Firearm rights restoration after prohibiting criminal convictions: a statewide descriptive study

Abstract

Background

In the US, federal law prohibits firearm possession and purchasing for individuals with convictions for a felony or domestic violence misdemeanor. If a prohibiting conviction occurred in a state court, then state law dictates how/when a prohibited person can later have their firearm rights reinstated. To our knowledge, there is limited empirical research documenting the frequency and characteristics of individuals who receive firearm rights restoration (FRR). Research is needed to guide policy, balancing the need to prioritize public safety while also fairly re-instating rights for formerly convicted individuals. We sought to address this research gap by examining FRR data from Washington state (2017–2022). During this time period, FRR granting was largely procedural (i.e., non-discretionary).

Methods

We used data from Washington State Patrol to identify who received FRR statewide. We linked these data to prior criminal convictions, arrests, and protection orders to understand the person’s characteristics and violent behaviors prior to FRR.

Results

We identified n = 11,516 individuals in the state of Washington who received FRR over the six-year study period (~ 1,900 people annually). FRR recipients were mostly male (88.4%, n = 10,182) and majority White (84.2%, n = 9697). Half (51.4%, n = 5921) had a prior conviction for a violent crime. On average, 15.3 years (sd: 6.1, median: 15.0) had elapsed between the last prohibiting conviction and FRR. In the 5 years prior to FRR, 710 individuals had been arrested, and 200 had been restrained in a protection order.

Conclusions

By understanding who receives FRR and how often, this research helps lay the foundation to improve our understanding of FRR legal practices. Our findings suggest that variability within the population of prohibited individuals could help inform FRR eligibility considerations to meet the goals of both preventing firearm violence and equitably re-instating rights to formerly convicted individuals.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40621-025-00635-4.

Introduction

Firearm violence is a serious public health and public safety problem in the US. Each year, about 17,715 people are shot and killed by another person with a firearm [1], and an additional 35,134 people are treated in the emergency department for firearm assault-related injuries [2]. Even more people are threatened with firearms without sustaining a physical injury, for example in an armed robbery or by an abusive intimate partner [3–5]. Exposure to firearm violence—whether as a victim, a witness, a grieving loved one, or simply through awareness of its presence in one’s community—can profoundly affect a person’s mental health, heightening fear, anxiety, psychological distress, and undermining overall well-being [6–9].

There are several known risk factors for violence perpetration, including prior criminal convictions [10, 11]. Among handgun purchasers, those with a prior criminal history have 4.9–7.0 times the risk for later conviction of a violent crime compared to handgun purchasers without a criminal history [11]. Those with a more extensive criminal history are at the greatest risk of perpetrating subsequent crimes [11]. Domestic violence-related misdemeanor convictions are also key indicators for violence; among handgun purchasers, they are associated with 45–80 times the risk of having a subsequent criminal charge for a violent crime, and 30 times the risk of having a subsequent criminal charge related to illegal or violent crimes with a firearm [12].

The risk for harm in a violent situation is also greater when a firearm is present. If the perpetrator has a firearm during an assault, the odds of fatality more than doubles [13]. When men with a history of abusing their female intimate partner have access to firearms, the risk that they will murder their partner increases by a factor of five [14].

One promising approach to curb firearm violence is by limiting access to firearms for individuals who pose an elevated risk of harming others [15]. Accordingly, federal law prohibits firearm possession and purchasing for individuals with prior felony or domestic violence (DV) misdemeanor convictions [16]. When these firearm prohibitions are properly implemented and enforced, research suggests that they can prevent violent crime and intimate partner homicide [11, 17–19]. Specifically, research shows that among individuals arrested for a felony, those who were not convicted and thus were allowed to legally purchase a handgun had a 1.2-fold greater risk of subsequent offending compared to those who were arrested, convicted, and thus denied the ability to legally purchase a handgun [19].

While the risk of violence perpetration among individuals with a prior criminal record is well-established, recidivism rates taper off over time [20, 21]. For individuals who abstain from perpetrating new crimes for 6–7 years, the risk of re-offending begins to approximate crime risk of the general population of people without a prior criminal record [20]. The duration for this tapering process varies. Younger individuals tend to have longer periods of criminal involvement before desisting compared to older individuals [20, 21]. Violence history also matters; someone with a prior violent offense has a longer, more-enduring risk for perpetrating violent crime compared to individuals who engaged in non-violent crime [21]. The tapering process may also take longer for people with prior firearm-related convictions [22, 23].

