Abstract
The extreme risk protection order (ERPO) was conceived initially as a civil restraining order to temporarily suspend access to firearms for individuals behaving dangerously who are not otherwise legally barred from gun possession by a felony conviction or other gun-disqualifying record. In practice, however, ERPOs in many states are being applied in a range of different kinds of cases in conjunction with discretionary criminal law enforcement and prosecution, essentially as a tool of deflection or diversion from the criminal legal system. In this article, we develop a typology of the discretionary uses of ERPOs by police officers, prosecutors, and judges, in cases where an ERPO may be initiated as an alternative to arrest, a diversion from prosecution, a mitigating intervention to soften criminal charging and sentencing (reducing incarcerations), a concurrent legal intervention, or as a complementary tool for robust law enforcement. We illustrate the typology with case vignettes from Indiana, Washington, Virginia, and Florida. Although many ERPO petitions are initiated in response to suicide threats, with or without the presence of public risk, the article invites the question of whether, in some cases, ERPOs can serve effectively as a mechanism of deflection or diversion from the criminal legal system and suggests that future research should carefully examine both the process and outcome of ERPOs used in this way.
Keywords
Highlights
● While ERPO was designed as a civil restraining order to remove firearms from risky individuals, police and prosecutors are sometimes seeking these orders as an alternative to arrest or prosecution.
● The article outlines and illustrates a framework of uses of ERPO in cases where risk of firearm injury converges with criminal law enforcement.
● These unintended uses of ERPO may reduce harm, but further research is needed to evaluate their effectiveness in populations involved with or diverted from the criminal legal system.
Introduction
Gun violence prevention scholars initially conceived the extreme risk protection order (ERPO) as a public safety intervention not necessarily involving the criminal legal system. 1 An ERPO was designed as a civil restraining order to temporarily remove or deny access to firearms for a person who poses a substantial risk of injury to self or others, but who has not necessarily broken the law and could otherwise lawfully possess firearms. In practice, however, this innovative legal tool, often called a “red flag law” in the popular lexicon and commentary,2 -4 is often being applied in a range of different kinds of cases in conjunction with criminal law enforcement and prosecution.5,6
In this article, we develop a typology of the uses of ERPOs in cases where police make an arrest or could have done so, but did not, or where police make an arrest and ERPOs are used in conjunction with amelioration of criminal punishment. We illustrate the typology with case vignettes from Indiana, Washington, Virginia, and Florida—states where risk-based firearm removal laws were enacted in 2006, 2015, 2018, and 2020, respectively. The article invites the question of whether, in some cases, ERPOs can serve effectively as a mechanism of deflection or diversion from the criminal legal system and it suggests that future research should examine both the process and outcome of ERPOs used in this way.
Background on Extreme Risk Protection Order Laws
Common Features of ERPO Laws
Animated by public concern over mass shootings in recent years, as well as the organizing efforts of national and local advocacy groups, 7 21 states, the District of Columbia, and the U.S. Virgin Islands have enacted ERPO laws as of October, 2025.8,9 Virginia’s “substantial risk order” statute provides an informative specimen of these laws, as it hews quite closely to an expert consensus model described in 2013, originally called a “gun violence restraining order.”10,11 The underlying ERPO scheme was inspired by precursor risk-based gun removal laws in Connecticut and Indiana, and relies on the established legal framework of a civil domestic violence restraining order.
The Virginia law, in particular, allows an attorney for the Commonwealth or a law enforcement officer to request a civil court order to remove a person’s firearms when there is probable cause to believe that the person poses a significant risk of bodily injury to self or others in the near future by possessing or acquiring a firearm. In most other states with ERPO laws, the statute also authorizes family members of a person behaving dangerously to petition the court directly for an ERPO; a few states add other categories of authorized petitioners, such as educators and clinicians.
In Virginia, as in most other ERPO states, the court may first issue an ex parte “emergency substantial risk order” authorizing law enforcement officers to remove the respondent’s firearms, after giving the person an opportunity to voluntarily relinquish them. Law enforcement must take possession of any voluntarily surrendered firearms; otherwise, they can obtain a search warrant to search for and remove the respondent’s firearms. The emergency order expires in 14 days following its issuance, and within that period the court must hold a hearing and the state must show by clear and convincing evidence that the respondent poses a substantial risk of harm to self or others; if the state meets this burden, the judge may enter a ruling for an extended substantial risk order for a duration of up to 180 days. The analogous statute in most other states authorizes an extended ERPO for a maximum initial duration of up to 12 months. However, as of 2025, several states, including California, Delaware, and Hawaii, allow for a maximum duration of up to 5 years. In Indiana, the order stays in effect until terminated by a judge.
