Context. Physicians have been encouraged to counsel families about risks associated with gun ownership, but the effectiveness of physician counseling regarding gun safety is unknown.
Objective. To determine the effectiveness of gun safety counseling during well-child care visits.
Design. Group randomized, controlled trial. Physicians, nurse practitioners, and physician assistants were randomized to either the intervention group or a control group.
Setting. Group Health Cooperative, a staff-model health maintenance organization.
Patients. Consecutive sample of families (n = 1295) seen for a scheduled appointment for well-child care for a child <18 years of age. Of the families originally scheduled for a visit, 80.3% were seen and completed the outcomes surveys.
Intervention. Each family in the intervention group was given a 60-second message by their practitioner that depended on the presence of guns in the home. Families without guns were informed of the health risks associated with gun ownership and given a standard information pamphlet. Families with guns were given the same information about risks and were told that if they chose to keep a gun, they should store it locked and unloaded. They were given instructions on storage and a folder with material, including the same pamphlet, a letter from the police department, written storage guidelines, and discount coupons for gun storage devices.
Main Outcome Measure. Changes in the following self-reported events: 1) acquisition of a safe storage device; 2) removal of firearms from the home; and 3) acquisition of firearms.
Results. There were no important differences between intervention and control groups in the rate of acquisition of new guns (intervention: 1.3% vs control: .9%) after the intervention. Among households with guns at baseline, there were also no differences between groups in the removal of guns (intervention: 6.7% vs control: 5.7%), but there was a fairly large nonsignificant difference in the proportion who purchased trigger locks (intervention: 8.0% vs control: 2.5%).
Conclusions. A single firearm safety-counseling session during well-child care, combined with economic incentives to purchase safe storage devices, did not lead to changes in household gun ownership and did not lead to statistically significant overall changes in storage patterns.
- STOP =
- Steps to Prevent Firearm Injury
Firearms are frequently involved in the death of children and adolescents in the United States. In 1996, a total of 4643 children and adolescents died from firearm- related injuries.1Eighty-five percent of these fatal injuries occurred among adolescents between 15 and 19 years of age. Twenty-eight percent of all firearm deaths among children and youth were suicides, and 62% were homicides. Only 8% were classified as unintentional, with the remainder being of undetermined intent.
Household access to firearms has been implicated as a risk factor for both youth suicides, as well as unintentional firearm deaths in the home among children.2–6 Firearms are common among US households. According to the National Survey of Private Ownership of Firearms, ∼35% of US households own at least 1 firearm.7 Children may be exposed to the risk of firearm injury by handling loaded weapons that belong to their parents or other adults in the household by being commonly exposed to either unlocked or loaded weapons. Several studies focused on households with children have reported that 10% to 20% of handguns are stored both unlocked and loaded.8 ,9
There has been recent interest regarding the role of pediatricians and other health care practitioners in addressing firearm safety risks within anticipatory guidance in primary care pediatric practice. Firearm safety counseling is currently uncommon among pediatricians and family physicians, but many physicians acknowledge that it may fit with other forms of injury prevention counseling.10–13 Because physicians frequently list “lack of knowledge” as an important barrier to counseling, some professional medical organizations, such as the American Academy of Pediatrics, have developed specific programs and materials to train pediatricians in firearm safety counseling. However, the effectiveness of firearm safety counseling by health care practitioners is unknown.
The purpose of this study was to determine the effectiveness of a firearm safety-counseling program delivered by primary care clinicians to parents of children and adolescents attending well-child visits.
The design was a group, randomized, controlled trial in which the unit of randomization was the individual practitioner. Using a computer generated random number, practitioners were block randomized within specialty practice clusters to balance physician specialty (pediatrics or family medicine) and clinic location between treatment groups. Clinicians assigned to the intervention delivered the intervention described below. Clinicians in the control group did not receive any training, materials, or specific information about the study. They were told to conduct their usual practice. Clinicians did not share rooms or nursing staff.
The study was conducted among primary care practices at Group Health Cooperative, the largest staff-model health maintenance organization in the state of Washington. Most Group Health practices in western Washington State are situated in large primary care clinics with many practitioners in each location. Although information on race, educational attainment, and income are not collected routinely for Group Health patients, several focused surveys using a random sample of the cooperative population indicate that the overall patient population is 90% white, with 75% having at least some college education. Approximately one quarter of children <5 years old at Group Health Cooperative receive Medicaid insurance.