Acknowledging this desistance process, a prohibited person can later seek to have their firearm rights restored. Most convictions occur in state courts where firearm prohibitions are implemented according to state law, and where firearm rights restoration (FRR) is also dictated by state law [24, 25]. Our prior review found limited empirical data documenting how often prohibited individuals receive FRR, and almost no information about who receives FRR in the US [26]. Studying FRR is an important area for research. There have long been tensions in public policy between the need for rehabilitation and risk management. FRR reflects a belief in second chances but also raises serious public safety concerns. Balancing these goals is critical for effective FRR policy.

While research has documented the effectiveness of firearm prohibitions in reducing violence [17–19], information is lacking on how often and under what circumstances these restrictions are lifted. Researchers and policy makers have been largely unable to study this topic given incomplete data collection and inconsistent recording practices in most states and localities in the US [26]. In the present study, we sought to address this research gap by: (a) identifying the number of people who received FRR in Washington state (WA), (b) characterizing the demographics of this population, and (c) describing their criminal history and recent violent behaviors prior to FRR. By documenting who receives FRR and how often, this research can help lay the foundation to improve our understanding of how FRR legal practices align with broader goals of balancing public safety and equitable reinstatement of rights for formerly convicted individuals. To do this, it is vital to understand the demographics of those who receive FRR, as well as their prior criminal and violence histories.

We employ several measures to characterize violent behaviors prior to FRR, including not only criminal convictions and arrests, but also considering whether the person who received FRR was restrained in a protection order (PO). POs are the most common legal intervention used by DV victim-survivors to address DV perpetration in the US [27]. By considering whether individuals who received FRR were restrained in a PO (i.e., the courts found that they perpetrated DV), we seek to more broadly conceptualize and describe possible harm and violence perpetration prior to FRR.

Data challenges are common in the field of firearm injury and violence research; to our knowledge, no comprehensive data exists regarding how many people have lost their legal right to possess or purchase firearms, nor is there comprehensive data about FRR eligibility, denials, or firearm purchasing/ownership among US residents [26, 28]. As a result of these limitations, we were unable to ascertain how many people were actively prohibited in WA during the study period, how many individuals attempted to regain their firearm rights, or how often people who received FRR subsequently became legal firearm owners.

FRR state policies

State approaches to FRR are heterogeneous—a 2020 review found that state FRR policies ranged from automatic restoration upon completion of a sentence, to requiring more formalized petitioning to the courts, to mandating that the prohibited person receive a full pardon or expungement for their prohibiting crimes before firearm rights could be re-instated [25, 29]. FRR eligibility and processes further vary within states based on the type of crime of which the prohibited person was convicted. Some states differentiate their approach to FRR based on whether the crime was violent or non-violent, whether it included a dangerous weapon, or whether the prohibiting conviction was for a felony or a misdemeanor [25, 29]. Depending on these factors, states may require that prohibited individuals remain crime-free for different amounts of time before they can receive FRR [25, 29]. Further differences exist both within and across states for reinstating the right to own long guns compared to handguns [25, 29].

FRR in Washington state

For the present study, we used WA data from 2017–2022. Several changes occurred to WA law in 2023 which primarily focused on clarifying legal ambiguities in FRR processes, adding new prohibiting convictions, and streamlining administrative protocols, with some changes to FRR eligibility and granting criteria [30]. In this section, we only describe policies and processes that applied during the study period, prior to the 2023 changes.

From 2017–2022, all felony convictions in WA state were prohibiting (including violent or nonviolent felony crimes), as well as specific DV-related misdemeanors including assault in the fourth degree, coercion, stalking, reckless endangerment, criminal trespass in the first degree, violation of a protection order, and harassment [30–32]. These prohibitors are largely consistent with federal law and with prohibitors in other US states [33]. To receive FRR, a prohibited person could petition either the sentencing court where their prohibiting conviction occurred or the court in the county where they reside [30–32]. They would be considered eligible if they had been crime-free for five years after a felony, or three years crime-free after a DV misdemeanor conviction [30–32]. Certain crimes disqualified someone from FRR including a Class A felony (which includes severe crimes such as first-degree murder, rape, or arson in the first degree), a felony sex offense, or a felony punishable by 20 years [30–32]. During the study period, if the prohibited person met the legal criteria for FRR eligibility, then the court was required to grant FRR after a hearing regardless of potential risks for re-offending and irrespective of the other underlying fact patterns that led to the prohibition [30, 31, 34]. Because of this non-discretionary approach to FRR in WA, we expected our sample of individuals who received FRR to be inclusive of a range of prohibiting crimes and circumstances.

