“What Will Happen If I Say Yes?” Perspectives on a Standardized Firearm Access Question Among Adults With Depressive Symptoms

Firearms are the most commonly used method in suicide deaths in the United States (1, 2). In the past 10 years, the age-adjusted rate of firearm suicide deaths increased 19%, from 5.8 to 6.9 deaths per 100,000 (3). Risk of suicide has increased among adults with access to firearms (4, 5), particularly when the firearms are stored loaded and unlocked (6, 7). Interventions to improve safe storage of firearms and ammunition can be effective (8–10), particularly counseling-based firearm safety interventions delivered by a health care provider (11–13). Thus, collaborative patient-provider discussions about lethal-means safety (e.g., firearms and medications) are a standard component of recommended care for patients identified as being at risk of suicide (14, 15).

Despite the potential benefits of counseling-based interventions (16), national debate remains about whether health care providers should routinely ask adult patients about access to firearms (17). For example, Florida legislation (later overturned by courts) prohibited these discussions when unnecessary for treatment (18, 19), and other states regulate how health care organizations collect and store firearm access information (20). No federal law or statute prohibits providers from asking about firearms when the information is relevant to patients’ health (21), but physician leaders have expressed fears that regulations, both proposed and enacted, may inhibit patient-provider conversations about firearms (19, 20). Moreover, there are no national clinical guidelines describing standard practice for firearm screening (22).

A nuanced understanding of patient perceptions is essential for developing patient-centered approaches to assessing firearm access, but only a few studies have addressed this topic, and their findings have been heterogeneous. A representative survey of U.S. adults found that about half believed it was sometimes appropriate for providers to discuss firearms with their patients (23), and a survey of 660 veterans receiving mental health care found that among those who reported firearm access, 92% endorsed health system interventions to address firearm access during high-risk periods (24). A qualitative analysis found that 550 Internet commenters generally disapproved of patient-provider firearm discussions, except in the context of suicide prevention (25).

Additional viewpoints are needed to optimize firearm screening in routine care. Therefore, as part of a broader qualitative study focused on how primary care patients with recently endorsed depressive symptoms experienced questions about suicidal thoughts (26), we explored their perceptions of a routine screening question about firearm access.

Methods

Study Setting

The study was conducted at Kaiser Permanente Washington (KPWA), which serves about 700,000 patients across Washington State. All primary care patients with a current diagnosis of a mental disorder receive a self-administered mental health “monitoring” questionnaire on paper, which includes the 9-item Patient Health Questionnaire (PHQ-9) to assess depressive symptoms (27), questions about substance use (28–31), and a single question about firearm access (“Do you have access to firearms?” yes/no). Primary care patients without a current diagnosis of a mental disorder receive a mental health “screening” questionnaire, including the PHQ-2 (followed by the PHQ-9 when the PHQ-2 is positive) and substance use questions but not the firearm access question. Patient-provider lethal-means safety discussions are part of standardized follow-up for all patients identified as being at high risk of suicide (32), regardless of self-reported firearm access.

Study Sample

Sampling criteria (33), designed for purposes of the broader study (26), used electronic health records (EHRs) to identify adult patients (N=100) with a primary care visit in the prior 2 weeks. For comparative purposes, we purposefully sampled groups of patients who answered the ninth question of the PHQ-9 across all positive response options (several days, more than half the days, nearly every day). (Question 9: “How often have you been bothered over the past 2 weeks by thoughts that you would be better off dead, or thoughts of hurting yourself in some way?”) The sample for the parent study included both patients who received and who did not receive the firearm access question (“screening” versus “monitoring” questionnaire). Patients were recruited to the study via a mailed invitation letter, which included the study purpose and procedures and a phone number to opt out. Interviewers attempted to reach all invited patients by phone within 3 weeks of the invitation. Patients who agreed to interviews provided oral consent for participation and received $50. We estimated on the basis of prior research a need to enroll 30–40 patients to reach thematic saturation for the parent study (34). The project was approved by the KPWA Institutional Review Board.