Subjects for this study were selected at 2 levels: the practitioner and the patient's family. Nine urban and suburban clinic sites were chosen, primarily based on the volume of pediatric patients and the geographic accessibility of the site to the investigators. Clinics were initially approached for possible participation by consultation with senior clinical staff. All clinics that were approached agreed to participate. Individual providers at the participating clinics were chosen based on the volume of pediatric patients seen in the previous 2 to 3 months. Practitioners were approached to participate if they saw at least 5 patients <18 years of age per month for well-child visit or physical examinations. All practitioners who were approached except 1 agreed to participate.
Families were eligible for the study if they had a scheduled visit during the study with a participating practitioner for a child or adolescent of 2 months to 18 years of age for well-child care or a sports physical examination. Families were contacted before the scheduled visit and asked to participate in a baseline survey of household injury prevention practices.
Physicians and other providers in the intervention group were trained by 1 of the investigators (J.M.), a practicing physician in Group Health, using the method of academic detailing.14Academic detailing is a method used for continuing medical education in which a peer physician (or other health provider) conducts a brief training session on a topic, with highlighting of important messages and subsequent reinforcement. During the session, which lasted 20 to 30 minutes, practitioners were given a standardized packet of firearm-counseling materials, including physician reading material, a bibliography, and an audiotape with suggestions on how to counsel about firearms. Most of these materials had been developed jointly by the American Academy of Pediatrics and the Center to Prevent Handgun Violence, as part of the Steps to Prevent Firearm Injury (STOP) program. Practitioners were asked to counsel all families coming for well-child care and physical examinations and to discuss firearm safety in the context of other child safety issues (Fig 1).
Practitioners gave parents 1 of 2 messages on firearm safety, depending on whether household firearm ownership was reported. The message for families without household firearms was intended to stress the risks associated with access to household firearms and to discourage these families from introducing firearms into the home in the future. The families of patients were given a STOP brochure to reinforce this message. The message for patients with household firearms was that the safest action for the family was to remove guns from the household, because the risks of ownership seem to be greater than the potential protective effect.15 Furthermore, parents were told that if they chose to keep firearms in the household, they could reduce the risk of injury by keeping firearms unloaded in locked locations. Families were advised that a trigger lock or a gun safe was the best choice for a rifle and that a portable lock box with a combination lock (that could be bolted to a floor or wall) was the best choice for a handgun. The practitioner gave these families a folder including a letter from a local police chief encouraging safe weapon storage, the STOP brochure, an information pamphlet describing firearm safety devices, and coupons for obtaining 1 trigger lock and 1 lock box at a discount. The coupons could be redeemed by mailing them to the research site; the selected device was then mailed to the home. Finally, patients in the intervention group were also exposed to STOP posters displayed in the examination rooms. Both standardized messages were printed on a laminated card for easy practitioner reference. No additional interventions occurred after the single clinic visit.
The study for each practice lasted from 30 to 60 days. Sample size calculations indicated that with 15 practitioners per group and α at .05, we had 80% power to detect a 5-percentage point difference in gun ownership in the intervention group. We also had 80% power to detect a 6-percentage point difference in the proportion of families who stored their guns locked in the intervention group.
Participating practitioners were notified at the start and end of the data collection periods for their clinics. Practitioners were reminded about eligible patients with a sticker clipped to the medical record. They were asked to place the sticker on the chart as an indication that counseling had taken place. In an effort to estimate the proportion of patients counseled, a sample of records was routinely audited to determine the proportion of eligible patients who had actually been seen with stickers on their charts. Of the 303 charts audited, 73% had a sticker placed on the chart.
The primary outcomes of interest for this study included: 1) the removal of firearms from households that owned guns at baseline; and 2) the acquisition of a safe storage device by gun-owning families. Because families without guns could potentially be led to believe that safely stored guns may sufficiently decrease the risk of ownership to induce a gun purchase, we also examined whether families that did not own a gun at baseline acquired 1 after the intervention.
Information about firearm ownership and storage was collected at baseline before the office visit and again 3 months after the office visit. Before the office visit, eligible families were mailed a 2-page, self-administered survey that was accompanied by a cover letter from their personal practitioner explaining its purpose. The survey included questions regarding motor vehicle safety, use of seat belts, bike helmets, smoke detectors, and firearm safety. Parents in both study groups were told only that the primary care practice was participating in a study of injury prevention practices among families with children and adolescents. Households not returning the survey after 1 reminder were telephoned and asked to complete the survey over the phone with a research study assistant.
