Several medical organizations, including the American Medical Association, the American Public Health Association, and the National Medical Association, have called for increased attention from the medical community to this issue. Moreover, the American Academy of Pediatrics, the American Pediatric Surgical Association, the American Trauma Society, and the Eastern Association for the Surgery of Trauma have adopted or proposed policies on control of firearm violence [4-9]. Medical journals have published editorials [10-18], hosted formal debates [19], issued calls for papers, organized special theme issues [10, 12, 20, 21], and published policy-oriented papers [22-29] on gun violence, all of which prompted brisk responses in letters to the editor [30-44]. The literature on gun violence has been growing with several notable studies done on the relation between gun violence and death [21, 45-58].

Despite the increased reframing of the issue, few studies have examined the attitudes and practices of physicians in relation to firearms. These include studies of pediatricians [59-61], subscribers to Physician's Management [62], physicians in Portland, Oregon [63], and directors of the nation's trauma centers [64]. Firearm injury appears in the clinical prevention literature. Physicians are urged to inquire about guns in the house and counsel patients about firearm safety measures [65, 66].

To describe and define more clearly the attitudes of physicians toward guns and prevention of firearm-related injury, we designed a survey and administered it to the membership of the American College of Physicians (ACP) and the American College of Surgeons [ACS]. These organizations are the two largest medical specialty societies in the United States, representing approximately 100 000 internists and 68 000 surgeons, respectively. We report descriptive results and add to the literature on physician attitudes by answering the following questions: 1) To what extent do internists and surgeons view firearm violence, injury, and deaths from a public health perspective? 2) What roles, if any, do internists and surgeons see for physicians with regard to firearm violence? 3) What public policies do internists and surgeons support for the regulation of firearms?

Methods

Top Methods Results Discussion Author & Article Info References

Study Sample

We drew a random sample of ACP and ACS members. Each association maintains its members in a random listing. In preparation for sampling, ACP and ACS removed ineligible members from the total list of members. Selected segments of the memberships were then excluded to yield a sample of actively employed, nonmilitary physicians living in the United States. The remaining members were sorted according to the four regions of the United States (northeast, south, midwest, and west) to assure regional representation.

A two-stage sampling technique was used. In the first stage, a random start was chosen and every nth member was selected (n was determined by dividing the number of members in each region by 300). The total number of physicians eligible for inclusion in the sample across both associations was 94 271. To achieve estimates with 95% probability of accuracy with an error rate of ± 5%, a total of 800 completed interviews (400 from each association) was required. On the basis of an anticipated 40% response rate, we drew a sample of 2409 physicians (1209 from ACP and 1200 from ACS) from all four regions. In the second stage, 2019 physician names were released to the interviewers; the remainder were held in reserve in case the requisite number of interviews could not be completed with the initial larger set. From this set, appointments were then made for interviewing.

We used a practice common in telephone interviewing: Once the sample was selected, we began interviewing a subset of participants randomly selected from the whole sample until the number of required interviews was reached. The total number of participants released to the interviewers from the whole sample then became the denominator in determining response rate; this denominator was less than the number of participants in the whole sample. A total of 1108 of the 2019 internists and surgeons were contacted and asked to participate before the target number of completed interviews was reached. Of the physicians contacted, 915 completed interviews (457 from ACP and 458 from ACS). The 115 interviews beyond the target of 800 resulted from interview appointments that had been made before the required number of completed interviews had been reached. The compliance rate was 82.5% (915 completed interviews ÷ 1108 physicians who were contacted [that is, 193 refusals + 915 interviews]) and the overall response rate was 45.3% (915 completed interviews ÷ 2019 physicians in the sample released to the interviewers).

Questionnaire and Data Collection

By using the literature on firearm violence and guidance from recognized experts in the field, we identified six domains of importance: physician-clinician experience with firearms, knowledge about clinical sequelae of firearm injury, knowledge about public policies on firearm violence, attitudes toward public policies on firearm violence, clinical practice behavior, and education and training. The questionnaire was revised after being pretested by telephone interviews with 31 ACP members and 20 ACS members. The final version was a 55-item instrument that took an average of 14 minutes to complete. Response categories included yes/no, favor/oppose, 3- and 5-point Likert scales, and categorical selections.

Data Analysis

The National Opinion Research Center, an independent research firm, converted the questionnaire into a computer-assisted telephone interview format and conducted the interviews. Data were collected between 19 January 1996 and 12 February 1996.