When a petition for FRR by a prohibited person was granted by a judge, the local courts would send the FRR order to Washington State Patrol (WSP) for processing. Once WSP received the order, WSP would record a note such as “right to possess a firearm restored” or “restored federal firearm privilege” as a subsequent disposition in their conviction data. This systematic process began in 2017; reliable data on granted FRR was not maintained prior to that date. To our knowledge WSP has not maintained consistent data on FRR petitions or denials, and thus we did not examine petitions or denials during the present study. During the study period, WSP was also responsible for running a background check and manually reviewing all granted FRR orders to confirm that the prohibited person was indeed eligible for FRR [31]. It was only after this last step that firearm rights would be officially re-instated.

Methods

Data

We used the WSP court record data to identify who received FRR (2017–2022). WSP data uses a person-level identifier based on fingerprints to systematically record individual arrests and convictions over time. We included any individual who had a conviction record attached to a subsequent disposition that indicated receipt of FRR. We excluded individuals with incomplete information about their prohibiting crimes, or who may not have met the final FRR eligibility criterion as verified by WSP. Data were only available for granted FRR orders in WA.

We also used the WSP court record and arrest data to describe the criminal history of these individuals, including prior prohibiting convictions and conviction history overall. WSP court record data capture any convictions that occurred in-state since 1972 for all defendants who were fingerprinted at the time of the arrest.

To complement data about criminal convictions and arrest history, we determined whether any FRR recipients had been subject to a PO that was granted prior to FRR. We linked person-level WSP records to granted PO records (2000–2022) from the WA Administrative Office of the Courts (AOC). In the linkage, we took a probabilistic and deterministic approach using name and date of birth as linkage fields with the Link King program [35]. With a deterministic approach, we assumed that any two individuals who had the exact same first name, last name, and date of birth across two WA state datasets were the same person. A probabilistic approach models the probability of each pair of records across datasets being a match using the Link King program algorithm [25]. This allowed us to capture matches with were minor misspellings or variations in names (e.g., “Robert” listed as the first name in one dataset, “Bob” listed as the first name in another dataset, all other name and date of birth fields matching for this set of records), in accordance with best practice [36, 37]. This study was approved by the University of Washington Institutional Review Board in December 2022.

Measures

Demographics. WSP data included information about age at the time of FRR, sex, and race. Information about race may have been self-identified or may have been assessed by a third party such as by a law enforcement officer. WSP race data included five categories: American Indian/Alaskan Native (AI/AN), Asian, Black, Unknown, or White. There was no measure for ethnicity (i.e., Latine/Hispanic) and no option to indicate more than one racialized identity for a person.

Prohibiting convictions. We only considered a charge as prohibiting if the disposition was listed as guilty, vacated, or if it was a sealed juvenile record. For each person who received FRR, we extracted details about their prohibiting convictions including the number of prohibiting convictions (based on unique case numbers), the region(s) where they were adjudicated, whether the conviction(s) occurred when the defendant was a juvenile, and how many prohibiting convictions were subsequently vacated.

We also recorded whether the date of the prohibiting conviction was available or unavailable from WSP; the date of the original conviction would be erased if the conviction was subsequently vacated. Convictions could have been vacated for good behavior [38], or because a prior charge was deemed no longer constitutional [39]. A comparison of FRR recipient characteristics based on who had all, some, or none of their prohibiting convictions vacated is included in Appendix 1. When reporting information about the timing of prohibiting convictions, we reported only on individuals where at least one prohibiting conviction date was known. Based on those dates, we calculated the person’s age at the time of their last prohibiting conviction, and the number of years that elapsed between the last known prohibiting conviction and FRR granting.