Approximately 3 months after the scheduled office appointment, families in both study groups were recontacted by mail to determine whether their firearm ownership or storage practices changed after the office visit. This survey was also accompanied by a cover letter from the primary practitioner and covered similar injury control topics and safety behaviors as the baseline survey. The firearm-related items included whether guns had been acquired or removed from the household and whether 1 of several safe storage devices, such as trigger locks, had been acquired since the last survey. Again, households not responding to the follow-up survey or a reminder were telephoned and asked to complete the survey over the phone. Survey data were linked with a limited number of demographic variables from the Group Health enrollment database. The surveys were confidential but not anonymous to allow for response tracking and linkage over time.
Outcome results were expressed as the proportion of respondents who had each outcome in the intervention and control arms. Significance tests and confidence limits were based on randomization tests involving 10 000 rerandomizations at the practitioner level.16 This approach accounted properly for the practitioner level (rather than the household level) randomization, while making minimal assumptions about the distributional form of practitioner level variation. We used a permutation method to calculate the 2-sided P value and 95% confidence intervals for any difference in outcomes between the 2 trial arms.17 ,18
The investigational review boards of the University of Washington and the Group Health Cooperative approved this study.
Nine clinic sites were recruited for the study. In these sites, 56 practitioners (34 physicians [including 23 pediatricians], 9 physician assistants, and 13 nurse practitioners) were enrolled, with 28 in the intervention group and 28 in the control group. Of a total of 2082 eligible households approached, 1673 (80.3%) kept their appointment with the scheduled practitioner during the study period. Of these, we were able to collect both baseline and follow-up data on 1292 households (77%); follow-up surveys were received from 1295 households and rates of follow-up were the same between groups. There were 618 households in the intervention group, and 677 in the control group.
At baseline, ∼24% (n = 309) of the families in both groups reported owning any firearm. Fourteen percent of all subjects reported owning handguns, and 17.6% reported owning a rifle. Households in the intervention and control groups were similar (Table 1). There were no statistically significant differences between groups of households with regard to the gender of the respondent, the prevalence of household firearms, or how weapons were stored in the home.
There were no statistically significant differences between intervention and control groups in the acquisition of a handgun or rifle in the 3 months after the index visit (intervention: 1.3% vs control: .9%; Table 2). Similarly, there were no important differences between groups in their removal of guns from the household after the intervention (intervention: 2.0% vs control: 2.0%). These results were similar both for handguns and rifles, and when the analysis was restricted to only those owning guns at baseline.
There were no large differences between groups with regard to the acquisition of firearm storage or locking device after the intervention. Among the intervention group, 7.9% of families acquired a safe storage device, compared with 6.3% in the control group (P = .34). Among households owning guns at baseline, there was a fairly large difference in the proportion that purchase trigger locks (intervention: 8.0% vs control: 2.5%; P= .06; Table 3).
Families in the intervention group redeemed a total of 41 discount coupons distributed to families for safe storage devices: 26 were for handgun lock boxes and 15 were for trigger locks. The median coupon redemption price for a lock box was $9.99 (range: $9.99–$45.00; average retail price: ∼$70.00), and the median for a trigger lock was zero dollars (range: 0–$5.00; average retail price: ∼$10.00). In some cases, relatives of the patient's family used the coupons.
This randomized, controlled trial did not provide evidence that a single, brief counseling session during a well-child visit caused families to either give up a gun or reduced the likelihood that they would acquire a gun. Gun-owning families who received the intervention were somewhat more likely to purchase a safe storage device, particularly a trigger lock, but these differences between groups may have been attributable to chance.
To our knowledge, only 1 other study of firearm safety counseling in a primary care setting has been published.19 This nonrandomized, preevalation/postevaluation study was also based on an office intervention using the STOP materials. Similar to the present study, no statistically significant differences were found with regard to gun ownership or storage patterns.
