Firearm locking device preferences among firearm owners in the USA: a systematic review

Abstract

Background

Preventing firearm-involved injuries is a critical public health priority. Firearm locking devices can prevent firearm injuries, such as suicide and unintentional shootings, as well as theft. Various firearm locking devices exist; however, little is known about firearm owners’ preferred locking devices for secure firearm storage. In this systematic review, we examined existing literature on preferred locking devices for secure storage of personal firearms among United States (US) firearm owners with the purpose of understanding practical implications and needs for future research.

Methods

We searched 8 major databases, as well as the grey literature, for English-language sources published on or before January 24, 2023, that empirically examined firearm locking device preferences. Following PRISMA guidelines, coders independently screened and reviewed 797 sources using pre-determined criteria. Overall, 38 records met inclusion criteria and were included in this review.

Results

The majority of studies measure and report on participant use of various types of locking devices, but few go on to measure preference between device options and the attributes and features that may contribute to an individual’s preference. Included studies suggest that a preference for larger devices, such as lockboxes and gun safes, may exist among US firearm owners.

Conclusions

Review of included studies suggests that current prevention efforts may not be aligned with firearm owners’ preferences. Additionally, findings from this systematic review emphasize the need for additional methodological rigorous research to understand firearm locking device preferences. Expanded knowledge in this area will result in actionable data and foundational best practices for programming that encourages behavior change concerning secure storage of personal firearms to prevent injury and death.

Background

In the USA, firearm injuries, spanning suicide, homicide, interpersonal violence, and unintentional shootings, are a major public health concern. Each year, over 45,000 individuals in the USA die due to firearm injuries, and tens of thousands more experience nonfatal injuries annually (Aitken et al. 2020; National Center for Injury Prevention and Control 2023; Rees et al. 2022). Multiple studies have demonstrated that when a firearm is stored securely (e.g., locked, unloaded, and separate from ammunition), risk for firearm suicide and other forms of firearm-involved injuries may be reduced (Monuteaux et al. 2019; Shenassa et al. 2004). Thus, to reduce the risk of firearm injuries, key stakeholders—including medical organizations (Butkus et al. 2018; McLean et al. 2019; Bulger et al. 2019), suicide prevention organizations (American Foundation for Suicide Prevention 2022), leaders in the firearm-owning community (National Shooting Sports Foundation 2022a), and other groups—encourage secure firearm storage practices.

An estimated 30% of US adults own one or more firearms and an additional 11% do not personally own a firearm but live with someone who does (Parker et al. 2017). One prominent approach to secure firearm storage is the use of firearm locking devices. While an estimated 36% of US firearm owners store all of their firearms locked (Parker et al. 2017), there are various types of firearm locking devices available to firearm owners, spanning keyed cable locks (which commonly rely on a steel cable) to biometric safes (which use biological data, such as fingerprints, unique to authorized users) (National Shooting Sports Foundation 2021). Prior research examining firearm locking device use has not considered the heterogeneity in locking device preferences.

Promisingly, interventions that distribute firearm locking devices increase end users’ secure firearm storage practices (Anestis et al. 2021d; Roszko et al. 2016). Work to date has largely focused on the distribution of cable locks, a relatively low-cost option, although there may be variations in firearm owners’ preferences for certain types of firearm locking devices (Stuber et al. 2021). Interventions that are attentive to firearm owners’ preferences may have the potential to increase their receptivity to recommendations to store firearms securely.

In this systematic review, we sought to examine published studies on firearm locking device preferences among US firearm owners to help inform public health and clinical decision-making regarding the distribution of firearm locking devices. In addition to examining studies specifically focused on firearm locking device preference, we also included studies examining firearm owners’ current use of specific locking devices (e.g., cable lock vs. gun safe), as one’s current choice to use a device may be a proxy, albeit imperfect, for current preferences. We also sought to identify gaps in the existing literature and present recommendations for future research in this area.