Statistical Analysis

We used descriptive and bivariate statistical techniques. Unstandardized and standardized weights were produced. The unstandardized weight represented the number of internists and surgeons in the target population who were represented by each participant in that stratum. Because the weights varied greatly, we used weighted values in the analysis. The average unstandardized weight across the strata was 103; this meant that on average, each study participant represented about 103 internists or surgeons in the U.S. target population. The standardized weights were used to estimate the SEs of the means and proportions and to perform Pearson chi-square testing for significance. Because our report involves comparisons of multiple responses from the same physicians, we used adjusted critical P values of 0.001, 0.002, and 0.003 to judge the statistical significance of individual comparisons within a table (Bonferroni correction) so that the family-wise type I error rate was limited to 0.05 for each set of comparisons.

Results

Top Methods Results Discussion Author & Article Info References

Demographic Characteristics

As a group, the respondents were representative of the ACP and ACS memberships, adjusted for age, sex, ethnicity, and regional distribution (Table 1).

Firearm Violence as a Public Health Issue

Most respondents (94% of internists and 87% of surgeons) agreed that firearm violence has become a major public health issue. Support for a public health perspective was greatest among physicians who were not raised in households that had guns, did not currently own guns, belonged to gun control organizations, and were not members of gun clubs. Physicians who endorsed the public health position tended to be female, younger, and currently practicing in large cities (Table 2). Although support for the public health perspective was not related to negative experiences with guns, attitudes toward some medical practice variables were (Table 3). Physicians who took a public health perspective on guns tended to think that it is appropriate for physicians to offer counseling on firearm safety, to believe the medical literature on guns, to think that physicians should be involved in firearm injury prevention, and to think that violence prevention should be a priority for physicians.

Firearm Violence and the Role of Internists and Surgeons

Respondents supported two roles for physicians. One concerned the involvement of internists and surgeons in community efforts that address firearm violence. Fewer surgeons than internists (64% and 84%) thought that physicians should support community efforts to enact legislation restricting the possession or sale of handguns.

The other role concerned the clinical implications of firearm violence and their bearing on individual physician practices. Most internists and surgeons (84% and 72%) thought that physicians should be involved in firearm injury prevention. Furthermore, 88% of internists and 78% of surgeons believed the medical literature that describes the increased risk for gun injury associated with having guns in the home (Table 3). Among respondents who provided direct patient care, few currently include firearm safety counseling in their clinical practice (Table 4). Less than 20% usually discuss firearm ownership or storage as part of safety counseling, 4% of internists and 2% of surgeons frequently talk to patients about having a gun in the house, and approximately 50% never discuss these topics. Few internists and surgeons received training and education about assault or homicide by adolescents (14% and 12%), domestic violence (47% and 34%), or conflict resolution (24% and 20%). More internists than surgeons (68% and 38%) received training in suicide prevention.

Sixty-three percent of internists and 52% of surgeons were interested in receiving education about prevention of firearm-related injuries. Physicians who were female, were younger, or knew a gunshot victim were more interested in receiving additional training.

Public Policies on Firearms

Most respondents in both groups thought that a wide range of appropriate policy reforms can help reduce firearm violence and that gun control legislation will reduce the risks for injury and death from firearms. Restricting access to high-capacity, rapid-fire weapons and dangerous new forms of firearms and handguns received support, and 84% of gun-owning internists favored registration of handguns (Table 5).

Several factors seem to be associated with support for firearm regulation (Table 6). Support was higher among physicians who were raised in households without guns, are members of gun control organizations (but not members of gun clubs), were raised and practice medicine in urban settings and urbanized regions (such as the northeastern United States), are female, are younger, and are internists. Most respondents (91%) knew someone either personally or professionally who had been injured by a firearm, and 27% of respondents in both groups reported that they or a member of their family had been threatened by someone with a gun. However, negative experiences with guns did not necessarily lead to greater support for firearm regulation. Seventy-four percent of respondents who reported knowing someone who had been shot had favorable attitudes toward gun control; however, 82% of respondents who did not know someone who had been shot also had favorable attitudes toward gun control. In addition, support for regulation of firearms was greater among respondents who believed the medical literature on firearms (82%) than among those who did not (29%) and among respondents who thought that physicians should offer safety counseling (76%) than among those who did not think that this is appropriate (50%).