While the order remains active, the respondent is prohibited from possessing or purchasing firearms; the state must return the person’s firearms when the order expires. It is worth noting that in many jurisdictions, respondents may choose not to contest the issuance of ERPOs, accepting the terms of temporary firearm restriction without proceeding to a full evidentiary hearing. This procedural dynamic—whether strategic or pragmatic—has implications for legal access, procedural justice, and perceived legitimacy.
Preliminary Evidence of ERPO Use in Criminal Cases
Early research on the effectiveness of ERPOs analyzed outcome data from gun seizure subjects in Connecticut 12 and Indiana, 13 the first states to enact these kinds of laws in 1999 and 2006, respectively. In both states, the laws resulted from a legislative response to a public outcry over a multiple casualty shooting, but as they were implemented, the majority (about two-thirds) of gun seizure cases involved suicide threats. Accordingly, the analysis of the effectiveness of these laws to prevent firearm deaths has focused on suicide outcomes.
Researchers conducted statistical analysis of matched death records of the gun-removal subjects in both of these states. Criminal behavior was identified in a small minority of cases. In Connecticut, 17% of ERPO respondents were arrested in conjunction with the firearm seizure process. Yet only about 1 in 4 of the arrested individuals (4% of the total study population) received a criminal conviction, suggesting that civil court-ordered gun removal could have been used to leverage dismissal of criminal charges. However, this pattern—that only a small proportion of arrestees are prosecuted and convicted—also reflects broader trends in misdemeanor enforcement. How does the finding from Connecticut compare to the rate of conviction without ERPOs in the picture?
While prosecutors often report conviction rates as a proportion of cases they choose to prosecute, a more relevant benchmark for evaluating ERPOs is the proportion of all misdemeanor arrestees who are ultimately convicted. In Suffolk County, Massachusetts, Agan et al 14 found that about 15% to 20% of nonviolent misdemeanor arrestees were convicted, with nearly 75% of prosecuted cases ending in dismissal or diversion. In contrast, violent crimes involving firearms are far less likely to be dismissed. Leaving aside suicide, the occurrence of a gun crime represents the most relevant ERPO counterfactual—the event that ERPOs are designed to prevent. In Washington, DC, the U.S. Attorney’s Office reported that in 2024, over 90% of arrests for violent gun crimes were charged at the time of arrest, and approximately 75% resulted in conviction. 15 This disparity suggests that, with the exception of suicide threats, ERPOs may be most impactful when applied to individuals who pose a credible risk but whose conduct falls short of triggering a prosecutable violent offense—a population of concern for whom criminal legal intervention is often discretionary.
Considering evidence from other states, in Indiana only 5% of gun seizure respondents were arrested during the police encounter; slightly over half of these arrested respondents received a criminal conviction. In Washington State, 21% of respondents were concurrently arrested, and conviction data were not available. 16 In Virginia, where the ERPO law took effect in 2020, the Virginia Mercury published a report describing several cases that resulted in a concurrent arrest and criminal charges being filed; while statistical data are lacking, the Virginia cases illustrated novel uses of ERPO by criminal prosecutors. 17
Based on the information gleaned from these case descriptions, along with previously unpublished reports of circumstances surrounding selected gun seizures in Indiana, we developed the typology described below. This typology is meant to serve as a conceptual framework to compare and evaluate various ways in which ERPOs are being applied, and to help guide future ERPO research.
Relation Between ERPO Orders and Criminal Law Enforcement
Table 1 presents the typology of key ways in which ERPOs are being used in conjunction with discretionary law enforcement actions by police and prosecutors. Five types of orders are identified based on combinations of legal actions associated with ERPO cases. Arrest and conviction are represented on the horizontal dimension of the table, while the involvement of criminal charges—not involving guns, or involving guns—is represented on the vertical dimension. Because our main interest pertains to use of ERPOs in criminal law enforcement, the typology described below excludes ERPO cases in which there is no basis for an arrest (eg, such as a suicide threat unaccompanied by an independent criminal act) or where the ERPO is used solely to reduce the risk of firearm injury with no accompanying criminal charges.
Typology of the function of extreme risk protection orders (ERPOs) in conjunction with criminal offending.
The research study that assembled the de-identified case vignettes presented here was reviewed and approved by the [Institutional Review Board].