Methods

Data sources and search strategy¹

Our search strategy was developed and implemented by a health sciences librarian (Rethlefsen et al. 2015). Records were eligible for inclusion if they were: (1) published in English; (2) empirically (quantitatively or qualitatively) examining US firearm owners’ firearm locking devices preferences; and (3) peer-reviewed studies, dissertations, and non-peer-reviewed publications/organizational reports or presentations. Studies had to be US-based. No limits on publication date were used. Conference abstracts and proceedings were excluded in the search strategies for databases with high conference proceedings coverage (PsycINFO, Sociological Abstracts, Web of Science). We queried 8 databases with an updated search on January 24, 2023: Ovid MEDLINE, Web of Science, PsycINFO, Public Affairs Information Service Index, Sociological Abstracts, Social Sciences Full Text, ProQuest Dissertations & Theses A&I, and Google Scholar (first 100 citations). The title, abstract, and subject headings for select databases (Ovid MEDLINE, PsycINFO) were searched for key vocabulary. See Table 1 for full search strategy and terms for Ovid MEDLINE.

Table 1 Search strategy

Additionally, we conducted a search for grey literature via custom, advanced Google searches developed using similar strategies. Our grey literature search included targeted searches of the organizational websites listed in Table 2, as well as several search strings, which were iteratively defined and developed by the review team.

Table 2 Grey literature organizations searched

Study selection

Study selection methods and procedures followed PRISMA guidelines. The initial database search yielded 1316 citations. Records were de-duplicated for identical citations and organized using the citation management software Endnote version 20 (Clarivate). After de-duplication, the remaining 797 records were uploaded to Covidence, a systematic review citation screening software, which identified three additional duplicates—leaving 794 for title and abstract screening.

Three reviewers—a trained Research Assistant with a Bachelor of Public Health and two trained undergraduate student supervisees—independently screened the titles and abstracts of all 794 records. The reviewers were supervised by senior members of the Review Team, including the Project Manager and the Principal Investigator who both have extensive experience in firearm injury prevention research and clinical and public health programmatic efforts. Following screening, a full-text review of the 231 remaining records was conducted by the same three reviewers. Each record was independently reviewed by at least two reviewers. Any discrepancies across reviewers were discussed with the larger Review Team and resolved with consensus. Following the full-text review, 37 records met inclusion criteria. The search of the grey literature resulted in the inclusion of one additional record that met eligibility criteria. Eligibility of grey literature was determined first by a title screen, followed by a full-text review conducted by two independent reviewers. Upon conclusion of the full-text review and grey literature search, 38 records were eligible for data collection (Fig. 1).

Fig. 1

PRISMA diagram

Data collection and quality assessment

The Review Team collaborated to develop and refine a data extraction form within Covidence. Two team members—the Project Manager and the Research Assistant—independently extracted the following data from each included record: citation, year, location, study setting, participant characteristics, study aims, methodology, and reported outcomes related to firearm locking device use, preference, and willingness to pay for devices. Data collection was conducted using Covidence. Any disagreements between the reviewers were resolved by consensus of the team members.

To determine if the identified studies were of sufficient methodological quality, we used the Critical Appraisal Skills Programme (CASP), a tool frequently used to assess the quality, utility, and relevance of studies (CASP 2023). Using this 10-item metric, the Review Team methodically examined each included article across various study domains (clear methodological aims and approach, appropriate research design and participants, data collection and interpretation, relevance, and utility of findings), providing a “Yes,” “No,” or “Can’t Tell” response to each item. All Review Team members used a structured methodological approach for assessing selected articles with this tool. Disagreements in assessment were resolved via discussion between reviewers and it was determined that only studies scoring a 7/10 or higher were to be included. All 38 studies were deemed to be of sufficient methodological quality and therefore met the threshold to be included.

Data synthesis and analysis

Due to the extent of differences across included studies in methodology, design, and outcomes, no meta-analyses were feasible. Instead, the study team chose to report findings as a narrative synthesis, summarizing and explaining the characteristics and results of the included studies pertaining to the primary outcomes of this review.

