Search tips
Search criteria 


Logo of jwhMary Ann Liebert, Inc.Mary Ann Liebert, Inc.JournalsSearchAlerts
Journal of Women's Health
J Womens Health (Larchmt). 2010 January; 19(1): 93–98.
PMCID: PMC3120094

Women with Protective Orders Report Failure to Remove Firearms from Their Abusive Partners: Results from an Exploratory Study

Daniel W. Webster, Sc.D., M.P.H.,corresponding author1 Shannon Frattaroli, Ph.D., M.P.H.,1 Jon S. Vernick, J.D., M.P.H.,1 Chris O'Sullivan, Ph.D.,2 Janice Roehl, Ph.D.,3 and Jacquelyn C. Campbell, Ph.D., R.N.4



The purpose of this study was to describe the perceptions of women who sought court protection orders for domestic violence (PODV) about actions to implement laws intended to disarm their abusers.


We identified female victims of intimate partner violence (IPV) in New York and Los Angeles primarily through family courts and records of police calls for domestic violence. Of these, 782 were surveyed and asked about their experiences seeking PODV from courts, judges ordering the removal of firearms from defendants, and if firearms were actually surrendered or confiscated.


Of the 542 victims who had obtained a PODV and knew whether their abuser owned a firearm, 82 (15%) reported that their abuser owned a firearm. Although state law either allowed or mandated judges issuing PODVs to require abusers to surrender their firearms, 21 victims (26%) reported that judges used this authority. Ten victims (12% of victims with armed abusers) reported that their abuser had either surrendered all of his firearms or had the firearms seized. When victims reported that the judge ordered their abuser to surrender his firearms, victims were more likely to report that all firearms were either surrendered by the abuser or confiscated by law enforcement.


Based on the perceptions of the IPV victims in this study, laws designed to disarm domestic violence offenders were either poorly implemented or failed to inform victims when their abuser's firearms were surrendered or confiscated.


Shootings are the most common method by which women are killed by an intimate partner in the United States. In 2005, 678 of the 1181 (57.4%) women killed by a boyfriend, spouse, or ex-spouse were killed with firearms.1 This is an undercount because FBI data do not include a category for ex-boyfriends. Children are also at risk of being killed with a gun in incidents of domestic violence.2 Perpetrators of intimate partner violence (IPV) also use firearms to threaten and intimidate their partners,3,4 and such threats are predictors of subsequent homicides.5 A study of risk factors for women being murdered by a current or former intimate partner after prior IPV found that the abusive partner's ownership of a firearm was associated with a 5-fold increased risk.5 A separate study of risk factors for women being murdered in their homes, primarily by current or former intimate partners, found that the presence of a gun in the home increased the risk of femicide 3-fold.6

In recognition of the inherent danger posed by IPV offenders with ready access to firearms, as of 2002, federal law and 24 states prohibited firearm possession by individuals who are subject to certain court orders of protection for victims of IPV.7 An evaluation of these laws found that firearm restrictions for defendants of protection orders for domestic violence (PODV) were associated with an 8% reduction in the rate of intimate partner homicide.7

In an attempt to ensure that IPV offenders are disarmed once prohibited from owning firearms, 16 states have passed laws that either allow or require judges issuing PODVs to order defendants to surrender any firearms in their possession.8 The impact of these laws is likely to depend on effective enforcement. A study of the implementation of two Maryland laws designed to disarm IPV offenders revealed several challenges. For example, some law enforcement officers reported they had little or no recourse if, upon serving an order with a firearm surrender provision, a respondent denied possessing any firearms.9

California law requires judges issuing most types of PODVs to order defendants to surrender their firearms. Sorenson and Shen10 examined implementation of this law by analyzing court administrative data for all PODVs in California in effect on June 6, 2003. Court documents indicated that 52% of the perpetrators were required to relinquish any firearm in their possession, and an additional 38% were prohibited from purchasing firearms. A subsequent study found that very few criminal justice agencies in California routinely confiscated firearms from PODV defendants when firearms were not voluntarily surrendered.11