Telephone Interviews

Three interviewers (two psychologists [U.W. and E.J.L.] and a health services researcher [J.E.R.]), recruited patients between August 30 and November 9, 2017, until the enrollment goal was reached. The interview guide was designed to help the interviewer understand the patient’s experience of being asked about suicidality for the parent study (26). Interview questions regarding the firearm access question (Box 1) were designed to assess all participants’ perceptions of the question, regardless of whether they had recently received it, and to elicit specific experiences answering the question from those who remembered it. To avoid interviewer bias, interviewers did not have access to participant characteristics at the time of the interview (35). Interviews were audio-recorded and transcribed.

Analysis

For the parent study, two health services researchers (J.E.R. and S.D.H.) coded transcripts uploaded to Atlas.ti (36); the coders were guided by a combination of directive content analysis (deductive), for which they used codes developed from the interview guide a priori, and conventional (inductive) content analysis, for which they used codes for themes that emerged from analyses (34, 37). Ten transcripts were independently coded by both coders for purposes of iterative comparison and achieving consensus (38, 39). Once all interviews were coded by at least one coder, codes were organized into a thematic network (i.e., affinity diagram) to facilitate depiction of salient themes (40). This process revealed that the breadth of data specific to perceptions of the firearm access question warranted further elaboration. Therefore, four coders (J.E.R., S.D.H., C.D.S., and C.L.) repeated processes and created a separate thematic network focused on this topic. The full investigative team met to review all coded data specific to the firearm access question to finalize themes and identify prototypic examples for presentation. Stratified analyses were later added to explore variation in themes by gender, age, and disclosure of suicidal thoughts.

Results

Participant Characteristics

A total of 37 participants completed semistructured interviews (mean±SD=15.4±6.0 minutes, range 5–29 minutes). EHR data indicated that most participants (N=28) reported suicidal thoughts (PHQ-9 question 9 score of 1–3), and approximately half (N=16) had received the firearm access question at the visit used for sampling, whereas others (N=8) remembered having answered the question in a prior visit. Aggregate characteristics are presented in Table 1, and participant-level information is presented in Table 2.

Qualitative Results

Three organizing themes were derived from deductive and inductive content analysis specific to firearm access disclosures: apprehensions about disclosing access to firearms related to privacy, autonomy, and firearm ownership rights; perceptions of relevance of the firearm question, informed by experiences with suicidality and common beliefs and misconceptions about the inevitability of suicide; and suggestions for connecting questions about firearms and other lethal means to suicidal intent. (A figure illustrating the themes is included as an online supplement to this article.)

Apprehensions.

Several participants described apprehensions about disclosing access to firearms because of fear of unknown consequences related to autonomy and privacy. One participant (P20) said, “When you just see it on this form, and you don’t know what they’re going to do about how you answer this form, for someone who is concerned about the government infringing on their rights, it gives you the feeling of, ‘Maybe I should just answer no.’ ” Similarly, another participant (P4) said, “That’s a Big Brother question for me, too, like if I was suicidal or whatever, would I tell someone that I actually had access to a gun?”

Several participants described their concerns about privacy. One participant (P21) said, “I waivered on answering that question. I suppose that you have to ask it in order to figure out whether or not how depressed I would be and things like that; however, I felt that it was really not part of your business whether or not I had a gun.” Another (P34) answered, “Yes, I have guns. So, some people are anti-gun, and some people are pro-gun. . . . If you’re asking a person that’s pro-gun, then they’d be like, ‘It’s none of your damn business.’ ”

However, gun owners did not always express concerns about privacy. As one participant (P36) recalled, “The question said, ‘Do you have access to a gun?’ and I just hit yes. . . . It’s very simple for me to just be honest.”

Participants also expressed concerns about losing access to firearms if they disclosed ownership in the context of mental illness. One participant (P10) responded, “If they have a permit to carry and are mentally unstable, then that opens up a whole new question. It’s one that I think most people would just answer no to.” Another (P25) described why retaining firearm rights might be important for those with depression: “While depression might be something you struggle with, it is not who you are. . . . I still believe that people who have depression have the right to carry a gun to defend themselves, because though they have days where they want to give up on life, they’re also at the same time spending every minute of every day trying to beat their depression, trying to focus on their survival.”