Results

We identified a total of 38 studies that reported on firearm locking device preferences among US firearm owners.² Included articles were published from 1996 to 2022. Two included studies examined the same dataset. Sample sizes ranged from 16 to 6404. Table 3 provides a list of all included studies, their reported aims, sample size, participants, setting, methods, types of firearm locking devices discussed, and whether devices were provided. In this section, we summarize the findings in the literature regarding (1) preferences for specific firearm locking devices; (2) device attributes and features contributing to preferences; (3) current use of specific firearm locking devices among firearm owners; and (4) firearm owners’ willingness to pay for devices.

Table 3 Included Records (n = 38)

What preferences for firearm locking devices exist among firearm owners in the USA?

Eight studies reported data that suggest which firearm locking devices firearm owners may prefer. In three studies, participants were offered a free locking device and allowed to pick between lockboxes and trigger locks. Lockboxes were chosen by more participants in each study compared to trigger locks (82–18% of firearm owners Uspal et al. 2021; 87–12% of all participants Simonetti et al. 2018b; 89–8.5% of firearm owners King et al. 2020), two of these studies—both with at least 90% of sample reporting firearm ownership—also reported less than 2% of all participants having “no preference” between the options (Simonetti et al. 2018b; King et al. 2020). Barber et al. (2022) also reported on participant selection and eventual use of either cable locks or lock boxes. Among the sample of parents whose child (10–17) was being evaluated in an emergency department for a suicide-related or behavioral health-related problem, fewer opted to receive offered cable locks (65%) compared to offered lockboxes (70%). A similar pattern emerged at follow-up, with more firearm-owning participants reporting use of the provided lock box (28%) compared to use of the provided cable locks (14%).

A study of 1292 families with children ages 2 to 18 years measured the use of coupons provided to participants to purchase subsidized devices (Grossman et al. 2000). Intervention participants who reported owning firearms received a coupon to purchase a lock box (median redemption price $9.99, range $9.99–$45.00; average retail price ~$70.00) and one to purchase a trigger lock (median redemption price $0.00, range $0.00–$5.00, average retail price ~$10.00). More coupons were used to purchase lock boxes (8.4%, n = 26/309) than trigger locks (4.9%, n = 15/309). In a study of 401 community-based firearm safety event attendees, Simonetti et al. (2019) found that a greater proportion of firearm-owning participants indicated they would never use a trigger lock, cable lock, or clamshell device compared with a lock box or gun safe. These studies may suggest a preference for larger. More expensive devices (e.g., safes and lockboxes) compared with cable and trigger locks, which are smaller in size and often cheaper in cost.

What attributes or features contribute to firearm owners’ preferences for various types of locking devices?

Ten studies reported on the device features and/or attributes that may influence firearm owners’ preference for and use of various devices. Several of these studies collected information from participants not on reasons for one device versus another, but rather on overarching features and attributes influencing the use of any locking device. These findings are consistent with previous research on motivations to use locking devices (Thomas et al. 30,24,; Hamilton et al. 2018; Crifasi et al. 2018) and emphasize the barriers to using a variety of locking devices among firearm owners who choose to own firearms for self and household protection (Warner 32,27,; Cao et al. 1997; Schenck et al. 2022). One study conducted with 147 firearm-owning parents and child caregivers reported 75% of participants indicated both the speed of being able to unlock and lock a device and being able to keep the firearm loaded when locked as “absolutely essential” features (Dennis et al. 2019). Simonetti et al. (2019) found that over 80% of community-based firearm safety event attendees with firearms in their homes reported the same features to be “very important” or “absolutely important”. This was supported by several additional studies that employed qualitative methodologies to collect information from participants, reporting hesitancy to use lock boxes and trigger locks due to delayed access in the event of a home invasion. Schenck et al. (2022) quoted one participant, “If someone’s in your house, you have literally seconds before they’re right there in your face. So, you have to find the key, get to the box, then you got to get to the ammo, unlock it, put it all together, I’m already dead at that point.”

