Medicine and Public Issues17 May 2016

Yes, You Can: Physicians, Patients, and Firearms

    Author, Article, and Disclosure Information


    Physicians have unique opportunities to help prevent firearm violence. Concern has developed that federal and state laws or regulations prohibit physicians from asking or counseling patients about firearms and disclosing patient information about firearms to others, even when threats to health and safety may be involved. This is not the case. In this article, the authors explain the statutes in question, emphasizing that physicians may ask about firearms (with rare exceptions), may counsel about firearms as they do about other health matters, and may disclose information to third parties when necessary. The authors then review circumstances under which questions about firearms might be most appropriate if they are not asked routinely. Such circumstances include instances when the patient provides information or exhibits behavior suggesting an acutely increased risk for violence, whether to himself or others, or when the patient possesses other individual-level risk factors for violence, such as alcohol abuse. The article summarizes the literature on current physician practices in asking and counseling about firearms, which are done far less commonly than recommended. Barriers to engaging in those practices, the effectiveness of clinical efforts to prevent firearm-related injuries, and what patients think about such efforts and physicians who engage in them are discussed. Proceeding from the limited available evidence, the authors make specific recommendations on how physicians might counsel their patients to reduce their risk for firearm-related death or serious injury. Finally, the authors review the circumstances under which disclosure of patient information about firearms to third parties is supported by regulations implementing the Health Insurance Portability and Accountability Act.

    In 2014, a total of 33 599 Americans died of gunshot wounds (1). Although the risk for firearm-related homicide is highest among young African American men, most firearm-related deaths are suicides, for which older white men are at highest risk (2). Public mass shootings account for a small percentage of firearm-related deaths but are occurring more frequently and could affect the character of public life in the United States (3). In early 2013, more than 40% of Americans worried that they could fall victim to firearm-related homicide or assault (4).

    Physicians seek to prevent important health problems at the individual and population levels. They inquire and counsel—routinely in some cases, selectively in others—about a wide range of health-related behaviors and conditions. In certain circumstances, they disclose otherwise confidential information to third parties to limit the risk an affected person poses to others.

    Physicians generally do not do well at firearm-related injury prevention, however. They ask infrequently about firearms and counsel poorly, if at all, although they are aware that the high lethality of firearms makes prevention efforts particularly important.

    In this article, we examine several commonly cited barriers to practicing preventive medicine for firearm-related injuries. Some physicians believe that it is against the law to discuss firearms; we show that this belief is unfounded. We then discuss the circumstances in which prevention efforts might be most appropriate. We briefly review other barriers to their more widespread adoption, patients' opinions about physicians addressing firearms, and the effectiveness of such prevention practices; more extensive information is available elsewhere (5, 6). We offer specific recommendations, based on the limited evidence available, for how physicians can incorporate firearm-related injury prevention into the care of their patients.

    You Can Ask

    No federal or state statute prohibits physicians from asking about firearms when such information is relevant to the health of the patient or others.

    Federal Statute

    The Patient Protection and Affordable Care Act (ACA) regulates the collection of firearm information, but the scope of these provisions is narrow (Table 1). Most important, the ACA prohibits required collection of firearm information by “wellness and health promotion” programs. The Obama administration's position was reflected in a January 2013 White House announcement that, to “protect the rights of health care providers to talk to their patients about gun safety,” it would issue guidance clarifying that the ACA does not regulate communication between physicians and patients about firearms (7). The U.S. Department of Labor repeated this clarification (8); to our knowledge, this guidance has not been issued.

    Table 1. Language From Relevant Federal and State Statutes*
    State Statutes

    The only existing gag law is Florida's medical privacy act concerning firearms (Table 1), enacted in 2011 after a series of inflammatory incidents occurred during clinician–patient interactions. In one widely publicized incident, a pediatrician in Ocala discharged from care 3 children whose mother declined to answer questions about firearms (9, 10).