If the increased trigger lock purchasing by gun-owning families was attributable to the intervention, it would correspond to a number-needed-to-treat of 18.20 This means that 18 gun-owning families would need to be counseled to persuade 1 patient to purchase a trigger lock. This level of counseling effectiveness is comparable with that of some other counseling interventions in medicine and public health. For example, based on the strongest evidence from randomized trials, 33 families need to be counseled for 1 family to buy a car seat and 13 families need to be counseled to result in 1 family's purchase of a smoke detector.21 An evaluation of smoking cessation counseling by pediatricians has shown that, depending on the length of the intervention, between 17 and 37 mothers need to be counseled for each mother who stops smoking.22 However, before firearm safety counseling can be endorsed, additional information is needed to provide more conclusive evidence of an effect on storage device acquisition. Furthermore, the efficacy of trigger locks needs to be determined to estimate the magnitude of injury risk reduction and the cost-effectiveness of device purchasing. Trigger locks vary in quality and in compatibility with firearms.23
Firearm injury prevention counseling by primary care pediatricians and other child health practitioners has been advocated by leaders and professional medical organizations, such as the American Academy of Pediatrics and the American College of Physicians.24 The US Preventive Services Task Force supported counseling families about firearm storage and safety with a type B (“fair evidence to support the recommendation”) recommendation. This recommendation was based on studies demonstrating an increased risk of death associated with access to firearms but not on studies showing effectiveness of this type of counseling. Few randomized, controlled trials have been performed to confirm the magnitude and duration of effectiveness of injury prevention counseling.25 However, systematic reviews demonstrate that most trials have revealed relatively small effects (<10% differences between groups) or the absence of long-term effects of counseling.22
This study had several limitations. First, the subjects received only 1 counseling session regarding firearm safety. Based on experience with the promotion of child occupant safety and other interventions, a more potent intervention may be needed, involving having families receive information and encouragement on multiple occasions.26 ,27Blinding was obviously not possible and could have affected the study in either direction. The good participation by practitioners and the relatively high proportion of stickers in medical records seemed to indicate good compliance with the intervention and makes it unlikely that the lack of effect was related to lack of practitioner effort or interest. The generalizability of these results to other settings is unknown. Families were aware of the intervention, but it is unknown whether they were aware of a link between the mail survey and the counseling session. If families were aware, it may have exerted bias toward demonstrating an intervention effect, because parents may wish to provide the most pleasing answer in response to their physician's survey. This study was conducted without evidence of the effectiveness of safe storage devices. This lack of evidence may have blunted the effectiveness of the intervention. Although there is evidence for the validity of self-reported firearm ownership, we are unaware of studies that have validated self-reported firearm storage practices.28 ,29 Finally, the power calculations presented did not correspond to the final outcome measurements. During the design of the outcome instruments, we became aware that it was impractical, in the context of this study design, to measure the status of every gun in the household. The final outcome measures may or may not have been more sensitive to changes. The 95% confidence intervals provide guidance regarding the precision of our estimates and the range of results that are plausible given the size and results of our study.
We conclude that a single firearm safety-counseling session during well-child care, combined with economic incentives for safe storage devices, did not lead to changes in household ownership of guns and did not lead to statistically significant overall changes in storage patterns. Based on these findings, it would be premature to recommend routine firearm safety counseling to primary care practitioners. Because of the potential for a small effect on trigger lock purchases, new trials should specifically determine whether an enhanced intervention program could result in improved gun storage patterns. Additional research is also needed to determine the effectiveness of gun storage devices in reducing child and adolescent firearm injury.
This study was funded by a grant from the Group Health Foundation, Seattle, Washington.
We owe gratitude to Kathy Plant, Marni Levy, Kathy Swart, and Madeline Caplow for their technical assistance.
We also express deep appreciation to the participating Group Health physicians, nurse practitioners, physician assistants, and nursing staff for their enthusiastic participation and assistance with this study.
Anne Matthiesen provided her expert assistance with the editing of this manuscript.
- Received June 9, 1999.
- Accepted October 25, 1999.
Reprint requests to (D.C.G.) Harborview Injury Prevention and Research Center, 325 Ninth Ave, Box 359960, Seattle, WA 98104. E-mail:
- ↵National Center for Injury Prevention and Control, Centers for Disease Control and Prevention. Injury mortality statistics. Available at: http://www.cdc.gov/ncipc/osp/usmort.htm
- ↵Cook PJ, Ludwig J. Guns in America: Results of a Comprehensive National Survey on Firearms Ownership and Use. Washington, DC: National Institute of Justice; 1997
- Haught K,
- Grossman DC,
- Connell FA
- Webster DW,
- Wilson ME,
- Duggan AK,
- Pakula LC
- ↵Good P. Permutation Tests: A Practical Guide to Re-sampling Methods for Testing Hypotheses. New York, NY: Springer-Verlag; 1994:29–32
- ↵Manly BFJ. Randomization, Bootstrap and Monte Carlo Methods in Biology. London, UK: Chapman and Hall; 1997:1–17, 80–84
- Oatis PJ,
- Buderer NM,
- Cummings P,
- Flietz R
- ↵DiGuiseppi C, Roberts I. Individual-Level Childhood Injury Prevention Strategies. Los Altos, CA: The Future of Children, Packard Foundation; 2000
- Wall MA,
- Severson HH,
- Andrews J,
- Lichtenstein E,
- Zoref L
- Bass JL,
- Christoffel KK,
- Widome M,
- et al.
- Copyright © 2000 American Academy of Pediatrics