No studies reported on the use of biometric devices, but one study of 1444 firearm owners reported on participants’ hesitancy to use biometric devices in place of more traditional devices. Biometric devices are commonly suggested as a solution to quick access concerns, but concerns noted in this study included vulnerability to hacking and the potential that the technology would fail or malfunction when needed (Crifasi et al. 2019). Cost was also discussed across several studies as a barrier to use of larger devices (e.g., gun safes) and/or biometric devices.

A study of 75 law enforcement officers with issued firearms reported on unfavorable attributes specific to cable locks that include a key. Officers expressed worry about losing the key and damage to or deterioration of the key mechanism (Coyne-Beasley and Johnson 2001a). A study conducted with 40 Alaskan firearm-owning households reported unfavorable features of trigger locks, with the most common reason for not using trigger locks being that they were “inconvenient” (27% of participants) (Horn et al. 2003). Several studies also reported motivating attributes including a device’s ability to be used for both handguns and long guns, ease of transfer (e.g., between vehicle and home), and ease of installation and use.

What firearm locking devices are used by firearm owners in the US?

Most (76.3%; 29/38) studies reported on participant use of locking devices at the time of study involvement, which we included because one’s current choice to use a device may be a proxy for current preferences. Across studies, the firearm locking devices reported on varied, as did the labels and descriptions used to define devices. Table 3 lists these locking device categories measured in each study.

There were notable differences across studies in study design, procedures, measures, participant inclusion criteria, and sample size. To facilitate comparisons with cross-sectional observational studies, we focused on the baseline proportions of firearm locking device use reported for studies that involved an intervention and/or a follow-up component in this review. Most studies (86.2%; 25/29) collected information on locking device use via self-report surveys. Eleven studies collected survey data from general populations of adults, with sample sizes ranging from 30 to 6,404. The use of gun safes was the most reported, with representation in nine survey studies of general adult populations, followed by lock boxes (n = 7), trigger locks (n = 6), cable locks (n = 4), and gun cabinets (n = 3). Only one study reported on the use of clamshell devices. No studies reported on the use of in-vehicle locks or biometric devices. Studies that reported on current device use only among firearm owning participants, showed the following ranges of use by device type: cable lock 18.7–29.2%, trigger lock 16.3–21.4%, gun safe 25.5–52%, and lockbox 6–20.1%.

Nine survey studies required participants to be parents or guardians of or to live in a household with children under the age of 18. Sample sizes across these studies varied (range 50−6990) and included both firearm owners and non-firearm owners, as did the devices included. The devices most asked about (4 studies) included gun safes (reported use ranging from 14.8 to 54.5%) and trigger locks (9–48.5%). The use of lockboxes (three studies: 9.2–48.8%), cable locks (two studies: 11% and 16.8%) and gun cabinets (two studies: 19.7–28%) were measured less often. Two studies included multiple devices in the same response option (e.g., safe or trigger lock), making it unclear which device participants were actually using (Aitken et al. 2020; Carbone et al. 2005).

What are firearm owners willing to pay for various firearm locking devices?

Six studies explored the cost of firearm locking devices and the role cost played in participants’ use of various devices. One study conducted with firearm-owning parents and caregivers of children under the age of 18 found that gun safes were often seen as too expensive to independently buy and use (Aitken et al. 2020; DeMello et al. 2020). In Simonetti et al.’s (2019) study of 401 community-based firearm safety event attendees, participants reported how important various features of locking devices were to them. 22% of firearm-owning participants felt device cost being less than $15 was “very important” or “absolutely essential”. Finally, a survey among 147 Oregon-based current and soon-to-be parents explored which devices firearm owners would prefer if “money was not an issue”. Options included a cable lock, life jacket locking device, lockbox with keyed access, quick access electronic lockbox, and biometric lockbox. 54% of respondents selected a biometric device as their first choice and 20% selected a gun lockbox with electronic keypad access. The majority of participants reported a gun lockbox with electronic keypad access as their second choice. Cable locks were reported overall as the least favorable choice (Dennis et al. 2019).