Perceptions of relevance.

Participants described the relevance of the question about firearm access, on the basis of its perceived utility for suicide prevention. Many participants described how asking about access to firearms was particularly relevant to clinical care for mental health. One (P5) said, “Yes, it is [appropriate] because somebody who’s going through this depression and anxiety—they can do things without thinking.” Conversely, participants also described their perception that questions about firearms are less relevant for women, because they rarely attempt suicide by firearm. For example, one participant (P21) said, “Obviously, a gun might be one way of doing that [suicide attempt]; however, as you probably know, most women would not use a gun.” Similarly, another participant (P25) described how the firearm access question was not relevant because of her belief that suicide was not preventable: “I feel like if a person really has their mind set on killing themselves, it doesn’t matter whether they have a gun or not. They will find a way.”

Other participants perceived relevance of the firearm question in the context of their personal experiences with depression. One (P6) said, “I would want someone to ask me because I’d say truthfully. Yeah, it’s shocking sometimes, but then after you’ve been depressed a few times and gone through things, you realize it has its purpose.” Another participant (P18) believed that the question was relevant when weighing the risk of self-harm with the value of protection: “Yeah. I do. I think it’s an important question. . . . I mean I purposely, no matter how afraid I would be or—I mean I have kids. And I wanna protect them. I wanna protect my house, but there was no way I would ever buy a gun. Just because I know if I ever got really, really depressed, I’d use it.”

Suggestions.

Some participants suggested expanding the question to address other common lethal means. One (P30) responded, “I think it’s an important question in addition to [other] implements that would kill you, like overdosing.” Another (P28) thought it would be important to specifically ask women about lethal means more broadly: “Women, I think, particularly say, ‘Of course not,’ while their purses are stuffed with sedatives and morphine derivatives and those things, which they might be planning to use when they get home.” Other participants shared examples about why answering general lethal-means questions would be more applicable. One (P16) said, “I overdosed on sleeping pills. So, even if I had a gun accessible, it wasn’t my way of thinking of how I would want to go.” Another (P9) said, “I do have a gun in my house. But it would get so complicated trying to open it and really, I would prefer not to go that way.”

Participants also suggested that more explicitly addressing suicidal intent could improve the utility of the question about firearm access. One (P29) said, “Technically, anybody has access to a gun. . . . ‘Would you shoot yourself?’ is how I always interpreted that question.” Another participant (P17) gave a detailed rationale for linking the question to suicidal intent for fellow veterans: “I don’t know any of my veteran friends who don’t have guns. And I know at least more than half of them have put the gun to their head and almost pulled the trigger. So, I feel like if there was some way to . . . pardon me for being blunt, but ‘Have you ever held a gun in your hand and turned it towards your face?’ or something to where the sincerity of that action is actually being verbalized.”

Participants also suggested specific examples of how information about firearm access information could be used for suicide prevention. One (P14) responded, “If I had access to a gun I would hope that the counselor I’m seeing there would either encourage me to lock it up in a place where I can’t get to it or encourage me to call her when I feel like using it.” Another (P20) expressed appreciation for a provider who used information about firearm access to discuss lethal-means safety and suicide risk: “She [provider] said, ‘The Second Amendment is just fine if that’s why you have the gun, but you’ve got to realize that the failure rate [margin of error] with a gun is so small. So, is there anything you can do to at least remove it from your immediate access?’ And I made an agreement with her, ‘Yeah, I can give it to this relative located 20 miles away from my house,’ . . . at least so that if I had any thoughts of using it, there’d be a few-hour cool down period before I could ever get access to it.”

Post Hoc Stratified Results

Variation in themes by gender, age, and disclosure of suicidal thoughts was also analyzed (see online supplement).