Protective orders are often initiated by the victim and are intended to increase victim safety. However, there has been little research examining IPV victims' perceptions about whether judges issuing their protective orders included a provision for their abusers to surrender their firearms or whether the firearms were actually surrendered or confiscated by law enforcement. Moracco et al.12 surveyed victims receiving ex parte PODV before and after a new North Carolina law went into effect that required judges to ask plaintiffs seeking ex parte PODV if defendants have any firearms and required defendants to surrender any firearms in their possession within 24 hours of being served with the order. Forty-five percent of the plaintiffs seeking ex parte PODVs after the new law went into effect reported that judges asked them about the defendant's ownership of firearms, a proportion similar to that during the prelaw period. Of the victims who reported that their abuser's restraining order prohibited their abuser from possessing a firearm, 14% said sheriff's deputies confiscated the weapons and another 5% reported that the defendant voluntarily surrendered the firearms to authorities.12

Victims who know whether their protective orders require the abuser to surrender his firearms and whether the abuser was indeed disarmed should (1) be better positioned to advocate for their abuser's adherence to court-ordered firearm prohibitions and (2) be better able to respond if the system fails to disarm proscribed abusers. However, this is an area that has received little research attention. Given this gap in the literature, the primary objectives of the present study were to describe for a sample of women who sought a PODV the frequency with which they reported (1) requesting that judges order the PODV defendants to surrender their firearms, (2) that judges asked PODV defendants if they possessed firearms, (3) if judges ordered PODV defendants to surrender firearms as a condition of their protective orders, and (4) if their abusers were disarmed in accordance with the orders.

Materials and Methods

Study sample

The data for this study were gathered as part of a larger effort to evaluate methods for predicting the risk of repeat assault among IPV victims. Baseline interviews were conducted in 2002–2003 with a convenience sample of 1307 adult female victims of IPV recruited in New York City and Los Angeles County. Most participants were recruited from either the New York City Family Courts (n = 630) or from 911 calls to the Los Angeles Sheriff's Department for IPV incidents (n = 397). Additional participants were recruited from shelters for victims of domestic violence (n = 233), hospital emergency departments (n = 30), and a service agency for crime victims (n = 17). Additional details of the recruitment methods have been published previously.13

Laws to disarm domestic violence offenders

New York and California each have laws restricting firearm ownership by individuals subject to IPV protective orders. In California, except for some emergency orders, judges are required to order respondents (abusers) to both temporary and final orders to surrender any firearms in their possession. Judges in New York are required by state law to order abusers to surrender their firearms if the incident that prompted the protective order involved the use or threatened use of a deadly weapon or if the abuser had a prior felony conviction involving violence, stalking, or failure to obey prior protective orders.8 New York law also allows judges to order abusers to relinquish their firearms if there is a substantial risk the abuser might use a firearm against the victim.

Data collection and measures

Two thirds (867 of 1307) of the baseline interviews were conducted in person (almost all from New York City), and one third (440 of 1307) were by telephone (almost all from Los Angeles County). We used the same survey instrument for both in-person and telephone interviews. During the baseline interview, each study participant completed one of two longer risk assessment protocols (Danger Assessment14 or DV-MOSAIC15). Each of these protocols included a question about whether the abuser owns a gun. We were able to reinterview 782 (60%) of the participants by telephone an average of 8.8 months after the baseline interview. Items pertaining to abusers' surrender of firearms were included in the follow-up interview only. Women who obtained protective orders against their abusers were asked: (1) Did you or your attorney ask the court to have the police take [abuser's name]'s guns from him? (2) Did a judge order [abuser's name] to give his guns to the police or sheriff's department, or did the judge order the police or sheriff's department to take [abuser's name]'s guns from him? (3) Did he [abuser] give his guns to the police or sheriff's department or did the police or sheriff's department take his guns from him?

To ascertain abusers' gun ownership at the time of a protective order, we assumed that any abuser who owned a firearm at baseline also owned one at the time of the protective order. We also assumed that the abuser owned a firearm if the participant reported that she had asked the court to order the abuser to surrender his firearms at the time of the protective order or that the abuser surrendered a firearm after the order.

Data analysis

We calculated the prevalence of victims' reports of actions taken by the court to remove firearms from IPV offenders subject to protective orders and their perceptions of whether these firearm surrender provisions were carried out. We used Pearson's chi-square statistic to determine the statistical significance of bivariate associations between categorical variables.