Conviction history – We characterized conviction history before FRR by counting the total number of convictions (including prohibiting convictions). To characterize whether crimes were violent (yes/no), members of the research team reviewed the crime literals and applied the World Health Organization (WHO) definition for violence, which conceptualizes violence broadly as "the intentional use of physical force or power, threatened or actual, against oneself, another person, or against a group or community, that either results in or has a high likelihood of resulting in injury, death, psychological harm, maldevelopment, or deprivation [40]." We also describe whether people were convicted of specific charges that may indicate an elevated risk for violence perpetration, including both violent and non-violent crimes such as animal cruelty, assault, driving under the influence, harassment, or kidnapping [41–43]. The category of “weapon law violations” in WA mostly included non-violent crimes related to providing weapon access to children or other prohibited people, possessing illegal weapons, or possessing legal weapons illegally [44]. The category of “domestic violence-related” crimes counted only convictions that had been flagged for DV in the WSP court record data.

Arrests– Based on WSP arrest data, we described whether individuals had any arrests in the five and 10 years immediately before FRR. We used the same process as described above to characterize whether these arrests were for a violent or non-violent crime consistent with the WHO definition [40]. We focused on the five and ten-year time period because violence perpetration risk usually remains fairly acute within five years after someone commits a violent crime [45], while recidivism tapers off over time and may be substantially lower after 10 years [20, 21].

Restrained in a protection order—Based on the linked AOC data, we describe whether individuals who received FRR were subject to a granted PO in the 10 and 5 years prior to FRR. This included several types of POs including for DV, sexual assault, harassment, stalking, vulnerable adults, extreme risk, and for criminal no-contact orders.

Results

During the study period (2017–2022), there were n = 12,676 individuals with information about FRR in their criminal conviction records. After excluding potentially ineligible individuals, the final analytical sample included n = 11,516 individuals who received FRR (Fig. 1), representing about 1,919 individuals who received FRR each year on average.

Fig. 1.

Identification of the cohort of individuals whose firearm rights were restored. ¹Information about the prohibiting conviction may not be recorded clearly if the conviction occurred in a municipal court or if the conviction occurred out-of-state. ²When the disposition for a criminal case was “unknown” or “dismissed,” it still may have been prohibiting if a guilty plea was entered. We did not have access to information about pleas and thus to be conservative with our inclusion criteria, we excluded those cases. ³A Class A Felony conviction should permanently disqualify someone from FRR, and thus anyone with a Class A Felony was removed from our sample. We did not have access to information about other potentially disqualifying criteria (e.g., involuntary treatment, prohibiting offenses from other states), or whether there were other extenuating circumstances that may have disqualified them for FRR (RCW 9.41.02). ⁴If a person submitted FRR paperwork without listing all prohibiting crimes, it may also have been rejected by WSP. We identified any individuals who had multiple prohibiting crimes and excluded any that did not have “firearm rights restored” recorded for each prohibitor, as this could represent incomplete filing paperwork

FRR recipients were mostly male (88.4%, n = 10,182) and majority White (84.2%, n = 9697, Table 1). The average age was 42.71 years (sd:10.14, median: 44) at the time of FRR. FRR occurred most often in Western WA (75.2%, n = 8665), the most populous region of the state. A large proportion of FRR recipients had some (14.4%, n = 1664) or all (21.2%, n = 2441) of their prohibiting convictions vacated by the time we received the data for this study. As a result, no prohibiting conviction dates were available for about a third (30.4%, n = 3497) of the sample (including the n = 2,441 FRR recipients who had all their prohibiting convictions vacated, plus an additional n = 1,056 FRR recipients who had both vacated prohibiting convictions and/or missing dates for unknown reasons). For most (69.6%, n = 8019) FRR recipients, at least some or all prohibiting conviction dates were known. A comparison of FRR recipient characteristics based on who had all, some, or none of their prohibiting convictions vacated is included in Appendix 1. People who had all prohibiting convictions vacated before receiving FRR were more often female, less often were convicted of a violent prohibiting crime, and had fewer prohibiting convictions overall (Appendix 1). Otherwise, these groups appeared to be relatively similar.

Table 1.