    Florida's law is seen as intruding on the patient–physician relationship (11). Its key provisions hold that health practitioners “should refrain” from asking about firearms and “may not intentionally enter” firearm information into medical records. There are broad exceptions, however. A practitioner who “in good faith believes that this information is relevant to the patient's medical care or safety, or the safety of others” may ask about firearms, and only information that “is not relevant to the patient's medical care or safety, or the safety of others” is excluded from medical records (12).

    Physicians and medical professional organizations challenged the law in U.S. District Court, contending that it unduly restricted their First Amendment rights. They were successful (9), but that decision was overturned by a divided 3-judge panel in the U.S. Court of Appeals for the Eleventh Circuit (10, 13). In February 2016, the full court vacated its panel's decision and agreed to rehear the case (14). The District Court decision remains in effect.

    Other States

    Three states—Montana, Missouri, and Minnesota—have statutes addressing firearm information acquired by health practitioners or agencies. None includes a gag provision. Montana prohibits requiring patients to provide firearm information as a condition of receiving health care (15). Missouri prohibits requiring that health professionals collect or record firearm information, but with an exception “if such inquiry or documentation is necessitated or medically indicated by the health care professional's judgment” (16). Minnesota prohibits collection of firearm information by its state health commissioner and MNsure, the agency administering its health insurance exchange (17, 18).

    More restrictive proposals have been considered elsewhere. In 2015, North Carolina's legislature debated prohibiting the collection of firearm information in any written form and the disclosure of responses to firearm-related questions “unless the patient has been adjudicated incompetent due to mental illness” (19). A true gag law proposal was introduced in Ohio in June 2015: “A physician shall not ask a patient or prospective patient … [any] question related to the ownership or possession of a firearm” (20). As of April 2016, no action had been taken on this bill.

    When to Ask

    Firearm information would be directly relevant to the health of an individual patient and that patient's close contacts under 3 general conditions (Table 2). Where physician preference or simple lack of time precludes routine screening for firearm access, physicians could identify patients for whom these conditions apply and proceed selectively.

    Table 2. Conditions When Firearm Information Might Be Particularly Relevant to the Health of a Patient and Potentially to Others

    The first arises when a patient provides information or exhibits behavior suggesting an acutely increased risk for violence, such as explicit or implicit endorsement of suicidal or homicidal intent or ideation. This is an emergency; the need to determine access to lethal means for such patients is self-evident.

    The second involves patients who possess other individual-level risk factors for future violence. A history of violence perpetration is a strong predictor of future violence in the general population (21) and among firearm owners (22). Similarly, patients hospitalized or treated in emergency departments for violent victimization are at high risk for committing violence (23, 24). Alcohol abuse is common in the general population and among firearm owners (25) and is a well-established risk factor for interpersonal and self-directed firearm violence (26). Drug abuse is another risk factor, but the magnitude of the risk may be drug-specific (27).

    Serious mental illness is a relatively minor risk factor for violence to others, and risk depends on the nature of the illness (28, 29). Alcohol and drug abuse, prior violence, and violent victimization account for much of the risk popularly ascribed to mental illness (30, 31). Self-directed violence is very strongly linked to mental illness, however. Suicide risk is increased by a factor of at least 10 across a range of psychiatric diagnoses, and an estimated 47% to 74% of suicides are directly attributable to mental disorders (30). Violence risk is particularly high among patients seen for mental illness in emergency departments, recently discharged psychiatric inpatients, and persons experiencing a first psychotic episode (30). Brain disorders associated with impaired cognition and judgment, such as Alzheimer disease, are associated with an increased risk for aggression and violence (32).

    Third, questions about firearms would be relevant for patients in demographic groups that are at increased risk for firearm-related injury. Middle-aged and older white men are at high risk for firearm-related suicide (up to 5 times higher than black men of the same age), and young African American men are 20 times as likely as young white men to die of firearm-related homicide (1). Children and adolescents may engage in risk behaviors with firearms because their judgment and cognitive skills are not fully developed (33).