Included studies also explored the provision of free and/or subsidized devices. One study of 164 parents or caregivers of pediatric patients reported that participants were more likely to accept a locking device that was free, compared with a locking device that was available at a reduced cost (91% vs. 52%) (Uspal et al. 2021). Overall, 12 included studies provided at least one type of free locking device: Six provided trigger locks, four cable locks, three lock boxes, and three gun safes or cabinets. One additional study provided free devices but did not specify which device type was given.

Discussion

Reducing firearm-involved injury and death will take a multilayered, community-engaged approach. In recent years, there has been increasingly more outreach to and leadership from firearm owners and the firearm industry to promote secure firearm storage (National Shooting Sports Foundation 2022b). Expansion of efforts to understand locking device preferences will provide practitioners, policymakers, and other stakeholders with useful insight on how to design effective safety interventions.

Review of the included studies suggests that current prevention efforts that employ the provision of cable and trigger locks—locking devices that are generally smaller and less expensive—may not be aligned with what firearm owners prefer. In fact, the review of reported outcomes indicates that a preference for larger devices, such as lockboxes and gun safes, may exist. Feasibility and scalability require a balanced consideration of cost and preference to ensure optimal implementation of interventions. In this case, a clearer understanding is needed regarding the proportion of firearm owners that would adopt secure firearm practices if their preferred—but more expensive—storage devices were made readily available. Such work would be useful in determining the degree to which preference findings should influence device distribution strategies. Future research might consider using customer-value-based pricing questionnaires (Garrison and Towse 2017) that enable an understanding of the price point at which the cost of specific firearm storage devices influences the likelihood that firearm owners would purchase and use specific devices. Such information could help establish not only a sense of the market for specific devices but could also enable cost-benefit analyses that aid in determining what types of devices specific outreach programs might opt to offer. Additionally, it is important to note that preference may be influenced more so by locking mechanism (e.g., key, combination lock, etc.) than by size and price, but the included studies did not include information regarding mechanism.

Our focus on preference for specific locking devices builds upon the growing body of research that seeks to understand motivations and barriers to the practice of storing and staging firearms locked. One theme that aligned with previous research is the prominence of firearms being kept unlocked to increase the speed of access in case of self-defense (Warner 2022; Cao et al. 1997). Individuals who are motivated to own firearms for self and home protection reportedly see locking devices in opposition to this motivation. Indeed, included studies reported participants using locking devices on some personally owned firearms while always leaving one or more unlocked. A possible solution may be biometric devices, which allow for quick access and prohibit unauthorized access and use. However, this systematic review revealed that little is known about firearm owners’ preferences for biometric devices over traditional devices. In fact, the few and limited studies that explored biometric devices with participants reported features and attributes that may make them undesirable for firearm owners, although additional research is needed. As research in this area moves forward, it will be important to understand the potential intersection between reason for firearm ownership (e.g., self- or home-defense) and specific firearm locking device preferences. To be clear, the findings from this systematic review are not dispositive regarding which firearm owners prefer which firearm locking devices and the associated characteristics thereof. Until more representative research can be conducted, practitioners may consider offering multiple device options for free or at a reduced cost.

There exists a variety of factors that may influence an individual firearm owner’s preferences for firearm locking devices (Hamilton et al. 2018; Crifasi et al. 2018; Ramchand 2022). These factors include motivations for ownership, number and type of firearms owned, household makeup, neighborhood characteristics, and more. Interventions that acknowledge the breadth of individual factors by allowing owners to choose their preferred device may be the most effective.