Characteristics of individuals who had their firearm rights restored 2017–2022 in the state of Washington after a prior prohibiting criminal conviction (n = 11,516)

	Overall n (%)
Person characteristics
Sex
Male	10182 (88.4)
Female	1287 (11.2)
Unknown	47 (0.4)
Race
American Indian/Alaskan Native	64 (0.6)
Asian	338 (2.9)
Black	954 (8.3)
Unknown	463 (4.0)
White	9697 (84.2)
Age at the time of FRR (mean, sd)	42.71 (10.14)
16–24 years	173 (1.5)
25–34 years	2350 (20.4)
35–44 years	4592 (39.9)
45–54 years	2808 (24.4)
55–64 years	1282 (11.1)
65+ years	311 (2.7)
Details about the prohibiting crime(s)
Region where the prohibiting conviction occurred
Central WA	1670 (14.5)
Eastern WA	1180 (10.2)
Western WA	8665 (75.2)
Unknown	1 (0.0)
Record status1
All vacated	2441 (21.2)
Some vacated	1664 (14.4)
None vacated	7411 (64.4)
Prohibiting conviction dates available
No dates known	3497 (30.4)
Some or all dates known	8019 (69.6)
Prohibiting conviction occurred when the person was a juvenile
All	848 (7.4)
Some	1398 (12.1)
None/unknown	9270 (80.5)

¹When a conviction is vacated the date of the original conviction becomes unavailable for research use.

FRR recipients’ overall criminal conviction history and specific prohibiting convictions prior to FRR are described in Table 2. Most FRR recipients had only one prohibiting conviction on their record in WA (68.1%, n = 7846) although several had two (17.0%, n = 1960), three (7.4%, n = 852), or more (7.5%, n = 858). Considering their overall criminal history prior to FRR (Table 2), FRR recipients had on average 3.5 (sd: 3.4, median: 3) prior criminal convictions. For half (51.4%, n = 5921) of the sample, one or more prior convictions were for a violent crime. Criminal conviction history included different types of crimes, including assault (43.2%, n = 4972), domestic violence-related crimes broadly (38.9%, n = 4484), driving under the influence (17.6%, n = 2030), PO violations (11.7%, n = 1344), and weapon law violations (10.7%, n = 1234), among others.

Table 2.

Conviction history before firearm rights restoration (n = 11,516)

	Overall criminal history n (%)	Prohibiting convictions n (%)
Count of prior criminal convictions (mean, sd)	3.5 (3.4)	1.6 (1.2)
One	4084 (35.5)	7846 (68.1)
Two	2196 (19.1)	1960 (17.0)
Three	1457 (12.7)	852 (7.4)
Four or more	3779 (32.8)	858 (7.5)
Convicted of a violent crime1
Yes	5921 (51.4)	5543 (48.1)
No	5595 (48.6)	5973 (51.9)
Convicted of any specific charge that could indicate risk for subsequent violence perpetration2,3
Assault	4972 (43.2)	4669 (40.5)
Domestic violence-related crime	4484 (38.9)	4053 (36.6)
Protection order violation	1344 (11.7)	1315 (11.4)
Driving under the influence (DUI)	2030 (17.6)	15 (0.1)
Weapon law violations	1234 (10.7)	877 (7.6)
Harassment	669 (5.8)	297 (2.6)
Robbery	274 (2.4)	273 (2.4)
Kidnapping	106 (0.9)	89 (0.8)
Animal cruelty	31 (0.3)	21 (0.2)
Manslaughter or vehicular homicide	16 (0.1)	16 (0.1)
Sex offenses	19 (0.2)	11 (0.1)

¹We assessed whether a crime was violent or not using the crime literals (i.e., brief text description of the offense) based on the World Health Organization (WHO) definition for violence.

²Categories are not mutually exclusive; one person may have multiple different types of criminal convictions. Percentages will add up to greater than 100%.

³Crimes were considered to be prohibiting during the study period if they were charged as felonies or they included the following DV misdemeanor crimes: assault in the fourth degree, coercion, stalking, reckless endangerment, criminal trespass in the first degree, violation of a protection order, or harassment [30–32]. Note that if a misdemeanor conviction fell into one of these crime categories but DV was not included as a charge element or a charge modifier, then the crime would not have been considered prohibiting [30–32].

Among FRR recipients where dates for their prohibiting crimes were available, the average age at the time of the last prohibiting conviction was 27.39 years (sd: 9.40, median: 25), and 15.34 years (sd: 6.18, median: 15.0) had elapsed between the last prohibiting conviction and FRR granting (see Table 3).

Table 3.

