To assess emergency department (ED) clinicians' attitudes and behaviors regarding identification, assessment, and intervention for youth at risk for violence in the ED.
Anonymous, cross-sectional written questionnaire.
The EDs of 3 urban hospitals.
Emergency medicine residents and faculty, pediatric residents, pediatric emergency medicine fellows and faculty, and ED nurses.
A total of 184 (88%) of 208 clinicians completed the questionnaire. Only 15% correctly recognized the lack of existing protocols for addressing youth violence. Clinicians reported being most active in identification of at-risk youth (93% asking context of injury and 82% determining relationships of victim and perpetrator), with pediatricians being more active than general ED clinicians (87% vs 68%; P<.01). Clinicians less often reported performing assessments or referrals of at-risk youth. Nurses and physicians were no different in their reported identification, assessment, or referral behaviors. Barriers identified include concern over upsetting family members, lack of time or skills, and concern for personal safety. Additional clinician training, information about community resources, and specially trained on-site staff were noted by respondents as potential solutions.
Emergency department clinicians recognize the need for evaluation of youth at risk for violence. They are able to identify violently injured youth, but less often perform risk assessment to guide patients to appropriate follow-up resources. Further investigation should address clinician barriers to the complete care of violently injured youth in the ED.
NUMEROUS studies1- 5 have described intentional injury as a chronic, recurrent disease. Injuries resulting from violence represent a substantial number of emergency department (ED) visits, and the proportion is particularly high in adolescent patients.6 Melzer-Lange and Lye7 reported that of all adolescents seen in a pediatric ED, 25% were treated for injuries resulting from violence. The ED staff are thereby positioned to intervene during acute episodes that either reflect an ongoing pattern of violence or may precipitate a cycle of violent retribution. In 1996, the American Academy of Pediatrics issued recommendations for the complete care of the assaulted adolescent, from the initial ED visit to rehabilitative and community services.8 These guidelines recommend that, in addition to providing injury care, ED clinicians thoroughly evaluate the patient's emotional status, potential for recurrence of the assault, retaliation, and acute psychiatric sequelae. In the case of injuries resulting from suicide attempts, child abuse, and domestic violence, established protocols usually result in more comprehensive evaluation and referral. In contrast, few protocols exist for injuries resulting from interpersonal nonintimate violence.7,8 It is not clear, however, whether protocols for youth engaged in interpersonal violence would be widely accepted by ED clinicians. This is an important first step in the implementation of such protocols.
We surveyed the ED clinicians of 3 EDs in our community regarding their attitudes toward this type of prevention effort. This survey examines clinicians' attitudes toward and involvement in each of 3 fundamental components of violence prevention in this setting: (1) identification of at-risk youth, (2) needs assessment for emergent and nonemergent services, and (3) linkage with social support resources in the hospital or community.
A questionnaire was distributed to ED clinicians at 3 hospitals within the same urban community: a tertiary care children's hospital, a general ED at the main campus of an academic medical center, and a community ED that is located further from the main campus. Respondents included emergency medicine residents and faculty, pediatric residents, pediatric emergency medicine fellows and faculty, and ED nurses. Clinicians at the 3 institutions have extensive experience treating victims of violence. In a study of the community surrounding the 3 participating hospitals, fully 50% of surveyed community members of all ages experienced an injury resulting in an ED visit, and up to 25% of the injuries were categorized as intentional or interpersonal.9 More than 70% of injured patients in the community are seen at 1 of the 3 participating hospitals.
The 12-item forced-choice written questionnaire was developed by The Philadelphia (Pa) Health Management Corporation in collaboration with the authors, and distributed by research personnel during staff meetings or through hospital mail. The questionnaire asked the clinicians to report how often they took specific actions in cases of youth violence. In the instructions for the questionnaire, youth violence was defined as assaultive behavior against a person younger than 25 years old, excluding cases of child abuse or domestic abuse. For analysis of self-reported behaviors, responses of "always" or "almost always" were considered positive, whereas "sometimes," "almost never," and "never" were considered negative. The anonymity of the participants was guaranteed.