Type 1 ERPO: Alternative to Arrest
Law enforcement officers often face situations in which they may have probable cause to make an arrest, but choose instead to take a non-criminal course of action. A common example is when police encounter a person who appears to be acutely mentally ill. In many such cases, the police may decide to transport the person to a hospital for evaluation instead of making an arrest, even if they have probable cause to arrest the person for a minor offense—a process known as pre-arrest diversion. The option to petition for an ERPO provides police officers with another form of diversion, which they may exercise as a stand-alone intervention to mitigate what appears to be risk of firearm injury, or may impose in conjunction with a short-term involuntary psychiatric “hold” if the person of concern with firearms also appears to be mentally ill. 18 In such cases, no arrest is made, no criminal charges are filed, and there is no resulting criminal conviction record.
The proportion of ERPO cases in which the ERPO functions as an alternative to arrest is unknown, but is likely to vary substantially across jurisdictions. What is known from some previous studies is that the large majority of ERPO cases—for example, 95% in Indiana, 83% in Connecticut, and 79% in Washington State—did not result in an arrest. Further research is needed to reveal whether police had probable cause to make a criminal arrest in some of these cases; whether the ERPO petition played any direct or indirect role in their decision not to make an arrest; and how common such a scenario might be. The following real case descriptions illustrate uses of ERPOs in which risk-based firearm removal may have been pursued as a substitute for arrest, and thus a front-end diversion of the case from the criminal legal system.
It is important to emphasize that the issuance of the substantial risk order is a civil process and that the court’s authority to issue it does not depend on whether the respondent had committed a criminal offense. In both of these cases, the ERPO fulfilled its primary intended purpose of forbidding firearm access to individuals who posed a public safety risk—a threat from a disgruntled former employer who owned 19 firearms, and an acutely intoxicated person who tried to discharge a recently purchased handgun inside a car. Firearm seizure may also have helped to avert an arrest and associated legal consequences for these individuals, in 2 ways: by providing a non-criminalizing safety intervention for the police to pursue in the immediate situation, and by temporarily removing the means to commit a gun crime.
In Mr. B.’s case, the risk order also appears to have played a role in achieving a complementary public health goal: steering an individual with a likely serious behavioral health condition away from the portal of the criminal justice system and into treatment. Along these lines, a 2019 Connecticut study of persons who had their firearms removed under the state’s risk-warrant law found that respondents’ chances of receiving behavioral health services in the community more than doubled in the year following a gun removal action compared to the year before—from 12% to 29%. Thus, it is possible that Mr. B’s crisis-encounter with law enforcement also connected him to ongoing treatment and supportive services in the community, which in turn may have mitigated his baseline risk of injurious behavior. However, it is important to note that while a mandated mental health evaluation frequently coincides with an ERPO, in no state is treatment legally ordered or formally leveraged solely by a firearm removal order; a specific finding of need for treatment to mitigate a safety risk must be made by a judicial officer or, in emergency situations, by a properly licensed medical officer.
Type 2: ERPO as an Alternative to Prosecution
When police make an arrest in ERPO cases, prosecutors in some jurisdictions occasionally exercise their own discretion to dismiss the charges. As mentioned earlier, a study in Connecticut found that about 3 out of 4 firearm seizure respondents who were arrested at the time of the seizure were not convicted. This could have occurred for one of several reasons, in addition to acquittal at trial. Perhaps the prosecutor thought these cases lacked merit and decided not to proceed. Or the prosecutor might have considered that the case had prosecutorial merit but that the public interest would be more efficiently served by a civil order for temporary firearm removal. An ERPO could be a viable and sufficient alternative, especially in situations that involve domestic turmoil with firearms in the mix, where there is a clear risk of harm but little public demand for—or interest in—prosecution. Additional reasons for declining prosecution include witness unreliability or lack of admissible evidence, defendant mental health needs, and risk thresholds. Importantly, trials are resource-intensive, and prosecutors often prioritize cases with higher stakes or clearer statutory violations. Thus, ERPOs may serve as pragmatic substitutes or supplements when criminal charges are unlikely to proceed.
