Renee Butkus, BA; Arlene Weissman, PhD
Disclosures: Authors have disclosed no conflicts of interest. Forms can be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M13-1960.
Reproducible Research Statement: Study protocol, statistical code, and data set: Available from Dr. Weissman (e-mail, firstname.lastname@example.org).
Requests for Single Reprints: Renee Butkus, BA, American College of Physicians, 25 Massachusetts Avenue NW, Washington, DC 20001; e-mail, email@example.com.
Current Author Addresses: Ms. Butkus: American College of Physicians, 25 Massachusetts Avenue NW, Washington, DC 20001.
Dr. Weissman: American College of Physicians, 190 N. Independence Mall West, Philadelphia, PA 19106.
Author Contributions: Conception and design: R. Butkus, A. Weissman.
Analysis and interpretation of the data: R. Butkus, A. Weissman.
Drafting of the article: R. Butkus, A. Weissman.
Critical revision of the article for important intellectual content: R. Butkus.
Final approval of the article: R. Butkus, A. Weissman.
Administrative, technical, or logistic support: R. Butkus.
Collection and assembly of data: A. Weissman.
Professional organizations have called for the medical community's attention to the prevention of firearm injury. However, little is known about physicians’ attitudes and practices in preventing firearm injury.
To determine internists’ attitudes and practices about firearms and to assess whether opinions differ according to whether there are gun owners in a physician's home.
Internal medicine practices.
573 internists representative of American College of Physicians’ members.
Respondents’ experiences and reported practice behaviors related to firearms and their opinions about contributors and public policies related to firearm violence, as well as physician education and training in firearm safety.
The survey response rate was 56.5%. Eighty-five percent of respondents believed that firearm injury is a public health issue, and 71% believed that it is a bigger problem today than a decade ago. Seventy-six percent of respondents believed that stricter gun control legislation would help reduce the risks for gun-related injuries or deaths. Although 66% of respondents believed that physicians should have the right to counsel patients on preventing deaths and injuries from firearms, 58% reported never asking whether patients have guns in their homes.
The generalizability of these findings to non–American College of Physicians’ member internists and other physicians is unknown. Responses may not reflect actual behavior.
Most respondents believed that firearm-related violence is a public health issue and favored policy initiatives aimed at reducing it. Although most internists supported a physician's right to counsel patients about gun safety, few reported currently doing it.
Recent events have prompted calls for medical community involvement in efforts to reduce firearm-related violence. Physicians’ attitudes about potential efforts at reducing firearm violence are not well-understood.
This survey found that most internists believed that firearm-related violence is a public health issue and that physicians should have the right to discuss firearm safety with patients, although few reported doing so. Most respondents reported favoring various public policies aimed at reducing firearm-related violence.
The study included only internists, and it could not evaluate actual practice.
Physician support may be useful in efforts to reduce firearm-related violence.
Table 1. Characteristics of Survey Respondents
Table 2. Importance of a U.S. Plan to Prevent Firearm Violence, by Demographic Characteristics
Attitudes of respondents toward firearm injury and prevention.
The y-axis shows the percentage of respondents with opinions about each category.
Appendix Table 1. Support by Internists for Specific Measures to Deal With Firearm Violence
Percentage favoring various firearm-related initiatives.