Time of the prohibiting conviction, among cases where date information was available for the prohibiting conviction 2017–2022 (n = 8,019)

	Overall n (%)
Age at last prohibiting conviction (mean, sd)	27.39 (9.40)
16–24 years	3462 (46.9)
25–34 years	2367 (32.0)
35–44 years	1104 (14.9)
45–54 years	358 (4.8)
55–64 years	83 (1.1)
65 + years	12 (0.2)
Time from last prohibiting conviction to FRR granting in years (mean, sd)	15.34 (6.18)
< = 4 years	113 (1.5)
5–9 years	1558 (20.0)
10–14 years	2247 (28.9)
15–19 years	1922 (24.7)
20–24 years	1406 (18.1)
25 years +	541 (6.9)

FRR: Firearm rights restoration.

We characterized arrests and reported whether FRR recipients had been restrained in a PO in the 10 and 5 years prior to FRR (Table 4). A subset had a recent arrest within 10 years (n = 2,354, 20.8%) or 5 years (n = 710, 6.2%) prior to FRR. More specifically, 7.0% (n = 807) had been arrested for a violent crime in the 10 years prior to FRR, and 2.2% (n = 253) in the 5 years prior to FRR. Hundreds of individuals who received FRR had been restrained in the 10 years (n = 536) or 5 years (n = 200) prior to FRR.

Table 4.

Arrests and protection order case involvement in the 10 and 5 years prior to FRR, 2017–2022 (n = 11,516)

	5 years prior to FRR n (%)	10 years prior to FRR n (%)
Arrests (mean, sd)	0.08 (0.41)	0.40 (1.11)
0 arrests	10,806 (93.8)	9126 (79.2)
1 arrests	554 (4.8)	1382 (12.0)
2 arrests	111 (1.0)	508 (4.4)
3 arrests	27 (0.2)	241 (2.1)
4 + arrests	18 (0.2)	259 (2.2)
One or more arrest was for a violent crime	253 (2.2)	807 (7.0)
Restrained by a protection order1 (mean, sd)	0.02 (0.15)	0.05 (0.26)
0 POs	11,316 (98.3)	10,980 (95.3)
1 PO	184 (1.6)	474 (4.1)
2 + POs	16 (0.1)	62 (0.5)

FRR: Firearm rights restoration; PO: Protection order.

Arrests are not included if the arrest was for the prohibiting crime itself.

¹A protection was ordered against the person who later received FRR (i.e., they were the respondent). This could include domestic violence protection order, anti-harassment order, sexual assault protection order, stalking protection order, extreme risk protection order, vulnerable adult protection order, or criminal no-contact order.

Discussion

To our knowledge, this was the first statewide study to empirically report on FRR among individuals who were convicted of a prohibiting crime. We found that > 11,500 people received FRR over the six-year study period (2017–2022). Understanding the population of FRR recipients in WA from 2017–2022, when FRR granting decisions were largely procedural (i.e., non-discretionary) can help policymakers both in WA and elsewhere learn about who might request FRR under similar state laws and circumstances. There has been renewed debate about the appropriateness of FRR policies, balancing the need to conservatively limit FRR to protect public safety with the need to re-instate rights for formerly convicted individuals so that they can re-integrate into society [46]. Our study suggests that variability within the population of prohibited individuals may be key to inform FRR eligibility criteria so that it is both specific and appropriate to meet each of these goals.

Our data suggest that 1,900 people received FRR each year on average in WA state. While we were unable to estimate the underlying number of people who were actively prohibited or were eligible for FRR, various sources suggest the number of annual convictions for prohibiting crimes in WA ranges somewhere between 13,000–25,000 if only considering felonies [47, 48], although to our knowledge there are no official numbers published that include prohibiting DV misdemeanor convictions. While this does not offer a perfect comparison, it provides some perspective on the relative magnitude of people who receive FRR, which is likely only a fraction of the total prohibited population. In the future, it would be important to study not only who received FRR as we did in the present study, but also to identify the population of individuals eligible for FRR, those who petitioned for FRR, and those who were denied. To facilitate further research, it would be important to enact and implement policies requiring systematic data collection for this information.

Our study did not have sufficient follow-up data to examine actions, violence, or behaviors that occurred after FRR, but a careful assessment of the legal, procedural, and public safety implications of our findings is warranted. We found that most FRR recipients were male, and a large proportion had a history of DV. While recidivism risk for perpetrating crimes tends to taper off over time [20, 45], the perpetration of violent crimes including DV may be more persistent throughout the life course [49, 50]. For that reason, it is possible that family members, current or former intimate partners, and children may be at the greatest risk of being harmed by individuals who received FRR.