In these kinds of cases, the use of an ERPO may function in a similar way to other forms of diversion in the criminal justice system: a formal legal mechanism—sometimes involving negotiation between prosecutor and defense lawyer—that moves a case from criminal court to some other type of adjudication, such as a supervised treatment program. Thus, in Type 2 ERPO cases, there is initially an arrest and a criminal charge, followed by a prosecutor’s negotiated or unilateral decision to dismiss the charges in conjunction with a time-limited civil firearm prohibition. In a few Type 2 cases, the respondent is arrested for resisting the police who come to serve the gun removal order, or for engaging in other illegal behavior that the police discover or observe on the scene. The following case descriptions illustrate the Type 2 use of ERPO, where risk-based firearm removal appears to have been pursued from the outset as an alternative to prosecution:
In both of these cases, police responded to a domestic crisis and were able to avert a potentially lethal incident; however, the confrontation itself precipitated criminal behavior—the respondents refused to cooperate with police, brandished or recklessly fired a gun—actions for which they were arrested and charged. Notably, the ERPO played no role in these cases until after the arrest, when the option of using a temporary civil firearm ban became part of a negotiation with prosecutor to dispose of the criminal charges. The solution aims to ensure public safety by removing firearms from a potentially dangerous individual in a time of crisis, while simultaneously providing a pathway for that person to avoid criminal legal consequences.
Type 3: Mitigated Criminal Sanctions Through Plea Bargaining
In Type 3 ERPO cases, the result is often a criminal conviction, typically for a lesser offense than the original charge. These cases are deemed to have prosecutorial merit, but removal of weapons through an ERPO is a determinative part of a tradeoff that the prosecutor is willing to make in return for agreeing, for example, to expungement of conviction. In jurisdictions where pretrial release conditions prohibit firearm possession, ERPOs may appear duplicative. However, ERPOs are flexible and can be deployed as civil complements to restrictions imposed in the criminal context (ie, as a pretrial release condition, probation condition, or collateral consequence of criminal conviction), potentially extending the duration of firearm prohibitions. Of course, this sort of negotiated disposition often occurs without ERPOs, but the ERPO statute and associated protocols can formalize and standardize the process of firearm removal in these situations. A vignette illustrates Type 3:
In this case, Indiana’s firearm seizure law served 2 main purposes. First, as a public safety measure, the law was invoked to remove several firearms from the possession of an individual who was considered to pose a significant risk to others, and forbade him from reacquiring firearms for the duration of the order. Second, as a civil legal disposition, the assumed safeguard of a gun seizure and temporary prohibition allowed him to avoid conviction for a more serious offense; the weapons charge was withdrawn, presumably as part of a plea agreement.
Notably, the gun removal order did not entirely divert Mr. E. and Mr. F. from criminal justice involvement; both acquired convictions on charges stemming from other illegal behavior. Still, the ERPO was associated with mitigating a criminal sentence, and possibly helping these respondents avoid a felony conviction and a prison sentence. Still, the outcome in Mr. F’s case was unsatisfactory, and he was rearrested and convicted for a subsequent gun-related offense during the period when the gun prohibition was in effect (It is possible, however, that merely reacquiring a firearm in violation of the gun seizure order constituted the offense.).
Type 4: ERPO as a Concurrent But Independent Risk-Reducing Intervention
In Type 4, a criminal case takes its normal course, but the ERPO is used to accomplish its preventive purpose during or even after the termination of the criminal proceeding. In this regard, firearm removal through an ERPO serves the same purpose that a bail condition could serve. The difference is that bail conditions are derived from the criminal adjudication process; the ERPO is not.
In Type 4, the function of an ERPO is neither to eliminate nor lessen criminal sanctions (these individuals are eventually convicted as charged), but to add a protective layer of firearm restriction in cases that would not otherwise result in a federal or state firearm disability, and to put such a restriction in place in a timely way while criminal proceedings take their course. The following vignettes illustrate a Type 4 ERPO case. This is informative insofar as the ERPO complements the criminal order rather than serving as a negotiation tool to avoid it.
In these vignettes, the ERPO does not prevent Ms. G. or Mr. H. from getting arrested or prosecuted. Rather, it appears that the civil order of gun removal and prohibition was used to reduce the risk of harm posed by a criminally charged individual who was being released from jail and might otherwise have retained legal access to guns. Notably, in Indiana, a non-felony criminal indictment alone does not entail loss of gun rights, nor do most misdemeanor convictions. Here, then, it appears that ERPO was used as an adjunct to criminal prosecution—as a way to strengthen restrictions attached to a criminal disposition rather than as an alternative to it.