    Persons with risk factors, including substance abuse disorders, diagnosed mental illness, impulsive anger, suicidal ideation, and dementia, report access to firearms at levels similar to those in the general population (34–39). Persons with multiple risk factors are a high-priority group, because risk is likely more than additive in such cases (22, 40).

    Questions about firearms may be appropriate when the patient is not a firearm owner, given that risk for victimization may extend to all household members if one is at risk for violence and firearms are present. For example, firearm-related questions would be appropriate if a patient's intimate partner exhibited violence and abused alcohol.


    High proportions of physicians (66% to 84%) believe they have the right to counsel patients about firearms (41) and a responsibility to engage in efforts to prevent firearm-related injuries (42, 43). The American Academy of Pediatrics has long recommended that physicians provide counseling (44); since the passage of the Florida law, many other medical specialty societies have made similar recommendations (45).

    Actual practice is another matter. Screening and counseling about firearms is uncommon, whether as anticipatory guidance (46) or a selective intervention (47–49). For example, in a study of veterans who screened positive for suicidal thoughts, only 15% of providers documented whether patients had access to firearms (50).

    Many barriers exist. Perhaps the most important is clinician unfamiliarity with the benefits and risks of firearm ownership, with what to say during firearm safety counseling and how to say it, and with firearms themselves (41, 43, 48, 51–53). It is prudent to avoid areas of practice when one is unfamiliar with the evidence, and in fairness, too little solid evidence exists. These barriers can be addressed through training (54) and through development of information about referral resources (55, 56).

    Physicians may worry about damaging the patient relationship (48) by asking questions that seem intrusive or invite discord (57). They may feel uncomfortable asking about firearms, even when they are well-informed (48, 51), or worry that patients will not be truthful (48). Some may believe that firearm counseling is outside their scope of practice (51, 57) or that discussing firearms could infringe on patients' Second Amendment rights (43, 58). Fortunately, studies of patient perspectives on firearm counseling have found that most—although certainly not all—patients seem to be open to nonjudgmental education, especially for members of high-risk groups (48, 57, 59). The American Bar Association has stated that Second Amendment rights are not affected by asking and counseling about firearms (45). Finally, clinicians may doubt the effectiveness of firearm counseling, especially given time limitations and competing clinical priorities (43, 51, 60).


    Research on practices to prevent firearm-related injury is in its infancy; too few studies have been done. For example, a recent systematic review identified no studies of the accuracy or predictive value of screening for firearm access among high-risk patients (5). Such absence of evidence is common. Except for a few behavioral screens endorsed by the U.S. Preventive Services Task Force (61), there are few clinically based studies of valid screening tools for other frequently encountered risk behaviors. Nonetheless, the Joint Commission has, for example, mandated screening for suicide among high-risk inpatient and emergency department populations (62). Because firearms are the most common means of violent death in the United States, the lack of rigorously validated tools should not, in itself, deter screening.

    Many, but not all, studies of clinical interventions to prevent firearm-related injuries have shown increased firearm safety behaviors among populations at high risk for injury. No interventions have resulted in harm (5, 6).

    Most studies have assessed interventions in primary care settings. In one randomized, controlled trial, a screening, brief intervention, and referral to treatment (SBIRT) intervention involving specially trained pediatricians led to increased safe firearm storage in households with children (63). Other observational and intervention studies have had positive results, including decreased firearm access among suicidal teens and adults (64, 65) and increased safe firearm storage in households with children (66–69). After brief counseling by a psychiatrist, for example, almost one third of suicidal adolescents' families removed a firearm from the home (68). Because there was no control population, we do not know how many families would have removed a firearm without intervention. Firearm counseling may be less effective when combined with other interventions or delivered universally (70–72).