Limitations and future directions: extant literature

Limitations of the extant literature include the limited number of studies and extent of differences in methodology, design, and outcomes did not allow for an analysis past a narrative synthesis. Because of this, our ability to compare the results of one study to another or to draw detailed conclusions that are likely to reflect the bulk of US firearm owners is severely limited. Future research can address this in multiple ways. First, efforts should be made to recruit large, representative samples of specific communities of firearm owners, allowing the research to accurately reflect the diverse array of firearm-owning communities (Thomas et al. 2022) and to highlight any differences that may emerge regarding storage preferences. This type of nuanced and generalizable understanding would facilitate optimization of resource distribution. Second, the research community should develop preferred standards for assessment methods that enable easier comparison of results and future meta-analytic consideration of these questions. Like any area in which limited research has been conducted, it is unsurprising to find substantial variation in how questions are framed and what information is assessed, but as the field advances, more consistent operationalization will be vital. Third, we were also unable to assess the extent to which device preferences vary by type of firearm owned (e.g., handgun vs. long gun) and reason for ownership. It may be that individuals have varying preferences based upon the degree to which they desire quick easy access (e.g., for a home protection firearm vs. one used primarily for hunting). This information will provide enhanced recommendations for practitioners and policymakers pertaining to the most effective approaches to device provision as part of lethal means safety effort. If, for instance, evidence emerges demonstrating that preferences for specific locking devices vary based on the extent to which firearm owners more readily envision themselves using the firearm to fend off an intruder than someone using it for another reason more likely to cause harm to the firearm owner or other household residents (e.g., suicide, unintentional shootings), this would speak to the need for increasing awareness about the actual risks for specific firearm-related outcomes in the home, particularly if further data indicate specific locking devices are more effective at preventing such outcomes. Fourth, it is vital that researchers systematically collect data on storage preferences across diverse samples, thereby clarifying if and how locking device preferences differ across demographics (e.g., race, gender, parenting status, geographic location). Fifth, additional research should examine the extent to which firearm owners’ current firearm storage practices align with their preferences for specific firearm locking devices and identify reasons for potential discrepancies. Sixth, going forward researchers should assess whether preferences for locking devices differ between firearm owners who do and do not currently use locking devices and between firearm owners who lock all their firearms relative to those who keep at least one firearm unlocked. The existence of any such differences is not clear based on the current literature and, in fact, no such differences may emerge following direct assessment; however, efforts to promote the use of locking devices focus specifically on those who do not currently lock their firearms, so understanding variation in preferences will be vital to the success of such campaigns.

Limitations: current systematic review

There are also limitations of the current systematic review that are important to note. First, we only included studies that were published in English; although the focus was on the USA, it remains possible that non-English speaking reports exist. Second, although our search was broad—across 8 databases—and conducted in consultation with a health sciences librarian, we did not examine every possible database; thus, it is possible that some studies may have been missed by this search strategy. Third, conference abstracts and proceedings were excluded in the search strategy from a subset of the databases, potentially missing studies that have not been submitted to or which were triaged in peer-reviewed journals. Additionally, although we developed our search terms collaboratively within a team of researchers with extensive experience in this area, it remains possible that we chose suboptimal terms and, because of this, our search did not reveal other instances of extant relevant research. Finally, we did not comprehensively examine industry and consumer data (e.g., purchasing patterns), which might provide critical insights into firearm locking device preferences. An additional important consideration is that our discussion of firearm storage practices—a variable related to but not synonymous with firearm storage preferences—is not based on a systematic review of the firearm storage practices literature. Our findings are restricted to results presented in studies that also report explicitly on storage preferences and, as such, numerous other studies that report firearm storage practices are not represented in our findings. Future efforts may be successful in accessing this information if partnerships with the firearm industry are cultivated and nationwide purchasing data are made more readily available for research purposes.

Conclusions

This systematic review provides important information and identifies knowledge gaps for future work. The findings from 38 total studies provide an initial summary of what data have previously been collected from firearm owners—including that firearm owners may prefer lock boxes or safes to cable locks, and that cost and access (speed and reliability) are concerns. More importantly, this review emphasizes the need for additional research to understand the topic and improve firearm-involved injury prevention efforts that involve the provision of free or reduced-cost locking devices. Until additional research can be conducted, practitioners should provide multiple device options for free or at a reduced cost to firearm-owning individuals and communities, as this may address the various factors that influence an individual’s decision to own firearms and, therefore, their preference for which locking device(s) to use on their weapon(s).