We grouped the self-reported behaviors into 3 categories representing increasing levels of involvement: identification, assessment, and referral (Figure 1). The successful completion of each of these levels required fulfillment of some basic components. For purposes of the study, successful identification required the clinician to inquire about both the context of the injury and the person causing the injury. Successful assessment of the youth's risk for further injury required the clinician to report asking about the youth's potential for retaliation, as well as inquiring about safety concerns or likelihood of repeat attack. Successful referral required the clinician to provide information on a referral program, contact the referral site personally, or involve a social worker or chaplain.
Division of reported behaviors into 3 levels of clinician involvement.
Differences between clinician subgroups with regard to forced-choice answers were determined by χ2 test or Fisher exact test for dichotomous data. Statistical significance was set at P<.05. All data were analyzed using SAS statistical software (version 6.12, SAS Institute, Cary, NC). Data are presented as percentages with 95% confidence intervals.
This study was approved by the Committee for the Protection for Human Subjects (institutional review board) at all involved institutions.
One hundred eighty-four (88%) of 208 potential respondents completed the questionnaire, including 107 physicians and 77 nurses. Although the 3 hospitals do not have an existing protocol for the treatment of youth violence, 23% of respondents incorrectly thought that there was such a protocol, and 62% did not know. Ninety-nine percent of physicians and nurses responded that health care professionals should determine the events surrounding a violent injury. Moreover, 99% of physicians and 92% of nurses believe that it is specifically their role to ask questions about the events that led to the violent injury. When asked if they felt comfortable asking about events related to assaultive injury, 11 clinicians (5%) reported that they did not. These clinicians reported concern about upsetting the family members, or stated that they lacked the time or skills to address these issues. Of these 11 clinicians, 9 expressed concern for their own personal safety in these situations.
Table 1 presents the percentage of clinicians who report asking specific questions or referring violently injured youth. In general, clinicians reported they are most likely to inquire about the context of the injury and the person causing the injury. More than two thirds of clinicians reported discussing safety concerns with the patient; however, fewer reported that they ask about the potential for retaliation or repeat attack. Although two thirds of clinicians reported that they refer to a social worker or chaplain when necessary, fewer than one third reported giving information about, or directly contacting, a violence prevention program.
Pediatric clinicians were significantly more likely to be rated as successfully identifying at-risk youth than were general ED clinicians (P<.01) (Figure 2). There was less of a difference between these groups with regard to referral (P=.08), and no difference in those reporting a full assessment (P=.86). Nurses and physicians are rated similarly with regard to these 3 levels of involvement (P≥.05).
Levels of screening and referral involvement reported by pediatric and general emergency department (ED) clinicians.
When asked what needed to be done by the hospital to help the staff refer these patients for nonmedical follow-up, 69% of clinicians surveyed requested further training in violence prevention efforts, 92% requested information about community referral services, and 81% requested an on-site staff member dedicated to this goal.
The results of this survey suggest that although ED clinicians indeed consider violence prevention efforts in the ED a worthwhile endeavor, by their own report they do not perform the actual behaviors needed to carry out this type of prevention. The goal of the present study was to investigate the willingness of ED clinicians to address these issues, and to begin to uncover the perceived barriers that limit their involvement.
In our sample in urban teaching institutions, most ED clinicians report that they identify important details of the violent event. Fewer report performing an assessment of the potential immediate and long-term risks for further injury. The majority of clinicians consult a social worker or chaplain to assist in the referral process. Rarely did a clinician report personally contacting or providing information on violence prevention resources.
Ideally, ED clinicians would have the time, training, and resources to appropriately identify and refer young people at risk. However, because these efforts compete for time and energy with other medical, social, and administrative tasks, ED clinicians may find it difficult to provide services beyond acute injury management. Identification of the important details of the violent incident requires only slight augmentation of those performed during the routine patient history. For an appropriate risk assessment to occur, additional interview skills, taking into consideration confidentiality issues, may help to successfully uncover the specific aspects of the medical and social history that reveal a youth to be at risk for further injury10,11 (Table 2). Knowledge of the risk factors for violent injury is a necessary component of this assessment.