Discussion

This study is one of the first to assess primary care patients’ perspectives on being asked a question about access to firearms as part of routine mental health follow-up care (22). In this qualitative substudy, participants reflected varying perspectives: some were wary of disclosing firearm access because of concerns related to privacy, autonomy, and firearm ownership rights; some conveyed beliefs regarding at-risk populations that might prevent acceptance of population-based screening for firearm-related suicide risk; and most offered suggestions (e.g., including assessment of other lethal means, such as prescription medications) and highlighted important patient-centered approaches to addressing firearm access in the context of suicide prevention.

Participants’ expressions of fears of surveillance or infringement on firearm access rights resulting from disclosures about firearms were unsurprising given the current national debate about “gun control” (41). Research exploring suicidality disclosures has similarly underscored fear of loss of autonomy, an important barrier to disclosure (26, 42–45). The apprehension participants expressed underscores the benefit of offering full transparency to encourage honest disclosures. For example, it may be useful to clarify that the firearm question helps health care teams provide appropriate guidance or resources to patients about firearm safety in times of increased suicide risk. Our finding that one participant appreciated a provider who explicitly expressed concern for his safety and was nonjudgmental about firearm ownership underscores recommendations that provider-initiated discussions about firearm access should not be motivated by or convey disapproval of firearm access or ownership (46). Similarly, contextualizing firearm access screening more broadly into injury prevention may be helpful to avoid stigmatizing firearm access while addressing the rationale for asking specifically about firearm access—that firearms are the most common method in U.S. suicide deaths and that an estimated 85%−95% of individuals who attempt suicide by firearm die (2, 47).

Findings also suggest that health systems may play a role in addressing beliefs and misconceptions about suicide. For instance, providers can address the belief that suicide is inevitable by describing numbers of unplanned suicide attempts and can explain that the desire to die may occur during intense but often temporary periods of emotional pain (48). Providers may also address the perception that the question about firearm access is relevant only for men by noting that although more men die by firearm suicide, women also die by firearm suicide, particularly older women and women veterans (49, 50). Additionally, providers could address common misconceptions about the frequency of means substitution (51–53), particularly in the context of discussions about securing highly lethal means such as firearms with patients at risk of suicide (5, 54).

Discussions about strategies that patients can use to increase time or distance required to access firearms (e.g., storage of firearms or ammunition) and other lethal means, as suggested by one participant, is a recommended component of safety planning interventions (14, 15). However, the quality of safety is variable in practice (55, 56), and there appears to be room for improvement in the quality of rates of lethal-means assessment for patients with suicide risk (57). Nevertheless, lethal-means counseling, particularly firearm access assessment, may reduce risk of suicide (58). Further, new firearm suicide prevention interventions are becoming available, including a Web-based decision aid for adults (59, 60). These studies provide a foundation for how health systems can utilize information about firearm access when patients disclose access.

This study had several limitations. Because the study was a substudy of another designed to understand how patients experience screening for suicidality, research specifically designed to assess patient perspectives on being asked questions about access to firearms is needed, particularly among populations for whom disclosing firearm access might be most salient (61–64) (e.g., those with multiple firearm suicide risk factors, such as veterans [65–68]). Additionally, the parent study oversampled participants with suicidal thoughts. Although inclusion of many participants who had experienced suicidal thoughts enabled assessment of patients’ experiences in a high-risk population for whom questions about firearms may be most pertinent, further research is needed to generalize results to broader primary care populations. Those wary of reporting firearm access may also have been less likely to be sampled for the study or to have volunteered for participation in an interview about suicidality screening. In addition, participants lived in Washington State where an extreme risk protection order (a process allowing temporary firearm removal from individuals who are legally determined to pose significant danger to themselves or others) (69, 70) may have intensified participants’ apprehensions about how their providers would use firearm access information. Also, the experience of being asked questions about firearm access may change as patients become more accustomed to such questions, and future research should explore how experiences change. Finally, the question asked routinely at KPWA has not been validated, nor have other questions about firearm access, and validation studies are needed.

Conclusions

Findings from this novel study suggest that clarifying the purpose and use of routine screening of firearm access, contextualizing firearm questions within injury prevention broadly, and addressing misconceptions about suicide prevention may encourage firearm access disclosure and facilitate patient-centered dialogue about safe storage. These findings should inform future research and development of standardized firearm access assessment practices.