Characteristics of study sample

Of the 782 women interviewed at follow-up, 595 (76%) reported ever obtaining a protective order against their abusive partner. Forty-three of these 595 reported they did not know whether or not their abuser had a firearm. Of the remaining 542, we identified 82 cases (29 of 146 from California and 53 of 398 from New York) in which there was a protective order involving an abuser with a firearm. Table 1 details selected characteristics of the women involved in these 82 cases compared with the 513 victims who obtained a protective order against abusers who victims reported did not possess guns. Demographically, both groups were similar with respect to marital status, presence of children in the home, race/ethnicity, education, employment status, and prevalence of experiencing severe physical abuse by a current or former partner. Victims with abusers who had firearms, however, were more likely to report no intimate or cohabitating relationship with the abuser at baseline (92.7% vs. 71.0%, p = 0.0001) and less likely to be foreign born (24.4% vs. 43.1%, p = 0.001). Approximately two thirds of the victims whose abusers had guns (56 of 82) experienced severe forms of IPV (e.g., being beaten up, attacked with a knife or gun, burned, strangled) in the 6 months before the baseline interview. Although the frequency of recent, severe abuse did not differ between victims who reported their abusers owned firearms and those who did not, victims whose abusers owned a firearm were more likely to have been victimized by a gun or knife and to have been victimized multiple times (Table 1).

Table 1.
Characteristics of Study Participants Who Obtained a Protective Order Against an Abusive Partner or Ex-partner at Baseline Interview (n = 595)

Victims' reports of firearm removal provision of protective order

Among the 82 cases in which the victim had obtained a protective order and reported that the abuser owned firearms, 37 women (45%) reported specifically asking the court during the protective order hearing to have their abusers' guns removed. Eighteen of these 37 respondents (49%) reported that the judge complied with their request, and 3 additional respondents reported that the judge ordered firearm removal without the victim requesting this relief. Thus, 26% (21 of 82) of victims whose abuser possessed a firearm reported that the judge ordered that these firearms be surrendered or removed from the abusers. Fourteen victims (17%) said that they did not know if the judge ordered the defendant to surrender his firearms. Victims were more likely to report judges ordering gun removal in Los Angeles (34%) than in New York City (21%). Victims were also more likely to report judges ordering firearm removal in cases in which the abuser had tried to kill the victim compared with less severe cases (35% vs. 18%, p = 0.073). However, prior threats or use of a weapon in abusive relationships were not associated with an increase in the likelihood that victims reported that judges ordered firearms removed (Table 2).

Table 2.
Hypothesized Correlates of Victim Reports of Judicial Orders for Abusers to Relinquish Firearms

Ten of the 82 (12%) victims with armed abusers subject to a protective order reported that their abusers either surrendered their firearms to authorities or had firearms confiscated. The likelihood of reported compliance with the firearm surrender provision was associated with whether the victim reported that the judge issued an order for firearm removal (likelihood ratio χ2 = 6.71, df = 2, p = 0.035). Among the 21 cases in which the victim reported that the judge had ordered firearm removal, 5 (24%) reported that all firearms were surrendered or confiscated, 5 did not know, and 11 (52%) reported that the abuser retained at least one firearm. Among the 61 participants who reported that the judge did not order firearm removal, 5 (8%) reported that all firearms were surrendered or removed.


California law mandates judges to include a firearm surrender provision in nonemergency domestic violence restraining orders. New York law requires judges to order firearm surrender if the incident prompting the protective order involved a firearm assault and allows (but does not require) judges to order firearm surrender if they deem a victim is at substantial risk of future gun assault. Yet in our sample, IPV victims from New York City and Los Angeles reported that judges issued orders for firearm surrender in only 26% of the cases involving protective orders against armed abusers. In some cases, victims reported that judges did not act despite their explicit request to have firearms removed.

There are many reasons why a judge might not order an IPV offender to surrender his firearms. Although New York and California laws have relatively broad inclusion criteria, some cases will not meet the legal requirements for judges to order firearm removal. In New York, judges have some discretion about when to order firearm removal in cases where guns were not part of the abuse. A recent study of court records in California found that about half of all PODV in the state included an order for the respondent to surrender any firearms in his possession.10 Forty percent of our sample was drawn from Los Angeles County, and 34% of those respondents reported that the judge had ordered their abusers to surrender firearms.