    Hospital- and emergency department–based interventions using SBIRT methods or comprehensive case management have shown promise in reducing the risk for repeated violence (73, 74). A randomized trial of a collaborative care program, including brief interventions, comprehensive case management, and as-needed mental health treatment, showed decreased weapon carrying among adolescents hospitalized after an injury (75). Many studies have relied on surrogate outcomes or had shortcomings in design or execution, such as frequent loss to follow-up. Controlled trials and a systematic review now in progress (74) will hopefully clarify the role of these interventions.

    How to Ask and Counsel

    The provider's attitude is critical (76). Patients are more open to firearm safety counseling when providers are not prescriptive but focus on well-being and safety—especially where children are concerned—and involve the family in respectful discussions (48, 57). Conversations should acknowledge local cultural norms (57); be individualized (48); and, when possible, occur within a well-established clinician–patient relationship (57). Acute care physicians, such as hospitalists and emergency physicians, should intervene when risk is increased or they encounter a teachable moment. We suggest, “Don't just ask, inform”; emphasize education, not just information gathering (48, 57).

    Giving a context for firearm-related questions can enhance the interaction. For anticipatory guidance, a firearm-related question can be included in routine screening for household hazards or risk behaviors (57). For selective screening in patients with other risk factors, we recommend that clinicians briefly explain why firearms might be relevant to the patient's well-being and safety.

    The first question might simply be, “Are any firearms kept in or around your home?” When the answer is “yes,” 2 follow-up questions are important: “Do any of these firearms belong to you personally?” and “Are any of these firearms stored loaded and not locked away?”

    What to do with the answers depends on the circumstances (Table 2). Questions and advice about safe storage might make use of the “5 Ls” mnemonic (Locked, Loaded, Little children, feeling Low, Learned owner [77, 78]), accompanied by a nonjudgmental educational handout (Table 3) and information on firearm storage devices (Table 4). The desired outcome is that firearms are stored unloaded and locked, with ammunition stored separately. Safe storage is less common when other risk factors are present.

    Table 3. Materials for Distribution to Patients
    Table 4. Firearm Safe Storage Options*

    For patients at risk for violence, counseling to reduce access to lethal means of harm (“lethal means counseling”) (69, 79) is indicated; mental health, social service, and substance abuse referrals may be appropriate. It may become important for the patient to temporarily relinquish custody of firearms to family members, gun shops, or law enforcement, as allowed by state law. California's new Gun Violence Restraining Order statute allows family members or law enforcement officers to petition for a court order that firearms be recovered for safekeeping in such cases (80). Connecticut, Indiana, and Texas also allow for mandated firearm recovery. Other states may soon follow.

    The advisability of having firearms at home may come up for discussion. Clinicians can point to a large body of evidence establishing that, on balance, firearms in the home (81) and purchasing a handgun (82) are associated with a substantial and long-lasting increased risk for violent death.

    You Can Disclose

    The Code of Federal Regulations, in provisions governing the Health Insurance Portability and Accountability Act, lists specific situations in which protected health information may be disclosed (83). These include occasions when disclosure “is necessary to prevent or lessen a serious and imminent threat to the health or safety of a person or the public and … is to a person or persons reasonably able to prevent or lessen the threat, including the target of the threat.”

    The Obama administration has emphasized that disclosure is permissible: “No federal law prevents health care providers from warning law enforcement authorities about threats of violence” (7). A letter from the U.S. Department of Health and Human Services added that “a health care provider may disclose patient information, including information from mental health records, if necessary, to law enforcement, family members of the patient, or any other persons who may reasonably be able to prevent or lessen the risk of harm” (84).

    There is no agreed-on definition of the key term “imminent”; in research on violence prediction, it describes time frames ranging from a few days to months (85). Elsewhere in the Code of Federal Regulations, an imminent hazard is defined as one that may cause harm before it can be alleviated by a formal regulatory action.