One-half of FRR recipients had been convicted of a violent crime (ever) and many had multiple prior convictions. Only a minority of people had been convicted of all prohibiting crimes when they were still a juvenile, although about one-half of the sample were convicted of their most recent prohibiting crime before the age of 25. Heterogeneity within this population might indicate differing levels of risk for violence perpetration, information which could be used to guide FRR policy and eligibility considerations. For example, on average 15 years had elapsed between a person’s last prohibiting conviction and FRR, but about 10% of people in our sample received FRR within 7 years or sooner after their prohibiting conviction. While some states focus on distinctions between prohibiting felony or misdemeanor crimes when defining FRR eligibility [25], there is more variability and nuance in people’s criminal records which could additionally be considered when outlining FRR eligibility to ensure public safety (e.g., laws could define FRR eligibility based on how long someone has gone without committing a violent crime, or based on how many violent crimes they were convicted for).

In our sample, 200 individuals were restrained in a court-issued PO within 5 years prior to FRR. For these individuals, the courts found them to pose a credible threat of violence based on the PO process, which could include findings related to threatened or actual physical violence, sexual assault, stalking, or other interpersonal harm [51, 52]1. People who are subject to a PO often perpetrate serious and escalating abuse [53]. Data from North Carolina suggests that 23.1% of men subject to a PO for intimate partner violence have threatened their current or former intimate partner with a firearm as part of the incident that led to PO issuance [3]. Our findings about PO involvement prior to FRR are not an adequate proxy measure for ongoing or recent DV; many victim-survivors do not report their experiences to police or do not seek a PO due to fears of violence escalation, retaliation, or other barriers to engagement in the PO process [54]. As a result, our findings almost certainly underestimate the prevalence of DV perpetration in the 5 and 10 years prior to FRR.

People who are subject to a PO in WA temporarily lose their right to access or possess firearms, but that PO-specific prohibition goes away when the PO expires or is terminated (often within a year or two of issuance) [55]. Under 2017–2022 WA state law, prior POs were not considered in FRR case decisions unless the PO was active [31]. Whenever possible, it may be appropriate for state policy to encourage courts to consider whether a prohibited person was recently restrained by a PO to assess whether those individuals may still be perpetrating violence.

We found that mostly White individuals (84%) received FRR, with a smaller proportion of Black individuals 8.3% receiving FRR. Approximately 72% of felony convictions in WA occur each year among White individuals, whereas 13%% occur among Black individuals [47]. This could suggest that a relatively larger share of criminally prohibited White individuals receive FRR than criminally prohibited Black individuals. This is consistent with the larger body of legal system research showing that minoritized individuals have been over-policed and face disproportionate harm stemming from legal system involvement [56]. Prohibited individuals may face greater scrutiny from law enforcement and thus risk re-incarceration [57]; prohibited individuals can be charged for illegal firearm possession even in non-violent circumstances and in cases when they may inadvertently access another person’s firearms [56]. Possible differences in FRR receipt by race could be due to distinct barriers related to systematic racism, different preferences regarding firearms, eligibility statuses, willingness to re-engage with the court system for FRR, access to legal resources, among other explanations. It is important to note, however, that our data on race may have been ascertained by a court official, and information was missing on ethnicity, making it challenging to draw inferences from our findings. Without data or a proxy on the number and characteristics of individuals who lose their rights each year, the insights we can draw from our analyses on race and access to FRR are extremely limited. For example, we cannot know the scope of how many individuals qualified for FRR but have not received it or from what demographic groups they belonged.

WA state law on FRR changed in 2023 [31]. Most of these changes focused on clarifying legal ambiguities in FRR processes, adding new prohibitors, and streamlining administrative protocols [30]. Some changes were made to FRR eligibility and granting criteria, such as extending the 3-year crime-free period to 5-years for certain DV misdemeanor offenses [30]. With this policy landscape, we still believe our findings hold relevance given current WA policy and will be valuable for other states with similar FRR eligibility criteria.