The difference between the findings of this study and those found in the study of California court records10 could be partly attributable to our relatively small sample drawn from a single jurisdiction in California compared with the prior analysis of the entire state. But our findings are more likely due to the different measures used in the two studies and to slightly different research questions being examined. In responding to interview questions about whether the judge ordered firearm surrender or confiscation, some respondents may focus on what a judge said at the hearing or what they recall about the protective order rather than on what was written in the order. Firearm surrender provisions for restraining orders in California were a standard condition that judges could apply by checking a box on the order.8 (These check boxes have since been eliminated, and firearm prohibition language is now a standard part of all California's PODV forms.) However, judges may not verbalize every condition of protective orders when explaining their decisions; thus, victims may not always know when judges check a box indicating that the abuser is not permitted to own firearms unless the victim carefully reads the protective order form itself.

Sorenson and Shen10 examined how commonly judges checked defendant firearm prohibition boxes on the protection order forms. These designations have important legal consequences; however, legal restrictions concerning an abuser's possession of firearms may not result in the intended response if the judge does not verbally order the abuser to surrender his firearms. This is what was found in a recent evaluation of a new North Carolina law designed to mandate judicial actions to disarm PODV defendants.12 Furthermore, because law enforcement agencies often do not take proactive steps to ensure that abusers have relinquished their firearms,9,11 victims' knowledge of their abusers' firearms restrictions contained in orders of protection is necessary if victims or their advocates want to press law enforcement agencies to confiscate abusers' firearms.

We found that fewer than half of the victims who obtained protective orders against armed abusers affirmatively asked judges to order their abusers' firearms removed. There are several reasons why a victim might not request gun removal from her abuser. The circumstances of some of the cases may not have permitted court action to remove firearms. In addition, some victims may not feel it necessary to request firearm removal if they believe judges will take such action on their own. Some IPV victims believe removing guns from their abusers will increase the likelihood and severity of retaliation, or they think that such action is futile because it is relatively easy for their abusers to obtain another gun.9 Despite information contained on the PODV form itself, some women may have been unaware that judges had the power to order firearm removal from their abuser. Further research is needed that explores victims' knowledge and beliefs concerning abusers' access to firearms and strategies for disarming them.

Relatively few (12%) participants seeking orders of protection against abusers who owned a firearm believed that the court order resulted in the removal of all of the abusers' firearms. Because few victims in our study continued to live with their abusers, they may not know for certain if their abuser surrendered his firearms to authorities. Of course, victims who are threatened or abused with a firearm while a protective order is in effect need not be living with their abuser to accurately report on the failed implementation of the firearm surrender provision. In addition, our results indicate that many victims may simply be uncertain about whether guns were removed, an uncertainty that can profoundly affect women's safety and sense of well-being. Despite these caveats, our findings indicating significant gaps in the enforcement of firearms surrender conditions of domestic violence restraining orders is consistent with other studies using other types of data9,11 and similar to a recent study that also used data from interviews of victims.12

Changes to existing firearm removal policies could facilitate more effective disarming of batterers. For example, states could follow California's approach and require, rather than merely allow, judges to order firearm removal from IPV offenders. Victim reports of judicial orders for firearm removal were somewhat more common among women living in Los Angeles, where state law requires judges to include firearms prohibitions in court PODVs than among women living in New York City, which only requires firearm prohibitions in orders of protection in more narrow circumstances. These differences in victim's perceptions by state (although not statistically significant in our relatively small sample) may reflect actual differences in judicial use of this removal authority. In states that allow judicial discretion for ordering firearm removal,8 it may be important to educate judges about the substantial increase in the risk of lethal violence when abusers have access to firearms.5 This conclusion is reinforced by our data indicating that victims reported that judges were not more likely to order removal where there had been a prior threat with a weapon against the victim.

IPV victims and their advocates have a role and an interest in improving the implementation of firearm restrictions in protective orders. Victims and advocates can encourage law enforcement to follow up when firearms surrender orders are issued to enhance compliance with the orders and attempt to hold law enforcement accountable if they do not act to ensure that defendants have been disarmed. Medical practitioners screening women for IPV should also be aware of the risks associated with an abuser's firearm possession and legal options available to women to remove that firearm.