Our results are similar to those reported by other investigators. In another recent survey regarding firearm injury prevention, 4 of 5 physicians (internists and surgeons) thought that physicians should be involved in firearm injury prevention, and even more thought that violence prevention should be a priority for physicians.12 However, of those who provide direct patient care, fewer than 20% of these physicians actually counseled patients about firearms in the home. This rift between attitude and action is not limited to those who take care of adults, nor are educational interventions the only answer. A recent educational effort at our institution provided pediatric residents with skill sessions and information to perform primary violence prevention counseling for their clinic patients.13 More than 90% of the residents thought that they would use these skills to counsel patients in subsequent encounters. After the intervention, the resulting 2-fold increase in violence counseling still remained below 20% overall.
The primary limitation to interpreting the results of this survey is the bias associated with self-reported responses. The answers to our questionnaire could have been affected by recall bias, as well as the desire to provide prosocial responses. Although the questionnaire did not request any traceable identifying information, the respondents may have been concerned about anonymity. Importantly, although this survey assessed reported behaviors and attitudes of ED staff, the possibility exists that the questions did not accurately assess the clinicians' attitudes and actions. Other methods such as chart review, patient questionnaires, and direct observation of the clinicians' behaviors can also reflect what actually happens in the ED. Finally, all clinicians surveyed work in urban hospitals, and the results may not necessarily be generalized to rural or suburban settings.
Clearly, comprehensive violence prevention counseling in the ED is a difficult task. The 3 steps of identification, assessment, and referral each require an increasing level of interest, skills, and energy on the part of the clinician. Physicians with both the time and the training to do so can provide brief interventions in the ED that may include assessment, behavioral change strategies, and, when appropriate, referrals for more comprehensive intervention.14,15 Alternatively, as with other "specialized" problems, clinicians in the ED may simply want to identify youth involved in the cycle of violence, and then consult appropriately.
Further investigations are needed to examine the nature and severity of the barriers that limit appropriate risk assessment and referral of these patients in the ED, and to explore, from the clinician's perspective, how to minimize these barriers. Then, educational interventions can be designed and evaluated to determine whether they effectively enhance ED clinicians' abilities to intercede in the acute stages of the cycle of violence. Finally, it remains to be determined what types of institutional support (eg, specially trained on-site staff) would allow trained ED clinicians to apply a consistent, comprehensive approach to youth at risk for violent injury.
Editor's Note: This is another study showing that clinicians recognize the need to do something that they don't actually do because of lack of time and/or resources. In the meantime, managed cost continues to turn the screws.—Catherine D. DeAngelis, MD
Accepted for publication October 6, 1999.
This work was supported by The William Penn Foundation, The Department of Pediatrics of the University of Pennsylvania School of Medicine, Philadelphia, and The Ethel Brown Foerderer Fund for Excellence, The Children's Hospital of Philadelphia.
Presented at The Pediatric Academic Societies Annual Meeting, San Francisco, Calif, May 1, 1999.
We thank the Emergency Department medical, nursing, social work, registration, and trauma chaplaincy staff of The Children's Hospital of Philadelphia, The Hospital of the University of Pennsylvania, and The Presbyterian Hospital for their cooperation with this project; Tara Nolan, Shapir Rosenberg, Helen Kang, and Gunjan Modi for data management; and Auria Rosa for manuscript preparation.
Reprints: Joel A. Fein, MD, Division of Emergency Medicine, The Children's Hospital of Philadelphia, 34th St and Civic Center Boulevard, Philadelphia, PA, 19104 (e-mail: email@example.com).
Fein JA, Ginsburg KR, McGrath ME, Shofer FS, Flamma JC, Datner EM. Violence Prevention in the Emergency DepartmentClinician Attitudes and Limitations. Arch Pediatr Adolesc Med. 2000;154(5):495-498. doi:10.1001/archpedi.154.5.495