We are hopeful that recent administrative changes implemented in 2023 regarding how FRR data are collected and stored in WA can help improve the accuracy and consistency of statewide FRR data, while also clarifying and streamlining the FRR petitioning process for prohibited individuals. The 2023 WA law called for the courts to take reasonable steps to notify the victim of the prohibiting crime about an FRR petition and to notify any person who has obtained a PO against the prohibited person [31]. This provides an important avenue for victims to become informed about the possibility that their perpetrator is re-arming themselves, which could have implications for their own safety and the safety of their families. In concert, the 2023 law introduced a mechanism for victims to provide a sworn written statement to the court regarding whether the prohibited person meets FRR eligibility criteria [31]. This gives voice to victims who may not yet have reported recent violence to the police. Victims may not report to police because they worry that the police cannot help them, they do not want to get the perpetrator in trouble, or the victim may fear they will not be taken seriously [58]. By providing an FRR-specific reporting mechanism, facts beyond the procedural WSP records check might be available to judicial officers to inform appropriate FRR granting.

Limitations

This study did not assess outcomes after FRR but instead provided a descriptive profile of individuals who received it—an essential first step for future research on policy impact and equity. It is challenging to put the number of individuals who received FRR into perspective, as the total number of people in WA who were criminally prohibited from purchasing or possessing firearms remains unknown. We could not confidently identify individuals with active prohibitions who did not have a recent FRR record. This limits our ability to compare characteristics of individuals (e.g. race, ethnicity, sex) who received FRR to their peers who also had a prohibiting conviction but did not receive FRR. Conviction dates were missing for approximately one-third of the sample, and most of the time, this was because their prohibiting criminal conviction had been vacated. People who had all or some of their convictions vacated may differ systematically from other people in our sample. When a crime is vacated in WA, information about the reason it was vacated (i.e., for good behavior; the statute was overturned) was not routinely recorded in centralized data systems. For that reason, we could not describe the context or reason behind why people’s crimes were vacated.

Instances of violence often go unreported and thus would not be reflected in arrest or PO data [58–60]. We did not have access to information about crimes, arrests, or POs from out-of-state. We also did not have access to information about other potential prohibitors related to civil commitment which could have impacted FRR eligibility. To our knowledge, no agencies in WA recorded comprehensive information about FRR applications or denials during our study period, data which would be important to collect in the future to assess potential barriers and inequities in FRR access and granting. We did not have information about FRR recipient’s motivation for FRR or their plans to acquire firearms. Interviewing FRR recipients as well as victims from prohibiting crimes is an important avenue for research moving forward to understand how this process impacts each of these parties.

Conclusion

There was considerable heterogeneity among FRR recipients in terms of the timing and nature of their prior criminal convictions, as well as in the five-year period leading up to FRR for prior arrests and for being restrained in a PO. These factors may be relevant to consider in future policy debates regarding FRR eligibility and granting criterion.

Author contributions

All authors made substantial contributions to the conceptualization and design of the study, as well as the interpretation of the data. All authors collaboratively obtained funding for the project. JK cleaned and analyzed the data. JK drafted the manuscript. Once the manuscript was drafted, all authors read and approved the final version.

Funding

This research was supported by a grant from the Fund for a Safer Future (Grant Number 019514–2022-12–12). The study was partly funded by the State of Washington. Partial support for computing servers and software in this research came from a Eunice Kennedy Shriver National Institute of Child Health and Human Development research infrastructure grant, P2C HD042828, to the Center for Studies in Demography & Ecology at the University of Washington and from the University of Washington College of Arts & Sciences, the Provost, eSciences Institute, the Evans School of Public Policy & Governance, College of Built Environment, School of Public Health, the Foster School of Business, and the School of Social Work. The views expressed in this manuscript are the authors’ alone and do not necessarily reflect the view of our funders or data providers.

Data availability

The data used in this research are restricted but can be accessed through formal requests to the Washington Administrative Office of the Courts and the Washington State Patrol.

Declarations

Ethics approval and consent to participate

This study was approved by the University of Washington Institutional Review Board in December 2022. The research involved analysis of administrative data and did not involve direct interaction with human subjects; therefore, informed consent was not required in accordance with IRB guidelines.

Consent for publication

The research involved analysis of administrative data and did not involve direct interaction with human subjects; therefore, consent for publication was not required in accordance with IRB guidelines.

Competing interests

The authors declare no competing interests.