This study is subject to certain limitations. First, our data are based on a relatively small sample drawn from two urban areas. Whether the findings are generalizable to other settings is unknown; however, our findings are consistent with those of a similar study in North Carolina.12 Differences in sample recruitment between New York and Los Angeles may also have influenced some study findings. Second, our data are based on victims' self-report and, as with all such data, are subject to recall biases. Some victims may not know of or be able to recall all restrictions imposed by the orders including firearm restrictions. As discussed, however, what victims believe to be the case may be critically important. Nevertheless, further research is needed that combines victim report of abuser firearm ownership with police or court records that show evidence of the surrender or confiscation of firearms from proscribed IPV offenders. Third, we did not always have data that definitively indicated whether an abuser was in possession of a firearm when a protective order was issued. Some abusers had multiple arrests for IPV and had been subject to multiple protective orders, some temporary and some long-term. We did not ask about the timing of each of these events, which may have taken place over many years, and our data indicate that abusers' firearm ownership can change over time. Finally, some abusers may have sold their firearms to comply with the PODV; some women may not have appreciated this distinction in responding to our interview questions about surrender of firearms to law enforcement.

Despite these limitations, this study fills a void in the literature about the implementation of laws designed to disarm IPV offenders. Our and others' findings suggest that the courts and law enforcement agencies are failing some women. Although California and New York laws mandating or permitting judges to order firearm removal from IPV offenders are more comprehensive than most states' laws of this type, our findings suggest that there are important gaps in enforcement that should be closed to protect IPV victims from severe injury and death by armed abusers. Enforcement might also be improved if efforts were made to be sure that women know the firearm-related provision of their protective orders and if they have, in fact, been implemented.


Data collection for this study was funded in by grant 2000WTVX0011 from the National Institute of Justice to the Johns Hopkins School of Nursing. Data analysis and manuscript preparation were funded by a grant from The Joyce Foundation to the Center for Gun Policy and Research, Johns Hopkins Bloomberg School of Public Health.

Disclosure Statement

The authors have no conflicts of interest to report.


1. Fox JA. Zawitz MW. Homicide trends in the U.S.: Intimate homicides. Washington, DC: Bureau of Justice Statistics, Office of Justice Programs, U.S. Department of Justice; 2007.
2. Websdale N. Understanding domestic homicide. Boston: Northeastern University Press; 1999.
3. Tjaden P. Thoennes N. Prevalence, incidence and consequences of violence against women: Findings from the National Violence Against Women Survey. Research in Brief from National Institute of Justice and the Centers for Disease Control and Prevention. Washington, DC: U. S. Department of Justice; 1998.
4. Sorenson SB. Wiebe DJ. Weapons in the lives of battered women. Am J Public Health. 2004;94:1412–1417. [PubMed]
5. Campbell JC. Webster D. Koziol-McLain J, et al. Risk factors for femicide in abusive relationships: Results from a mulitsite case control study. Am J Public Health. 2003;93:1089–1097. [PubMed]
6. Bailey JE. Kellermann AL. Somes GW. Banton JG. Rivara FP. Rushforth FP. Risk factors for violent death of women in the home. Arch Intern Med. 1997;157:777–782. [PubMed]
7. Vigdor ER. Mercy JA. Do laws restricting access to firearms by domestic violence offenders prevent intimate partner homicide? Eval Rev. 2006;30:313–346. [PubMed]
8. Frattaroli S. Vernick JS. Separating batterers and guns: A review and analysis of gun removal laws in 50 states. Eval Rev. 2006;30:296–312. [PubMed]
9. Frattaroli S. Teret SP. Understanding and informing policy implementation: A case study of the Domestic Violence Provisions of the Maryland Gun Violence Act of 1996. Eval Rev. 2006;30:347–360. [PubMed]
10. Sorenson SB. Shen H. Restraining orders in California: A look at statewide data. Violence Women. 2005;11:912–923. [PubMed]
11. Seave PL. Disarming batterers through restraining orders: The promise and the reality in California. Eval Rev. 2006;30:245–265. [PubMed]
12. Moracco KE. Clark KA. Espersen C. Bowling JM. Preventing firearm violence among victims of intimate partner violence: An evaluation of a new North Carolina law. Final report submitted to the National Institute of Justice by Pacific Institute for Research and Evaluation. Chapel Hill; NC: Sep, 2006.
13. Campbell JC. Webster DW. O'Sullivan C, et al. Intimate Violence Risk Assessment Validation Study. Final report submitted to the National Institute of Justice. Sep, 2005.
14. Campbell JC. Assessing dangerousness. Newbury Park, CA: Sage; 1995.
15. Gavin de Becker Associates. Domestic violence method (DV-MOSAIC) Studio City, CA: Author; [Jul 20;2006 ].

Articles from Journal of Women's Health are provided here courtesy of Mary Ann Liebert, Inc.