Type 5: ERPO as a Tool for Disarming Dangerous Criminals
In Type 5, ERPO is used essentially as a tool of criminal law enforcement by facilitating gun dispossession from individuals who are suspected of ongoing criminal activity. This category describes the use of ERPOs in unusual situations where a suspected dangerous criminal has evaded law enforcement’s attempts to gather sufficient evidence of felonious criminal activity to support an arrest. However, some of the evidence may be sufficient to support an ERPO petition alleging a risk of violence. If an ERPO is issued, it may save lives and it also provides a predicate for a criminal arrest if the respondent violates the ERPO prohibition by continuing to purchase or possess guns. In effect, the violation of the civil order becomes a criminal offense for which the respondent can be prosecuted. Two examples of Type 5 ERPO cases follow:
Type 5 cases are presumably quite unusual and could be controversial. Contrary to the main purpose of ERPOs as a tool to protect the public from firearm injuries, the essential function of ERPOs in these cases appears to be incapacitating known criminals who have evaded prosecution. In these contexts, ERPOs are being used to address a unique policy problem—providing a preventive tool when other legal interventions are seen as ineffectual or insufficient. There is no compelling argument against using ERPOs as a device for curtailing criminal activity by known criminals. Admittedly, this category is conceptually different from the others as an injury prevention measure used at the interface with criminal law enforcement; still, we include it here because it involves the deliberate use of ERPO statutes by prosecutors in criminal cases.
Discussion
ERPO laws fill several policy gaps between the permanent gun restriction categories based in federal law (mainly persons with felony criminal convictions or mental health commitments), and the many individuals in the community who manifest a worrisome and arguably predictable risk of firearm violence or suicide at certain times, but do not fit into any of those prohibiting classifications. The examples provided above are consistent with the goal of reducing risk but they also reinforce and amplify criminal justice operations. Research is needed to better understand the use and consequences of ERPOs in different contexts where the police, prosecutors, and judges may be implementing the law in a variety of ways not initially anticipated by ERPO’s legal architects.
Although the typology emphasizes intersections with the criminal legal system, suicide risk remains a central concern in ERPO petitions. Many Type 1 applications originate from suicidal threats that do not involve criminal conduct. In Type 2 or Type 3 cases, suicide risk often co-occurs with ambiguous threats to other people or presents broader public safety concerns. This overlap between self-harm and interpersonal risk highlights the need for further conceptual delineation and empirical study of ways in which ERPOs might be effectively deployed in these kinds of cases.
Jurisdictional variation in how local police and prosecutors are using ERPOs in criminal cases also invites attention to the larger questions of equity that play a prominent role in criminal justice reform. When enhanced legal authority to remove citizens’ firearms is placed in the hands of police officers, and when the discretionary use of that tool depends to some extent on the officers’ subjective perceptions of risk (which are notoriously susceptible to unconscious bias), will ERPOs be applied disproportionately in communities of color where relations with police are already fraught? Or, conversely, might the judicious use of ERPOs provide a non-criminalizing (or at least less criminalizing) alternative to prosecution—one that could support rather than undermine racial equity in the legal system?
In many situations where an individual with access to a firearm is behaving dangerously, a timely ERPO to remove the weapon can help the person in crisis avoid breaking the law or getting arrested; the ERPO may even provide an occasion for them to access needed treatment or social services in the community. Further cross-jurisdictional research is needed to bring empirical evidence to bear in considering some of these difficult questions. At this point, we know that criminal prosecution properly serves risk-reduction aims as well as punitive ones, and that prosecutors and judges have at their disposal a variety of discretionary legal tools—for example, conditions of bail and probation—that they can use to reduce risk of firearm injury in criminal cases. ERPOs provide a potentially valuable addition to this legal toolkit, especially given the limits of the Brady law and related state laws restricting access to firearms at the point of sale.
Footnotes
Acknowledgements
Support for the preparation of this article was provided by the Elizabeth K. Dollard Charitable Trust and the Fund for a Safer Future, New Venture Fund.
Ethical Considerations
The Institutional Review Board (IRB) of the Duke University Scho0l of Medicine previously determined that the de-identified and anonymized information pertaining to persons included in this article did not involve human subject of research and declared the study exempt from IRB review.
Consent to Participate
The information pertaining to individuals presented in this study was obtained from public records or in de-identified form. Consent was not obtained because the study did not constitute research on human subjects.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Research described in this article was supported by the Fund for a Safer Future/New Venture Fund and the Elizabeth K. Dollard Trust.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Author Statement
Both listed authors contributed to conceptualization, interpretation, review of literature, writing, and editing of this manuscript.