilization and impact can be properly evaluated. Possible remedies include standardized public reporting of common de-identified data elements, including jurisdictional information, relationship between respondent and petitioner, case disposition, and presence or absence of documentation of firearm relinquishment or other dispossession. Extreme risk protection orders (ERPOs) are a promising tool to reduce firearm deaths; work to evaluate their use, implementation should be ongoing to understand how to make these laws most effective.

5. Disclosure of funding

Drs Barnard, Betz, and Ms McCarthy reported receiving grants from the Colorado Department of Public Health and Environment and National Collaborative on Gun Violence Research, during the conduct of the study. Dr. Knoepke reported receiving grants from the Fund for a Safer Future, and the National Heart, Lung, and Blood Institute during the conduct of the study. No other disclosures were reported. This research was funded in part by the National Collaborative on Gun Violence Research. The views expressed in this manuscript are the authors' and do not necessarily reflect the view of the National Collaborative on Gun Violence Research.

Funding statement

Ms Barnard, Thies and McCarthy and Drs Knoepke and Betz reported receiving grants from the Colorado Department of Public Health and Environment during the conduct of the study. Dr. Knoepke additionally receives grant funding from the Fund for a Safer Future and the National Heart, Lung, and Blood Institute (K23HL153892).

This research was funded in part by the National Collaborative on Gun Violence Research. The views expressed in this manuscript are the authors' and do not necessarily reflect the view of the National Collaborative on Gun Violence Research or authors’ employers.

CRediT authorship contribution statement

Leslie M Barnard: Writing – original draft, Methodology, Formal analysis, Data curation, Conceptualization. Nisha Batta: Writing – review & editing, Data curation. Megan McCarthy: Writing – review & editing, Data curation. Kimberly Thies: Writing – review & editing, Data curation. Caitlin Robinson: Writing – review & editing, Data curation. Marcus Schultze: . Marian E. Betz: Writing – review & editing, Funding acquisition, Conceptualization. Christopher E. Knoepke: Writing – review & editing, Methodology, Funding acquisition, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Data availability

Data will be made available on request.