Appendix Table 2. Frequency of Discussions About Gun-Related Issues
Appendix Table 3. Interest in Education
Thomas F Heston MD
Assistant Professor (Adjunct), Johns Hopkins University
April 13, 2014
A Statistical Solution to Low Response Rates
This study had a low response rate that potentially resulted in a self-selection sampling bias. Although the authors suggested that the non-response bias was overcome by showing demographic data that respondents were similar to the non-respondents, we currently have no consistent way to deal with non-random missing data (1). I believe that the method I propose here is a valid, conservative, and new solution that can be utilized to overcome non-response bias.In this study, 324 out of 573 internists surveyed responded. Out of these, 85% (275) believed that gun violence was a pubic health issue. A conservative way to handle non-random missing data in such surveys is to assume that the bias in non-respondents is equally strong, but in the opposite direction. Using this method, we apply the positive response rate in the respondents to the non-respondents, but in the opposite direction. In this study there were 249 non-respondents. Assume that 85% of non-respondents believe that gun violence is NOT a public health issue and that 15% (37/249) believe the opposite. Combining the results of the respondents with the non-respondents, we estimate that 312 out of 573 (55%) believe gun violence is a public health issue. The 95% confidence interval for this proportion is 44% to 65%, and the alternative hypothesis (that most internists believe gun violence is a public health issue) can be rejected.By extension, this method also provides a way to determine an adequate response rate. In this case, a response rate of at least 66% is necessary as demonstrated below.A 66% response rate would mean that 378 out of 573 surveyed responded. Of the respondents, if the 85% rate held true, 321 would state gun violence is a public health issue. Of the 195 non-respondents, we estimate that 29 (15%) agree. Combining the results, we estimate that 350 out of 573 (61%) believe gun violence is a public health issue. The 95% confidence interval for 350/573 is 51% to 72%, and the alternative hypothesis is confirmed.Since the response rate from this survey was only 56.5%, it does not meet the required threshold of 66% demonstrated above. Therefore, the conclusion that internists as a group believe that gun violence is a public health issue can be hypothesized, but not confirmed.REFERENCES1. Altman DG, Bland JM. Missing Data. BMJ. 2007 Feb 24;334(7590):424.
James Webster, MD, MS, MACP
June 30, 2014
This ACP policy paper (1) is a comprehensive, authoritative, compendium of facts and
recommendations which describe why and how the U. S. can achieve significant reductions in firearm
related injuries and deaths. It correctly identifies gun violence (GV) as a major public health issue. With
more than 30,000 deaths and over 70,000 injuries each year GV most certainly qualifies as a major
epidemic. It is crucial that the recommendations of the paper now be prioritized into action initiatives.
Specifics could include:
• Convening a permanent ACP sponsored task force on the topic. This multidisciplinary coalition[recommendation 1.a (1)] could provide leadership for the entire profession as it generates
Specific goals, objectives and tactics on a continuing basis.
• Mobilizing the considerable resources and prestige of the ACP to craft and present draft
legislation to Federal lawmakers to reduce GV. Such templates could also be used by
Chapters on a local State basis. These bills could address the regulation of gun sales [recommendations 3, a,b,c,d,e, (1)] and improving the safety of weapons for owners [recommendations 7.8 (1)].
• Developing educational materials for physicians to distribute and discuss with patients and
families on the risks, responsibilities and safety measures for gun owners [recommendations 2 b and 5 (1)].
• Producing and disseminating curricular materials for physician education on various aspects of
GV. This would empower internists to better be able to educate, advise, and counsel patients
(2) about issues of prevention surrounding GV. This could be used at Chapter and
National meetings, in CME venues and by training programs [reco0mendation 2, b,c (1)].
• Supporting broad research on the topic of GV, even perhaps using seed money grants.
[recommendation 9 (1)].
There are clearly other projects that ACP could consider regarding GV, an area that is calling out
for leadership and collective action. These activities are totally consistent with the ACP’s
mission and how the ACP has responded to other epidemics and medical crises such as obesity,
Diabetes and the public health disaster in Haiti. This initiative would be a very cost efficient use
of ACP resources. Unless the position paper is followed up with action steps it may only gather
dust and tragically could all too soon be forgotten.
1. Butkus R. Doherty R, Daniel H. Reducing firearm-related injuries and deaths in the United States:
Executive summary of a policy position paper from the American College of Physicians. Ann Intern Med.
2.Butkus R, Weissman A. Internists’ attitudes toward prevention of firearm injury. Ann Intern Med.
Butkus R, Weissman A. Internists’ Attitudes Toward Prevention of Firearm Injury. Ann Intern Med. 2014;160:821–827. doi: https://doi.org/10.7326/M13-1960
Download citation file:
Published: Ann Intern Med. 2014;160(12):821-827.
Healthcare Delivery and Policy.
Results provided by:
Copyright © 2020 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use