    There is also no clear guidance on informing patients of such disclosures. The American College of Physicians (ACP) Ethics Manual states, “In general, individuals have the right to full and detailed [information]” (86). Timing and other specifics will vary because “upsetting news and information should be presented … in a way that minimizes distress.” Citing the American Medical Association (AMA) Code of Medical Ethics (87), ACP's manual adds that decisions not to inform patients (once referred to as therapeutic privilege) should be rare. However, the AMA Code stresses that “therapeutic privilege does not refer to withholding medical information in emergency situations” and allows for delayed disclosure even in nonemergency situations (87). As in domestic violence and child abuse cases, providers should proceed from a right-to-know position, determining when and whether a patient is informed on the basis of the provider's assessment of the patient's cognitive status and “the balance of benefits and harms” (87) expected to result from providing the information.


    Medical specialty societies recommend asking and counseling about firearms during routine patient care or as an element of risk assessment (45). Neither the law nor the Second Amendment prohibits such activities, and the First Amendment may protect them. The limited evidence on effectiveness is encouraging, and the need for additional research should not impede efforts to provide patient care consistent with the best information available (Table 5).

    Table 5. Additional Resources

    More training and better resource are needed. A systematic review identified only 4 reports on the development and assessment of training for clinicians (88). Input from clinician and nonclinician firearm owners is important (51). Collaborations with firearm retailers to prevent suicide are being tested in New England, the Rocky Mountain states, and Washington and may become important (89).

    We recommend that, at a minimum, clinicians determine access to firearms for patients who fall into any of the risk categories discussed previously. Depending on the circumstances, interventions may include education; counseling in support of behavior change; or more direct efforts, such as disclosure to others, to prevent death or serious injury.



    Sign In to Submit A Comment
    Garen J. Wintemute, MD, MPH; Marian E. Betz, MD, MPH; Megan L. Ranney, MD, MPH4 October 2016
    Physicians, Patients, and Firearms: In Response to Dr. Sklaroff
    September 29, 2016

    To the Editor:

    Dr. Sklaroff (1) suggests that we advocate asking about firearms without any prior effort to establish either the need for such questions from the physician’s perspective or their reasonableness from point of view of the patient. He is incorrect. We discuss these preliminaries in our article (2) and elsewhere (3) and use that approach in our clinical work. We wholeheartedly agree with the importance of non-pejorative phrasing—again, a point we have made repeatedly. Perhaps he sees something pejorative in “Are there any firearms kept in or around your home?” We do not.
    We take issue with his use of the loaded term “profiling” to describe making an initial and tentative assessment of an individual patient’s risk for an adverse health event based in part on characteristics that patient shares with others. Physicians adopt this approach all the time—of necessity—are aware of its limitations, and ask questions precisely to refine and individualize those initial assessments. Middle-aged and older white, non-Hispanic males are a prime example. Firearm suicide risk in that large group is high and increasing; to a surprising extent, firearm violence is now an old white guy problem (4). Does that high risk apply to every member of the group? Of course it doesn’t. Does it provide sufficient basis for further inquiry, particularly when other risk factors are involved? We believe it does.
    Our choice of a title came from the circumstances that led to the writing of the article itself. In recent years we have given presentations on firearm violence at national meetings of medical professional societies and have repeatedly been asked whether it was even permissible under the law for physicians to ask patients about firearms. The first draft of the article focused primarily on a discussion of state and federal laws on this matter, and the title was chosen to summarize our answer to that repeated question. As both our article and this response to Dr. Sklaroff should make clear, we agree entirely with him that physician-patient relationships should be “untethered from mandates.” This includes mandates that would restrict the free exchange of ideas and information about firearms.