Discussion

Top Methods Results Discussion Author & Article Info References

Consistent with the results of earlier studies [25, 52, 54, 58-61], most internists and surgeons who responded to our survey viewed gun violence as a public health issue. They thought that the problem of gun violence was worsening and considered violence prevention a priority issue. Most of these physicians also thought that firearm regulation can reduce injuries and deaths and that physicians should support community efforts to restrict the ownership and sale of handguns. Moreover, most respondents in both groups supported a wide range of measures to register and restrict firearms. Only the most stringent regulations (such as denying handguns even to people who have undergone a background check and a waiting period) were not supported by most respondents. Internists and surgeons thought that it is appropriate for physicians to provide counseling on firearm safety, but few currently cover firearms ownership or storage in their injury prevention counseling. Because our sampling technique was designed to reliably capture membership opinion, we are confident that as a group, the respondents were representative of ACP and ACS membership, adjusted for age, sex, ethnicity, and regional distribution.

Discussion of firearm risks and safety measures has not traditionally been part of the practice of preventive medicine, but the growing epidemic of death, injury, and disability related to firearms suggests an important role for physicians in patient screening, counseling, and education. It is important for physicians to become educated about the risks to patients and their families of gun ownership and the need for safety measures. Broader understanding of risk factors for suicide and identification and treatment of depression are also important. Most physicians who responded to our survey said that they would like to receive more education in these matters and more training in preventive counseling skills. This is an important message for medical schools and organizations that provide continuing medical education.

At the same time, attitudes are not uniform across all types of physicians with respect to gun violence as a public health issue and the regulation of firearms. Perhaps not surprisingly, support for gun control measures and perception of gun violence as a public health issue seemed to be shaped by exposure to guns and by certain medical outlooks. First, physicians who were raised in households with guns, who currently own guns, and who are members of gun clubs are less supportive of policies to regulate firearms and of a public health perspective on gun violence. In addition, believing the reports in the medical literature about the epidemiology of firearm injury and death and considering safety counseling an appropriate role for health care providers are associated with support for firearm regulations and for a public health position on gun violence. Specific counseling practices and having at-risk patients do not seem to be related to these positions. Internists are somewhat more likely than surgeons to support firearm regulation and to take a public health perspective on gun violence.

Our study had several limitations. First, although the compliance rate (82.4%) was high, almost half of the refusals (41.4%) were made by gatekeepers and not by the physicians directly. The gatekeepers' attitudes may differ from those of the physicians, and the views of physicians who were kept from participating in an interview may differ from those of the physicians who were interviewed. However, for there to be a bias of any consequence with the low nonresponse rate, the attitudes of physicians who were not contacted or who refused to participate would have had to differ substantially from the attitudes of those who participated. If we assume that all 193 physicians who refused to be interviewed think that firearm injury is not a public health issue, the combined rate of internists and surgeons who support the public health perspective would decrease from 91% (832 of 915 physicians) to 75% (832 of 1108 physicians). Because such unanimous opposition seems unlikely, the potential for sampling bias in our results is small. Second, our survey was designed to touch on many variables but not to investigate them in depth. The questionnaire may therefore have not been specific enough in such areas as details of medical training and current counseling practices. Further investigation is needed of the relations among the variables that we identified. Finally, in designing the instrument, we may have overlooked significant variables.

Despite these limitations, our study provides evidence that most members of ACP and ACS agree that firearm injuries are a public health issue of growing importance; that physicians should join community efforts to regulate guns; that safety counseling, including discussion of firearm risks, is within the scope of physician practice; and that a wide range of specific gun regulation measures should be adopted as public policy.

From Mount Sinai Medical Center and Harlem Hospital and Columbia University College of Physicians and Surgeons, New York, New York; American College of Physicians, University of Pennsylvania Medical Center, and Educational Commission foreign Medical Graduates, Philadelphia, Pennsylvania; and National Opinion Research Center, Chicago, Illinois.

Ms. Nelson: Research Center, American College of Physicians, Sixth Street at Race, Philadelphia, PA 19106.

Dr. Smith: General Social Survey, National Opinion Research Center, 1155 East 60th Street, Chicago, IL 60637.

Dr. Schwab: Division of Traumatology and Surgical Critical Care, University of Pennsylvania Medical Center, 3400 Spruce Street, Philadelphia, PA 19104.

Dr. Barlow: Pediatric Surgery, Harlem Hospital, MLK-17103, 506 Lennox Avenue, New York, NY 10037.

Dr. Gary: Educational Commission foreign Medical Graduates, 3624 Market Street, Philadelphia, PA 19104.