    1. Sklaroff RB. The privacy of physician history-taking mustn’t be violated, at gun-point. Ann Int Med. 2016 Aug 6. Available at
    2. Wintemute GJ, Betz MD, Ranney ML. Yes, you can: physicians, patients and firearms. Ann Intern Med. 2016;165:205-213.
    3. Betz ME, Wintemute GJ. Physician counseling on firearm safety: a new kind of cultural competence. JAMA. 2015;314:449-450.
    4. Wintemute GJ. The epidemiology of firearm violence in the twenty-first century United States. Ann Rev Public Health. 2015;36:5-19.
    Robert B. Sklaroff, M.D.8 August 2016
    The Privacy of Physician History-Taking Mustn’t Be Violated, at Gun-Point
    Just as was the case two years ago (1), because “public health research predictably guides generation of public policy, it is necessary to scrutinize the political science underlying the paired systematic review (2) and editorial on gun control.” Inasmuch as litigation pends before the SCOTUS, the focus is properly trained on slippery-slope rhetoric subtly embedded in (and omitted from) discussion of what is portrayed as inherent in the patient-physician relationship.

    It is suggested that an initial query “might simply be, ‘Are there any firearms kept in or around your home?’ (2).” No, the topic should be gently introduced just as is accomplished with other concerns such as tobacco-use, namely, “Would you like to discuss….?” Non-pejorative phrasing is welcoming without being intrusive; the more aggressive approach “simply” may trigger communication cessation and, thus, preclude invoking the presumed motivations for such questioning, education and counseling.

    In addition, “profiling” of demographic groups allegedly carrying increased-risk for violence includes citation of a generic CDC-website to support the claim that middle-aged and older white men are “at high risk for firearm-related suicide (up to 5 times higher than black men of the same age).” Yet, the specific article addressing this topic included a graph that depicts a three-fold increase (3) and another article covering the same decade (1999-2008) reported a datum that was ignored: “Among racial/ethnic groups, the greatest increases in suicide rates were among white non-Hispanics (40 percent) and American Indian and Alaska Natives (65 percent)” (4). Targeting millions of Caucasian males is unjustified.

    Not surprisingly, an article published last year that conveyed opinions consistent with those being promulgated was cited uncritically, failing to refute the referenced-claim in an accompanying letter that the “onslaught of ‘scientific’ support for gun-control laws is flawed” (5). The abstract even admitted specific recommendations were generated by “proceeding from the limited available evidence.” Such abuse of academic discretion—having bipartisan roots—is more ideological than definitive; it poisons what should be disinterested inquiry devoid of subliminal messaging.

    Finally, the title (“Yes, you can…”) was overtly reminiscent of the political campaign of the incumbent president, seemingly at-one with the big-government themes that have been emblematic since 2009. Polemics have no place in the composition of scholarly articles; paradoxically, their suasive capacity is diminished as a result.

    Clinical interactions must uphold patient-physician privacy-rights and be untethered from mandates, both regarding history-taking and how it is recorded.


    1. Sklaroff, RB. Guns, Suicide, and Homicide [Letter]. Ann Intern Med. 2014;160(12):876-877. doi:10.7326/L14-5012-2

    2. Wintemute GJ, Betz MD, Ranney ML. Yes, you can: physicians, patients and firearms. Ann Intern Med. 2016;165:205-213. Doi:10.7326/M15-2905

    3. Centers for Disease Control and Prevention. QuickStats: Death Rates from Suicide* for Persons Aged 45–64 Years, by Black or White Race and Sex — United States, 1999–2008. Morbidity and Mortality Weekly Report (MMWR) 61(01);21 (January 13, 2012). Atlanta, GA: Centers for Disease Control and Prevention. Accessed at on 5 August 2016.

    4. Centers for Disease Control and Prevention. CDC finds suicide rates among middle-aged adults increased from 1999-2010. Atlanta, GA: Centers for Disease Control and Prevention; 2014. [Press Release]. Accessed at on 5 August 2016.

    5. Sklaroff, RB. Onslaught of "Scientific" Support for Gun-Control Laws is Flawed [Letter]. Ann Intern Med. 2015;162(7):513-516. doi:10.7326/M15-0337. Accessed at on 5 August 2016.